Yeah I am 90% (more?) sure I got COVID at a big state high school sporting tournament where ~100,000 people attend over a few days the first week of MAR where at the time supposedly there was just 1 or 3 or 5 or something people with COVID in the entire state (and they were all Chinese nationals or recent visitors).
Not sure there was a "cover-up". But it definitely went here from "only worry if you have direct contact with people who were recently in China", to "OMG it is everywhere", in about 3 days +/-. Which makes me think they were not really giving accurate descriptions to the press at the time.
They did not gracefully transition from "be normal" to "OMG panic" as the evidence was likely suggesting, and instead clung to normalcy as long as possible until it was untenable.
For what it’s worth, I remember lots of people saying exactly this about getting COVID prior to ~March 2020 and then when the antibody tests became available I saw a lot of posts says “I guess I actually never had COVID”. I can’t put the numbers to that that you would need to make any sense of it, but I’m pretty skeptical of any specific claims. It’s pretty hard to distinguish COVID from other infections since the range of symptoms across individuals is pretty big (I never had smell loss when I had it and that’s the most identifying symptom). That said, there were probably early undetected outbreaks somewhere and your anecdote is only slightly earlier than official numbers and we all know testing wasn’t being done.
I know some people who strongly believe they got COVID November of 2019, which doesn't seem possible unless there was a massive coordinated coverup between both Chinese and Western medicine to hide that it spread earlier than thought. Certainly by now we would have some people admit that it spread earlier than expected. The most recent that I heard, there was some potential updates to how early it was breaking out in China, to November, but that there was no way it was spreading widely anywhere else - especially random continental US.
To figure out whether a local outbreak of COVID-19 happened before health authorities said a Pandemic was in the air, maybe you should look at these factors:
A. Is this an area with any amount of international travel incoming, or outgoing?
B. When local officials began officially noticing COVID cases, how large was the first surge of known-cases?
I was trying to figure out whether a large surge of cases or a small surge of cases, post-Declaration-of-Pandemic, would indicate wide circulation pre-Declaration-of-Pandemic. It actually depends on the timing: if the Declaration-of-Pandemic happened midway through the first surge of cases, then the case-count might go from single-digit to many-thousands within a week or so. On the other hand, if the first surge of cases came and went pre-Declaration-of-Pandemic, then the case-count after the Declaration-of-Pandemic would look like a splutter.
This might also be confounded by the age-curve of the first wave. Though it was not fully appreciated at the time of Declaration-of-Pandemic, sufferers of COVID in the age bracket of elderly-retirees had much higher death rates than sufferers of COVID who were not in that age bracket. Thus, an area with a large number of international students might see lots of cases of COVID, and very few deaths-due-to-COVID.
Do you think that COVID was introduced in your area sometime before December 2019, or during January 2020?
Was there an increase in rates-of-death in nursing homes before the official Declaration-of-Pandemic?
"The initial batch of cases were tiny and we had low case counts throughout the entire pandemic. It was often commented on how the East Coast of Canada was a pocket of safety, when in actuality they failed completely to protect us and didn't even notice. And then when they finally did notice they didn't tell us."
I am also in Halifax, Nova Scotia, and it was my personal experience, and that of my social circle, that Eastern Canada was a significantly safer place for Covid than the rest of Canada, and even more so when compared to the US.
This was obvious when looking at the official statistics, and also anecdotally...I didn't know anyone who got it...no one...until the Omicron wave. (We've probably started to revert to the mean in the last year or so, but I have not been tracking the data as closely recently)
So I would be really interested to hear what evidence you have for these accusations of a cover up.
- is there excess mortality data? The US has this: https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm. It shows excess mortality jumping in late March 2020 as you'd expect, and later waves roughly when people thought there were waves. I am very willing to believe there were isolated outbreaks of COVID in the West before March 2020, but also would want more evidence than anecdotal.
- what exactly is the "anti-mandate" movement in the context of the present day in Western countries? Does it mean against vaccine mandates specifically, or COVID rules generally? For me and people I know COVID is basically over, except for a few people who continue to be worried, and structural changes that I think persist for non-COVID reasons (e.g. work from home, outdoor dining). But of course I'm not everywhere. It's not clear to me there are a lot of mandates in the US at least right now, though I got my vaccine the first day I could so I never had to worry about it and never paid that close attention. Not sure the situation in Canada.
Anecdotally: My wife came back from a trip to San Francisco in late January with what seemed to be a case of the flu. She was pretty miserable for about a week. We both suspect she had covid then. It was officially not in the US at that point, largely because nobody was allowed to do their own covid test--they had to use the broken CDC-provided one for bureaucratic turf reasons.
And there's always a fuckton of people who actually have the (ordinary, miserable) flu in any January. Absent testing, there's no real way to know. And thanks to the CDC, we didn't have the tests when we needed them.
Picking only from the people I think Elon would plausible pick (long time superfans, essentially) Jason Calacanis is probably not too terrible. I sometimes watch his 'This Week in Startups' show and he's won me over a bit in the longer format.
Maybe somebody *could* have fixed Twitter before the acquisition, but is it even possible now with the new debt?
>In its last quarterly filing as a public company, for the period ended June 30, revenue was $1.18 billion, down slightly year-over-year.
>Twitter paid $23.3 million in interest expense in the quarter ended June 30, according to a filing.
>Now, the company will have to pay at least $9 billion in interest to banks and hedge funds over the next seven to eight years, when the $13 billion in debt matures, according to a review of Twitter’s loans by Mr. Davies, the former credit analyst.
It appears they may be down about 5,000 employees from when Musk took over. Assuming salary and benefits per employee of about $150,000 per year (which I think is conservative), that's $750,000,000 a year - no small change! We don't know if Twitter's income has changed or if there are other expenses that changed (higher or lower), so it's not readily possible for us to estimate their ability to pay off the loans, but Musk has made some really big steps towards it.
There was some talk of socialising it , as a public space. This would be a very unusual thing for the US to do and then adds another problem - can anybody be ever banned again?
> There was some talk of socialising it , as a public space. This would be a very unusual thing for the US to do and then adds another problem - can anybody be ever banned again?
Sure, pedophiles. If you can ban them from speaking in private spaces, you can also ban them from speaking in public ones.
The logical next step is to streamline the process of getting someone a child pornography conviction when you want to curtail their speech.
Who was talking about public ownership of Twitter?
Anyway, in theory, if the US government bought Twitter, couldn't Congress then donate it to a foundation which is set up to manage it in perpetuity? Couldn't that foundation then ban users, and so on, without any First Amendment concerns?
The $13 billion shouldn't be an issue if they make their payments for those 7-8 years, someone will (barring drastic interest rate changes) float them another $13 billion if they have managed those payments for 7+ years. Will they be able to make those payments? That probably depends on the structure of the debt and when those payments have to be made.
Also using quarterly data instead of annual data for revenue and then using a large lump sum for interest is disingenuous and makes it easy to misunderstand the situation.
Any predictions on what Elon will do after the poll closes?
Result: No, Elon should remain in charge of Twitter.
Elon changes his profile to "The People's CEO". He is spotted wearing a tshirt with that on it in the next week. Two months from now Elon says he needs to focus on Tesla and SpaceX. Elon appoints one of the hosts of the 'All-In Podcast' as CEO, most likely David Sacks or Jason Calacanis.
Result: Yes, Elon should step down.
Elon announces he will step away in two months and appoints one of the hosts of the 'All-In Podcast' as CEO, most likely David Sacks or Jason Calacanis.
Long Shot: Elon sells a bigger chunk of Twitter to the Saudis and actually really steps away and let's someone else deal with it. They choose some random CEO the new investors approve of. https://twitter.com/LauraLoomer/status/1604540593354735616
The poll closed this morning with 17M votes cast and it went 57-43 in favor of him stepping down. No response yet from Musk.
Tesla shares, which had dropped 11% recently as Musk sold shares, rose 5% this morning after the above Twitter poll closed.
Last night Snoop Dogg posted a poll "Should I run Twitter?" As of this morning it had 1.3 votes, split 80-20 "yes".
Calacanis posted a poll asking who should become Twitter’s next CEO: himself, Sacks, or Calacanis and Sacks together as co-CEOs. That poll's "other" option won with 39% of the 14,000 votes cast.
Predictions update? Starting to think I completely misread the situation.
I though Elon's poll was part of a plan that works no matter the result. But the poll closed 8+ hours ago and Elon hasn't said a thing. Maybe he's just hungover or whatever, but it's not feeling as much like a plan as I thought. Really no followup ready to go upon the poll ending?
Elon finally reacts. He's going with the deep state botted the vote. Looks like the plan wasn't to be voted out as an exit strategy? Seem like they might even redo the entire vote and only allow Twitter Blue users.
I'm curious, are there any other people who started out very pro-Musk and ended up anti? Where was your tipping point? Mine was, I think, when he banned the account that posted his jet's location, though my confidence had been eroding for a while before.
I don't know about very pro, but definitely a positive valence, which is now negative (more annoyed than disgusted or morally appalled). It's very odd (and a bit disheartening) to see his somewhat reasonable side publicly clash with his anti-left edgelord side (well, maybe not edgelord, but facebook-tier boomer meme, at least). Seems like buying Twitter was a bad impulse that he now can't escape from.
I think he has an unhealthy interaction with Twitter, and that involving himself deeply in it was a mistake for that reason. But I also don't require Musk to be a god or a demon, just a pretty smart and industrious dude who has accomplished a lot and will hopefully keep on accomplishing a lot in the future.
He's addicted to Twitter just any common social media flunkey addict. If his skill is "entrepreneurial engineering," then all the the time he spends on his silly tweets has got to distract him from his Real Productivity.
And it shows a weakness in our society that we're vulnerable to being so dominated by the autistic and eccentric whims of billionaires.
Why would someone be "pro" or "anti" Musk? He is just a guy, and as a standard bearer for an ideology or viewpoint he doesn't work very well because he is all over the place.
He is pretty clearly a genius businessman/investor, but perhaps a bit off his rocker these days. Though people were also saying that same thing 10+ years ago, and he is now one of the wealthiest people in human history so...
Blue tribe has rallied around hating him because they are tribal morons, but I assume that doesn't include you.
Anyway, like everyone else I think his recent shenanigans and Twitter acquisition are likely big mistakes and will fail, but I trust his judgement more than mine in such matters. Plus Twitter is blight on society so if some super rich guy wastes a bunch of money and Twitter gets destroyed that sounds like a win-win?
Phrasing is hard, especially having just gotten off a transcontinental flight. What I meant by "pro- and anti-Musk" is simply "in favor of, or against, him running Twitter". I'm a free-speech nutcase by today's standards (definitely not blue tribe), and I thought he was too, but now I'm thinking he's in favor of "free speech for people who agree with me" and am sad.
What I'll say about Musk is that he's intemperate. Which can be harnessed to good use when you're trying to lead and direct a team towards huge, almost preposterous achievements, like technological breakthroughs or military conquests. But it's a terribly counterproductive personality trait when you're trying to promote and referee a place for broad, thoughtful public discussions.
I think he sincerely wants Twitter to be the broadest possible place for thoughtful online discussion, but he just can't get out of his own way. By contrast, I don't think the previous regime truly wanted Twitter to be that place, but it certainly was a lot more temperate in how it went about it.
Even that might be better than having all big social media and traditional media companies reading off exactly the same page w.r.t. moderation, including stuff like kicking Trump off, blocking right-wing social media networks, blocking covid disinformation, etc. Having some stuff become unpublishable/invisible in US media and social media is dangerous on all kinds of levels.
The lockstep of traditional and social media post-2020 seems very similar to the lockstep of traditional media post-9/11 attacks, and seems equally unhealthy. Making sure nobody hears any debate or dissenting views about major policies (invading Afghanistan, invading Iraq) is a great way to help the country walk off a cliff, and indeed, that's pretty much what happened.
OK, that changes the meaning dramatically. I've always been pro-most-of-what-Elon-Musk-does, and admired his ability to get things done. That hasn't changed very much. I've always been strongly anti-Musk-running-Twitter, because that really isn't well suited to his skillset or his temperament and it is going to severely handicap his ability to do other things. Things that he's actually good at and that I want to see done well.
I'd also like to see Twitter either destroyed or radically transformed, but anyone but Elon Musk would be a better choice for that. Who's the #2 investor in that cabal? Mohammed bin Salman? Yeah, OK, let him run it - the mutual destruction of MBS and Twitter would be an absolute net positive.
I think banning the jet was an instance where a button got pushed with him. His kid was in danger, the mother of the kid was probably there, my diagnosis of that is that the mother of the child has an “Elon obey me” button that still works even though they’re divorced - this is not that uncommon. He probably got an awful phone call in which it was demanded that he Do Something, so he did. I think it did not have much to do with his thoughts or ideas on free speech, probably bypassed all of it. Is it a little upsetting that there’s an Obey button? Kind of, but I don’t think it comes up that often. Probably only a few people have one on him. The question becomes who has them & when are they pushed. Leading Twitter does mean he’ll be surrounded by people all trying to get to the buttons. I think people are trying to use him here and the sooner he leaves the better for him and Tesla. I see him as someone who gets things done where others don’t - but projects and goals. Not evolving balances of political rhetoric. Knowing that Twitter has been used as a tool of governments is helpful. Making that stop will be very hard and boring for him, plus he’ll have to admit the power brokers don’t like or respect him, there’s no mutual engineer esteem, it’s manipulation.
I think he should appoint a thirteen-person council and bail ASAP. I still respect him.
Yeah, from what I understand, this jet-tracking account was going for years, but then some rando went after Musk's kid so Dad booted it all off Twitter. Understandable reaction, whatever your opinion on doxxing or jet tracking or whatever.
I thought the Thursday Night Massacre was much more fun (exactly how full of yourself do you have to be, to invent a grandiose title for "we got suspended for two days"?)
I mean, to be fair, the name was probably coined before it was known whether the bans would be indefinite, like the case of the people who made Musk parody accounts (and only Musk parody accounts, I don't recall hearing about a single other impersonator account being banned) after Musk said comedy was legal on twitter.
It is still rather self-important, though; how many people have been banned permanently from Twitter or other social media sites? The guy I saw lamenting about it opened up with "I and eight other journalists" - stop the world, nine journalists whose names possibly nobody would recognise have (one of) their social media account suspended! Truly Democracy Dies In Darkness!
Meanwhile, if Mr Thursday Night was fired by his employer and lost access to the Twitter account that goes with the job - would we know or care? Would journalism grind to a halt? Does anyone, apart from journalists, really follow what someone says on Twitter as their main/only source of information?
I suppose some people do, but if Twitter imploded in the morning, some other site would take over from it. Everyone would go to Mastodon or somewhere. Or a new gossip website would spring up, as TikTok took over from Vine.
> this jet-tracking account was going for years, but then some rando went after Musk's kid so Dad booted it all off Twitter. Understandable reaction, whatever your opinion on doxxing or jet tracking or whatever.
Reminds me of the how a certain ex-president's account was going for years until one day a bunch of randos went after congress and the VP at his urging.
I'm a father myself, twice over, and agree about that being an understandable reaction....but it doesn't seem clear that the incident, whatever it was, actually had anything to do with the @ElonJet Twitter account. From the Washington Post:
"A confrontation between a member of Elon Musk’s security team and an alleged stalker that Musk blamed on a Twitter account that tracked his jet took place at a gas station 26 miles from Los Angeles International Airport and 23 hours after the @ElonJet account had last located the jet’s whereabouts....Police have said little about the incident but say they’ve yet to find a link between the confrontation and the jet-tracking account....
The incident occurred in South Pasadena, a Los Angeles suburb, on Tuesday at about 9:45 p.m. South Pasadena police were called to the gas station, according to the business’s manager, but made no arrests. South Pasadena police have not responded to requests for comment.
The Los Angeles Police Department said in a statement Thursday that its Threat Management Unit was in contact with Musk’s representatives and security team but that no crime reports had been filed...."
I'm not even "anti" but I still voted for him to step down. The (now reversed) policy change banning advertising accounts on other platforms was completely out of touch, and the whole rationalization of his actions at the helm of the platform went down for me. I still agree with like 90%-95% of the changes, but I have to wonder if it was mostly luck, or if I am still rationalizing away bad choices.
If anything, I totally understand and support the banning of the jet account / new doxxing policy, you cannot create an actual marketplace of ideas if it's too easy to go back down to ad-hominems/meatspace threats.
It's actually the reverse for me, I started very anti-Musk in about 2015-2016, less "angry" per se and more annoyed, utterly exasperated and eye-roll-strained with the incessant "oMg guys he's jUsT like ToNy StaRK" screams, which completely misses the 2 points of
(1) Tony Stark started as an immature and immoral character stuck at 16 years old mental age, he gradually knew better and grew out of the "I'm amazing I can do anything I want" mentality by bitter experience. Iron Man is about how Tony Stark gradually learnt to not be Tony Stark.
(2) Iron Man is a hilarious/infuriating misdepiction of how technology and engineering work, at some point in the first movie Tony is like "oh yeah ? the suit freezes at high altitudes ? here's an alloy I just pulled out of my ass*, start coating the armor with it" just fucking no, this is not how any of this works. Real science and Real technology, like Real mathematics and Real programming and Real every knowledge field that matters, long ago stopped fitting inside a single person's head, no matter who that person is. This seems to bleed into Elon Musk fanboyism, where some fan boys think he's personally writing every line of code and welding every piece of metal.
I started softening on him a little bit when he began toying with Blue Tribe and flirting with all of its traditional taboos, I'm a spiteful guy and sometimes the fastest route to my heart is to be the enemy of my enemy. It was, I admit, highly cathartic to watch the people who memorized "It's a private company it can do whatever it wants" by heart suddenly discover that Free Speech is a generally desirable value and that Corporations are no less bound by it than States are, even if our current, very flawed, laws don't always reflect that. I didn't want those people permanently booted off twitter, just banned and toyed with a couple of days to make them learn their lesson.
That said, it didn't last long. I don't remember when exactly did Elon outstay my tentative second chance, but it was before the Jet thing, probably when he reinvented shadow bans and then acted like this is a genius solution that is better than outright banning somehow. The Jet thing is a camel straw. His internet-celebrity-style polls aka "dear followers give me a crumb of attention" are also very see-through and cringe.
But regardless, it's always funny and good to destroy a social media giant, so Elon got all of us to laugh a good laugh *and* made the internet's idea pool better by an immeasurably tiny amount, all at the cost of 44$ slightly-fictional billions. I call that a good deal.
* : The dialogue tried to justify this by implying that his company used the alloy on a satellite before if I remember correctly, but this is no better. Stark industries make thousands upon thousands of weapons, gadgets and systems, so how did Stark manage to remember the one alloy that will solve his problem ? and how does the top company scientists that the villian father-figure later commissioned to solve the same problem didn't remember or discover the same thing ? and how does an alloy made to work on an unmanned satellite in orbit just magically worked on a manned in-atmosphere flight system ?
I started off disliking Musk, particularly the techno-worship around him. I don't think he's much of anything, except I have to give it to him that he's been very, very successful at business.
But all the fraught reactions (Thursday Night Massacre, you know, like Kristallnacht?) are making me very, *very* begrudgingly like him.
It's the same way J.K. Rowling has fully leaned into the TERF thing (which has become an accusation like 'racist'): oh, so you say I'm a TERF? Okay then, I'm a TERF!
Slightly off topic, but I recently saw someone referring to a clearly anti-feminist conservative that said something anti-trans as a "TERF", and my reaction was "ah, we've reached the point where that term is starting to lose all semantic value, got it."
(For those following along at home, TERF stands for Trans-Exclusionary Radical *Feminist*. The last part is still relevant.)
Technically, "radical feminists" are those who believe that there are no differences between men and women... other than the obvious anatomical ones. (Which is why they logically have to reject the entire concept of "trans", because there can be no such thing as "female brain in male body" if there is no such thing as "female brain" in the first place.)
Of course, this should also exclude conservatives.
That trend started a very *very* long time ago, 5 or 7 years at least, where have you been all this time ?
As an anti-feminist, it's annoying as hell to me too. Feminists do not suddenly become ok when they oppose one particular wing of the wider ideology around them, if anything, TERFy feminists seem to double down on their misandry harder. Labelling them with the same label potential allies are labelled with is just more work for me to sift through.
Right, but my point is that its value as a distinct term was that it captured the concept of "virtue signals like a progressive but irrationally hates trans people". Without that, it's just another leftie slur to add to the pile of thought-terminating cliches.
Then why do people still say "TERFs and transphobes"? Are you suggesting it's just one of those emphatic pleonasms, like "rogues and vagabonds" or "far and away"?
Every term that denotes a philosophical position (doubly so for a term relevant to a currently-waged human rights struggle) becomes woolly and imprecise in its actual usage, but my perception is that the correct use of the term, referring to self-proclaimed feminists, still predominates. Maybe the process is beginning, as you say, but I haven't seen much of it.
I'm sure it depends on which echo chamber you're most familiar with. If you grew up reading magazines in your college's womyns resource center you'll use the original meaning. If you discovered the acronym on twitter, you'll never know it as meaning anything other than "irreedemable bad person who should be cancelled immediately."
TERF was originally an autonym, much like Social Justice Warrior. But it seems (and I'm sure someone can find counterexamples) that it's very effective to take an autonym and turn it into a smear.
I think Radfem hub is still active. And of course, the philosophy of "Feminism is a leftist class struggle political movement" goes back more than a century.
If you mean regarding the Twitter buyout, then no, I was opposed to Musk's plan from the start. (Though I also don't care about Twitter and think all the predictions about how the buyout will "change social media forever!" or "put democracy itself at risk!" are absurdly overblown. I also think Twitter itself is bad, has always been bad, and can't be made not-bad due to its very design, so on the off chance that Musk actually does cause Twitter to collapse entirely, I'll consider it a win, albeit one that was unintentional on his part.)
In a more general sense, though, I used to have a very positive opinion of Musk and his ventures, which has gradually gotten more negative over the years. The first time I started having doubts about him was during that fiasco with the trapped miners, when he accused one of the rescuers of being a pedophile, though I was give him the benefit of the doubt and write it off as a one-time oddity. Since then, various other things made my opinion of him lower considerably, though up until just a few months ago it was still more positive than negative.
I'd say the big turning point was when he started backing pro-Trump Republican candidates and parroting right-wing Culture War arguments. The Twitter buyout and the subsequent fiasco that followed were just the cherry on the cake.
Yeah, I suspect Twitter is very bad for Musk (and many other people)--it amplifies his worst qualities, gives him instant feedback from his fanboys cheering everything he says and does, and sort of builds a feedback-reward system in his brain for saying stuff that gets cheers. But Musk is actually a genius businessman. Training himself to follow what his fanboys want to hear him say is a terrible idea--he's smarter than almost all of them in the domains in which he has the power to act.
I think that just because he is running this poll, it doesn't mean he is serious about stepping down, regardless of the results.
He seems to like messing with people on Twitter who get all high horse with him. "Do you want me to quit?" poll seems like a perfect opportunity to get all the preachy types who finger-wag about him being a baddie going, and then crushing their little hopes with a "ha ha, fooled you!" afterwards.
If that was the intention, than he made it unnecessarily difficult for himself by using the following exact phrasing: "Should I step down as head of Twitter? I will abide by the results of this poll."
Sure, he still can change his mind. Although he has run several polls lately, and this is the first one he explicitely added the second half. It *would* have been much easier, to say, oh, haha, changed my mind or even: sorry, something really important new has come up and changed my mind, without this second sentence. So I assume, he at least planned to abide by it, when posting this.
Jeez, people are taking this seriously? In my imaginings, he just took another vape hit and achieved a cannabis level 7.2 perspective, a perspective from which running this poll looks like a deeply clever idea. In a while he'll slide down to level 5.8 or zoom up to 10.5, where instead he will think of a dick joke or a change to Twitter rules that looks awesomely awesome. And so on.
Twitter's financial woes are largely--I don't know the percentage, but it's significant--due to being boycotted by companies who oppose Twitter becoming politically neutral.
Why is there no counter-movement by people who support Musk, to boycott the companies boycotting Twitter?
Generally speaking, the companies that oppose "Twitter becoming politically neutral" are those that don't want their ads to be seen alongside far right content.
People who are actually politically neutral (ie don't care about politics very much) seem to either fall in the camp of "I don't care what politics are being shown" (and thus, don't care that far right content is being allowed back onto the platform) or "I don't like seeing any type of politics" (and thus, only care about far right content being allowed back on the platform inasmuch as it increases the chances that they have to see anything political at all, and would likely mildly oppose Musk's changes, if anything).
Those who were previously banned and are now allowed back on the platform, if they were the type of people who would participate in a boycott (ie willing to forgo some level of personal convenience to send a message; I think this is wildly less common than simply saying you participate in boycotts and then not actually doing so, regardless of political leanings) are likely already doing so due to advertisers pulling out of actually right-leaning platforms (Fox News/Breitbart/etc), and so if you're already ideologically not buying an Audi or General Mills products, then you can't feasibly "boycott more" when they also pull out of advertising on Twitter.
Companies that oppose Twitter becoming politically neutral, and companies that oppose Twitter becoming unmoderated, or close enough as makes no difference. "Unmoderated" is a sort of political neutrality, but not the only sort. If I'm in the business of selling bicycles or baby food or whatever, I probably don't want my advertisements showing up next to a thousand variations of "I hate you [expletive deleted] [deleted expletives] and I want you all to die!".
Elon's comments and behavior on taking over Twitter would give a reasonable businessman cause to wonder whether or not all that fun stuff would be going unchecked on the grounds that "well, he only said he *wanted* all those people to die, he didn't say he was going to *kill* them, so it's not a death threat so Yay Free Speech!" The sensible plan would be to put your advertising budget elsewhere for a few months and see how it plays out.
Also, an unmoderated free speech zone probably shouldn't expect to support itself by advertising.
But that should have applied to Twitter for all these years already. There were plenty of people calling for death on Twitter who didn't lose their accounts because they were smiled on by the left, like Ayatollah Ali Khamenei, who used Twitter to renew the fatwah against Salman Rushdie, and whose account wasn't suspended after Salman Rushdie was consequently nearly killed; or like Taliban members and supporters.
In any case, it doesn't address what I meant, which was, Why don't individuals boycott the companies boycotting Twitter? There's been a devastating and persistent refusal of the victims of Twitter mobs to retaliate against Twitter mobs, which I would like people to recognize is unproductive. This is the same thing on a larger scale.
He definitely should. The best Elon Musk company - SpaceX - is the one where he got someone talented to do operational control of it for him while he focused on "big picture" design goals and promotional stuff for it.
About a year ago, archaeologists digging at a site called Sayburc in south east Turkey uncovered an 11,000 year old wall containing some carvings of animals and people:
It is pretty easy to search and find articles on this, along with images. But for the reader's convenience, here are a couple:
Any guesses what these carvings were meant to convey? After so many thousands of years I would have thought there's next to no chance of being able to clearly relate them to any recorded myth or religion. But all the same, it's fun to try and guess.
For what it is worth, I think they might have been meant to represent rulership and justice, and the area where they were found was a sort of council chamber where the tribal leader and his senior counsellors or village elders met.
There are two apparent "frames". The first shows a standing man sholding his penis, and surrounded on either side by a pair of leopards facing him in a rather menacing manner. Perhaps the leopards were meant to represent his royalty, a symbol used down the ages for this, and the penis his fecundity. A king in primitive societies was expected to be fertile in his offspring, and to bestow fertility on the land, and if he failed in either, if for example a drought caused a crop failure, then the king was soon for the chop, literally!
Also, as one of the papers cited above points out, the figure between the leopards is wearing a two-braid necklace. Assuming that could be a crude form of regalia, this seems further evidence that the figure is supposed to be a chief.
The other frame, to the left of the leopards shows another man holding up what looks like a snake and facing a bull. I think in primitive societies a snake often signifies cunning and wisdom (c.f. the snake tempter in the Biblical creation story), and presumably a bull is meant to represent the opposite, i.e. bovine obstinate stupidity combined with mindless violence. So I suspect that frame is meant to represent justice contending with barbarism and primitive kneejerk responses.
We should remember that this site was one of the very first where communities had settled to start farming. So the inhabitants would have been fairly wild nomadic types and quite a handful for the ruler, and regulating them would have been fairly challenging, and thus a major if not the main preoccupation of whoever was running the place.
All that said, perhaps this symbolism is a bit advanced for those times, and the carvings just show some myth, long forgotten or transformed unrecognisably in the millennia since. For example, is it obvious the figure between the leopards is a man and not a monkey? He seems to have quite an ape-like face.
Hehe! There's plenty of stone age female "porn" though, some far older than 10,000 BC, those strange smooth looking carvings of ultra-fat women. FWIW, I reckon most of those, the smaller ones anyway, were used as dilators in female coming of age ceremonies. After all, child birth is pretty dangerous, especially that of a first born, and considering the young age girls those days often started bearing children.
The central figure reads to me as a sort of god, not a human ruler. After all, the human ruler is physically present, presumably very nearby, so you can just look at him with no need for carvings.
Notably, the "Bull" in the carving is either a recently domesticated cattle or a soon-to-be-domesticated wild auroch. This was carved around the time and place that humans first invented cows. I like to see the bull as an auroch, as wild and dangerous as the leopards (also present in the time and place, as far as I can tell)
So the mural depicts a local diety standing against the wild animals that would threaten a newly-sedentary society. This god has nothing to fear from leopards and aurochs. The phallus feels to me like a symbol of power. This god is virile and healthy.
I imagine whatever ceremonies or gatherings took place here were intended to encourage this diety to protect the village from wild animals.
Maybe whoever tamed the first aurochs was imagining this god when he did so, and gave thanks to him afterwards. Or maybe it already happened and the carvers believed that this god was responsible, and hoped he would do the same with the leopards
That all sounds pretty plausible, but in reading it I had a random thought: Do bulls, like some other animals, have an instinctive fear of snakes? If so then perhaps holding up a snake and wiggling it menacingly would have been a useful technique in controlling them: A furious wild two ton bull could be charging towards a guy, who would only have to hold up a dead cobra for the bull to screech to a halt in a cloud of dust and start hoofing it away in the opposite direction! :-)
Just a brief note: the snake and the bull are particularly associated with the Eteocretans or Minoans, pre-Greek inhabitants of Crete. Maybe some connection there; I think genetic sampling implies the Eteocretans came from Anatolia.
Snake and bull would also be Shiva/Rudra as Pashupati, "Lord of Animals" (the snake is associated in various ways with Shiva and the bull is Nandi, his mount):
There are a lot of "Lord of Animals/Mistress of Animals" motifs throughout Indo-European and indeed all mythologies. The famous seal from the Indus Valley is one:
Can I be the one to say that I'm not convinced that he's even holding his penis? He might be holding something else and archaeologists are just a bunch of perverts. The people of eleven thousand years ago would be scandalised to find out that they spent all that time on a beautiful carving of a man holding the sacred stone of whatever and we just think it's his dick.
Perhaps the object is more obviously penisy on closer inspection than it looks in the photos, though, so I guess I'll have to take their word for it.
"The people of eleven thousand years ago would be scandalised to find out that they spent all that time on a beautiful carving of a man holding the sacred stone of whatever and we just think it's his dick."
Probably not. Bronze Age petroglyphs from the same era are full of little stick guys with A, spears and B, humongous boners. Penis shame is a fairly recent thing culturally, and unlike most things blamed on Christianity, it can probably be blamed on Christianity -- before that ol' popery came along, Romans of both sexes wore little dick-shaped bell pendants on necklaces as good luck charms (look up "fascinus"), and one of the religious tasks of the Vestals was to care for the Phallus of the Roman People.
In other words, at worst the people of eleven thousand years ago would probably just think it was hilarious that we misinterpreted the <carefully carved object> as a dick.
Is it true, or just an artifact of attributing nature to nurture? Is it even scientific in the sense of having testable predictions? Are there good studies about it?
In practice, it seems like astrology for people who took an undergrad psychology course (that is, the version commonly used doesn't seem to have much attachment to reality beyond "sometimes people are like this).
Possibly there's a much more carefully hedged version that academics know that makes testable predictions with nontrivial accuracy, but if so none of the attachment theory proponents I've met have managed it give an example.
Not entirely disagreeing with you here but can't this be said of any scientific (especially in psychology) theory? Almost a motte-and-bailey where science lies in a nuanced and carefully articulated motte and the popsci version of it leaks out into a field of vague assertions.
A quick scan of the wikipedia article linked me to what seems like a pretty rigorous large longitudinal study ascertaining a relationship between the attachment style during childhood and clinical mental health outcomes later in life - https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3861901/
Thanks, that's interesting (and an actual specific claim!).
(Mildly skeptical of the result in this paper based on complexity penalty (the predictive factor there seems fairly specific and with p hacking potential), but haven't read the paper through and giving it props for making a specific claim that an attempted replication could disprove).
For the popsci version though - I think the popsci claim is "people have an attachment style that reflects how their young relationship with their mother was", but since attachment styles seem to be strictly ranked (avoidant is worse than non avoidant and insecure is worse than secure), it seems to just boil down to "if you had a bad relationship with your mother your relationships in the future will also be bad [in similar ways]", which seems false-ish (partly because people's bad relationship dynamics are generally not even consistent between their own relationships).
I think a distinction your summary misses is that attachment theory would only predict that if your relationship with your mother is bad in specific ways that this would have impacts on your future relationships.
So I kind of elided this point, but I think it's a weakness of the theory that its description of the ways in which that relationship would be bad is pretty vague and flexible.
Isn’t it the case that there are many phenomena that have pretty divergent results? An auto-immune disease, for example, can show up with any number of different symptoms and pathologies. I tend to think of attachment in a similar way — that poor attachment inhibits a person’s ability to trust future relationships, and that inability to trust manifests in lots of different relationship disfunctions.
Hmm. Wasnt there an experiment where a monkey was brought up with in a cage with a mechanical food feeder, and a soft toy or something he could cuddle. Pavlovian theory would suggest he would attach to the food giver but he cuddled the toy.
Also scientifically you could test with children sent to cold institutions - are they less socialised and happy?
As I understand it, attachment theory's claim is that the monkey with the mechanical feeder would grow up to have bad relationships later in life. Open to seeing this experiment done but I'm going to guess it doesn't pan out.
> In an experiment called the "open-field test", an infant [monkey] was placed in a novel environment with novel objects. When the infant's surrogate mother was present, it clung to her, but then began venturing off to explore. If frightened, the infant ran back to the surrogate mother and clung to her for a time before venturing out again. Without the surrogate mother's presence, the monkeys were paralyzed with fear, huddling in a ball and sucking their thumbs.
> Another study looked at the differentiated effects of being raised with only either a wire-mother or a cloth-mother. Both groups gained weight at equal rates, but the monkeys raised on a wire-mother had softer stool and trouble digesting the milk, frequently suffering from diarrhea. Harlow's interpretation of this behavior, which is still widely accepted, was that a lack of contact comfort is psychologically stressful to the monkeys, and the digestive problems are a physiological manifestation of that stress.
I actually see this as a minor point against the popular conception of attachment theory. It suggests that despite complete absence of parents, a teddy bear was enough to give the monkeys a secure attachment style.
Yes, but attachment theory claims much more than this, and you do not need attachment theory to explain those findings. This seems somewhat analoguous with the nature/nurture-results that yes, children who are clearly abused and treated horribly will have long-lasting negative effects, but at the same time the normal variation in child-rearing doesn't seem to affect much to any outcomes we have been able to measure. The different attachment styles and their claimed effects fall to the latter "normal variation".
It seems fairly probable that lack of secure bonding in very early childhood has long term effects.
I suppose, teasing apart, the physical deprivation from the emotional deprivation, could leave room for discussion. Given the nature of people and our general need to bond with others I don’t think it’s a stretch to imagine that one’s first experience with that negotiation can have long-term affects.
I don't know that article seems like a cloud of things that are truisms and brute tautologies about what all those words mean, but mixing in some comments and predictions about things that I suspect are false.
For example the idea that there is just one attachment seems sort of wrong, or that it is tied to high quality social contact, more than quantity.
I would say for both my boys other than nursing, the "high quality social contact" was 100% more from me from ages up to ~4. I love playing with small kids, and we spent tons of high quality time on the floor of their room. Nevertheless while we have great relationships, they are definitely both more attached to their mother. Both because of the nursing, and because she is easier on them and more "mothering" of them. I am not resentful of that or anything, it is jsut the way of biology I suspect.
I do think that as the theory goes into, kids without a mother to attach to or a real poor quality mother may attach onto others, but I doubt it is often just one, and likely is a whole host of people depending on the exact balance of socializing over time.
The idea that kids with poor parentage and poor attachments as youth develop "non-standard"/pathological ways of socializing and attaching as adults seems both true, and a pretty mundane/boring/uninteresting observation.
i certainly had a crummy childhood, and that certainly is a part of some abnormal ways I relate to others compared to most, but I don't think you need some fancy "theory" to make any connections there.
IDK that "theory" seems just as much noise as signal.
My current theory about nurture is that while childhood experiences don't seem to *determine* future outcomes, they do seem somewhat *causal* in a roundabout sense
I mentally attribute all my personality traits to my parents actions, but half of them are "I was raised this way so obviously I'm like this" and half are "I was raised this way and it always bothered me so now I compulsively do the opposite." I think this is common enough. An obvious hypothesis is that our parents have nothing to do with it and it's just a natural but false assumption. An alternative hypothesis is that humans really do naturally build our personality around our childhood experiences with our parents, but there's a 50-50 chance (perhaps determined by genetics) that any given experience will result in us forming one personality trait or its opposite. This would still result in a lot of studies showing that parental quality "doesn't matter," which would arguably be true.
Its not entirely clear to me the two hypotheses are distinct. In either case, I can see the value in talking through with a psychologist the ways in which your parents have shaped your current behaviors. It's a real and natural mental construct, and changing your understanding of it could change your personality. But parenting still isn't predictive of future personality.
My view, as a psychologist: There aren't very many things that are all nurture, and it is very unlikely that this is one of them. I am sure wiring plays a part. Many people who are later diagnosed as autistic resist cuddling as infants -- they arch they backs and struggle. And I betcha if you measured things like how much time infants log looking at faces, and how they respond to cuddling you'd find a bell curve, with autistic infants at one end. On the other hand, it seems implausible that infants whose parents are absent or just cold and distant would not be affected, even if born wiring that gives them good capacity for attachment. There was a study done that so far as I know is still well thought of called the Still Face Experiment. In it, mothers were asked to interact in their usual way with the baby for a few minutes, except that they were to keep their faces absolutely blank. Infants reacted very powerfully to this, first increasing the behaviors that usually engaged their mother, then becoming frantic and howling with distress. As I recall, the effects continued to be visible for a few hours. I'm sure a chronically tuned-out, unresponsive parent would have a substantial negative effect on their infant.
On the OTHER other hand, seems unlikely that what happens with family is the only thing that can make a long term difference. Once kids are spending a lot of time with other kids the quality of kids' relationship with peer group is enormously important. I have had many patients who had loving, involved parents but bad experiences with their peers, spending years between age 8 & 18 being bullied or just friendless. They ended up avoidant, mistrustful, full of quiet rage at most others & convinced they were repellently weird. So I believe strongly that someone's attachment patterns can be greatly affected by experiences other than and later than the parent-child relationship.
Sacrilegious commentary ( following "the rest is commentary " line on the main Astral Codex motto ( smart and undoubtedly attractive and self filtering for a certain kind of audiences ). Deliciously subversive . Not necessarily universal, and
assumingely universal.
P(B), is in practice, harder to determine than one may assume.
It is "just" the denominator and hence merely a scaling factor.
Nevertheless, this 'hard' guessing of P(B) in most applied settings, makes Bayesian inference a relative and not absolute matter: the equal sign is typically replaced by a "proportionial to" symbol.
For example, if If A is lung cancer and B radon exposure, one may have a pretty good idea of the prevalence of lung cancer in the general population i.e. P(A).
One may also be able to estimate, from appropriate sampling the radon exposure history among those who have lung cancer i.e. P(B|A).
However, to estimate radon exposure in the general population is harder -> what is an adequate sampling frame for non lung cancer individuals?
Probability is not my strong suite in the slightest, but in all the settings that I have seen Bayes' theorem applied, the denominator is either :
1- Handwaved away, because we're only *comparing* probabilities, so only sign and relative magnitude matters.
2- Asserted to be calculable from the normalization condition (all probailities sum/integrate to 1), this is essentially the same as (1) : if you have 2 "semi-probabilities" that are 4 and 6, you can either compare them directly or - if you really want to - normalize them to true probabilities by asserting they sum or integrate to 1, and thus the underlying true probabilities are 0.4 and 0.6. It's simply a change of units.
3- Asserted to be calculable by the total probability theorem : P(B) = P(B|A)*P(A) + P(B|~A)*P(~A), where ~A is the event "not A".
Of course, for option 2, we need to be sure that the two options are the only options: if in addition to the 4 and the 6 we have a 90 that we've forgotten about, then we get 0.04 and 0.06 instead when we normalize.
(A detail, but the sort of detail that the devil proverbially resides in.)
Some more thoughts about why the crypto post bothered me so much: debates over the object-level correctness of its arguments aside, it basically pattern-matches to the arguments crypto scammers (which Scott admits are ubiquitous) make for getting money from people. I don't think Scott himself is a crypto scammer, but he does have a lot of reputation as a reasonable guy and it seems likely that him putting his weight behind arguments like this helps make people more vulnerable to them.
There's a counterpoint here that we should judge arguments based on their object-level merits, not whether they pattern match onto stuff bad people say (and while I do have some object level disagreements, I admit I'm not making them here and don't consider them overwhelmingly convincing). I think this is wrong: In cases where you know there are ubiquitous successful scammers, you should avoid trusting your object-level intuition and knee-jerk reject things that match to them if you don't have good reason to think you're better at detecting scams than they are at convincing you of scams (if you can't reliably outperform the rock with a sticker saying "crypto is scams", you're better off just listening to the rock). And as the FTX stuff showed, both Scott and a lot of other people in these parts are not reliably good at detecting crypto scams - probably good enough to detect the obvious ones, but crypto has a lot of really talented and hardworking scammers, and you're still more likely to be convinced by them than to chance into a non-scam based on object level reasoning.
"And as the FTX stuff showed, both Scott and a lot of other people in these parts are not reliably good at detecting crypto scams"
I didn't weigh in on the other thread because I got to the post too late, but this was my main takeaway -- that Scott has apparently not updated nearly enough on seeing FTX collapse to shit.
What part of the FTX debacle was due to the obviously crazy practice of just funneling FTX monies into Alameda and not setting up really anything like a robust exchange and what part was due to "crypto"? I personally weight the scam part of FTX pretty high (>50%) in terms of why it all collapsed. The fact that it was crypto was a pretty small part of it. Maybe crypto attracts this behavior?
A huge part. The Alameda money shuffle was based on:
1) FTX minting their own crypto token, then keeping the vast majority of available tokens for themselves
2) Scamming regular people out of their real money by trading it for their worthless tokens (this is how regular people lost money on the crash)
3) Taking that real money to fund Alameda
4) Selling a bunch of their trashtokens to Alameda in exchange for real money
All of this shit served to pump the apparent liquidity of both companies, which would not and could not have happened without the fakery of a cryptocurrency, inventing false money out of thin air. Crypto is the cornerstone of both these castles in the air.
(This is a very, very brief encapsulation of course, so most details and specifics are left out.)
2) Regular people just deposited their money with FTX, thinking that it would operate like a responsible bank (except with very high "no risk" return because wishful thinking). These customers didn't actually get tokens in return for their deposits, but rather simply a balance in their favor. Even if they did get tokens, it doesn't seem like the heart of the scam.
4) The main scam wasn't selling tokens to Alameda. FTX simply took money it was supposedly safekeeping on behalf of others and ran away with it. Yes, Alameda got a bunch of tokens. But the purpose of this was inflating its assets on paper, so that FTX could then pretend to take that inflated number at face value and give Alameda money based on this imaginary "collateral". But Alameda doesn't seem instrumental here. FTX wasn't regulated as a bank, so SBF could have just taken off carrying a huge sack with "$$$" written on it, and it would have been the same result.
So yes, there was a castle in the air. But this was used for nothing more sophisticated than tricking gullible people into thinking FTX was solvent / trustworthy. Any other fake source of legitimacy would have fulfilled the same purpose. It's a classic con (= confidence trick).
This is also consistent with the fact that FTX spent large amounts of money propping up some of their bad investments. Again, all about projecting an image of being successful.
As much as I think crypto is a huge scam and the correct market price of bitcoin is 0, the FTX scandal is only tangentially related.
Seems that you're right about 2) and I accept the correction unreservedly. The high "no risk" return is the big blinking four-story-tall neon sign that says SCAM if you just bother to read it, though.
4) I don't think is correct. The money/tokens switcheroo was a way to inflate *both* companies' value plus a way to convert consumers' money from theirs to FTX's without it just looking like naked theft.
"But Alameda doesn't seem instrumental here. FTX wasn't regulated as a bank, so SBF could have just taken off carrying a huge sack with "$$$" written on it, and it would have been the same result."
In that case, why did he bother with the whole Alameda setup? A burlap sack can't be more expensive than an LLC. Why complicate things needlessly?
No, I think it's clear that the money shuffling at the very least allowed SBF and his cronies to hold façade and collect more money from the gullible while living it up in the Bahamas.
"Any other fake source of legitimacy would have fulfilled the same purpose."
I don't think there's currently any fake source of legitimacy equivalent to crypto. It's the South Sea Bubble of our time, you can't just switch to a West Sea Bubble or some shit.
It's possible it was one of those "slip into becoming a scammer without meaning to" scenarios. Going from genuinely running FTX to "I just need to take this money here and put it all on black, that'll save my hedge fund and no-one needs to know", etc, all the while thinking themselves too special to worry about things being "technically illegal" because after all "they mean well".
From my very superficial impression, SBF sounds exactly like that. He's a psychopath. Low empathy, low impulse control. Good at saying all the right nice phrases. And unlike in TV shows, very little self-insight. Even after the whole thing collapsed, he was going on interviews and saying "Oh shucks, I'm trying to raise money to help out our customers as much as possible" as if people would somehow forget that the only reason the customers need to be "helped out" is the crimes he committed.
FTX's scam was not honoring deposits. Any other "fintech" company, say Paypal, can pull the same scam. No difference between spending customer deposits on blow and gambling on cryptos and blow and gambling in Vegas. The only difference is that crypto scammers somehow where succesful in normalizing trusting hundred-thousand-plus dolars deposits in the exchanges when people feel iffy about keeping 100$ in their fintech app of choice that's actually regulated or at least backed by companies with actual assets to go after.
I believe that's why the scam needed the fake inflated crypto to work. Putting your money in a company worth a few million feels less safe than putting your money in a company worth billions that's making so much money they don't know how to spend it all. If anyone noticed that a significant amount of the value of both FTX and Alameda were solely their own made up coins, that confidence would have dropped a *lot* and people would have pulled their money/never put it in to start.
It's like scamming 101 here, it was not a Nigerian farmer trying to give you free money, but a Nigerian Prince! He doesn't really need the money, so he's willing to help you out through his largesse! That's why you can trust him, when you would normally be concerned about giving out your bank account information to a stranger...
This analogizes perfectly to the internet during the 01 bubble -- companies attached “dot com” to their names to pump their stock, the field was full of scams. Good thing we abandoned that area
I don't think the rates are comparable, this is like comparing alcohol and fentanyl. Dotcom era and crypto both have scammers, but dotcom had a significant chunk of companies with sound underlying business structure (and thus was able to have a much larger fraction of people who genuinely, if wrongly, believed that their underlying business was sound or soon would be). Since crypto basically has no way to be profitable except hype, it's almost pure hype people.
What evidence do you have for the claim, "crypto has no way to be profitable except hype"? Why do you believe that? Ethereum, for example, generated about 200M of revenue in Q3 2022.
A huge amount of the FTX fraud relied specific on the mechanisms of crypto, the whole industry's culture (which is full of scammers and pump and dump, and pyramid schemes, and get rich quick suckers), and the lack of regulation in the area. All of which are issues which carry over to other projects.
Scott seems to be trying to define FTX and the collapse to a different category from "crypto" and is actively updating a different set of domains instead. I agree that this is a type of failure mode for him.
Although I agree that FTX (and MtGox before) are not exactly the same thing as "crypto," there is a lot of overlap in people and practices, and that neither of those scams/collapses was possible without crypto.
That would be looking at a different set of financial categories. Indeed we should evaluate all investments based on how often or likely they are to fail. The answer in 2008 was "more likely than many thought" but not the apparent rates of fraud and loss in crypto, which seems to me closer to "you knew what you were getting into here, don't come crying to me!"
I'm seeing enormous amount of financial scam ads in Gmail and YouTube. This is probably related to the fact I write and consume content in Russian, and Google stopped caring about it.
Still, these ads are not crypto ads, but they are all scam. Banking, investment. Like 90%.
So, it would be hard to compare arbitrarily. There is a lot of scam out there. I even fell to scammy US entry documents submission website once.
I don't really understand the whole "Crypto Scammers" angle. Scammers infest a whole lot of things, I subscribe to just 2 or 3 "Reverse-Scammers" youtube channels (people who scam scammers) and its astonishing how much they post. Are there people against reddit or facebook ? Those 2 together probably host something like 50x the scammers crypto can ever hope to attract.
I'm not convinced crypto is a game changer for scammer, at most it's just an "all in one" solution that does the same things as gift cards or international bank transfers, just (very slightly) better. The best "Crypto is for bad people to do bad things" argument for me is Ransomware.
I would absolutely tell anyone to avoid a "get rich" or even just investment businesses they found marketing to them over reddit or Facebook like the plague. So yes it is the same.
What Blank said. Remember that whole Gamestop thing? Those guys weren't even scammers, just trolls who had expanded to trolling the financial system, and there were *still* a whole bunch of imbeciles hopping on that train thinking they'd get rich. I did not have a palm big enough to bury my face in. Don't take investment advice from fucking Reddit. Especially if they tell you to ride Dogecoin to the moon.
(Now imagine if Scott and other seemingly sensible people had been talking about how HODLing was actually a viable business strategy and actually Gamestop really was criminally undervalued, etc. etc. I would try to say something then, too.)
But a short squeeze is absolutely a legitimate way to make money, large Banks do it all the time whenever they notice short squeezable circumstances, and the people shorting GameStop were absolutely on the hook for more gamestop stock than existed.
Is your argument that random people on Reddit are incapable of coordinating well enough to pull off a short squeeze? Or is your argument that short squeezes themselves do not earn money?
My argument is that the Redditors consisted essentially of two cohorts: one that was coordinating to short-squeeze, not to make money but to ratfuck the hedge fund guys who'd done a Producers on the stock out of some quasi-leftist eat-the-rich sentiment (although that one guy did get rich and then investigated by the SEC and all that; never really paid enough attention to find out what ultimately happened to that guy, or even if he's gotten out of the Modern Times-esque machine of grinding gears yet), and another one that couldn't even spell short-squeeze to save their mamas' lives and who thought the stock would just keep going up indefinitely and they should hang on to it way past the date when the original shorters needed to be covered. This second group was boosting and encouraging one another like idiots and did not have anything resembling a sensible strategy; they were just the regular rube hoping that this time, *this time* he'd gotten in on the ground floor of the get-rich-quick scheme.
Investing in complicated cryptocurrency stuff is also an entirely legitimate way to make money that can allow you to make a bunch of money, but can also help you lose your shirt.
Yes, there are tons of scams all over the place. I have a standing rule never to buy anything (or share personal information, etc.) to anyone who reaches out to me - phone, email, whatever - rather than me reaching out to them. If I get a call from a company that I recognize and should be open to dealing with but don't know the person I am talking to (picture VISA or my bank calling about a recent purchase), I hang up and call their regular number. So yeah, there are lots of scams going around. We should *not* update to think only crypto is full of scammers, but that crypto is *also* full of scammers.
That said, there's also something to the percent of crypto-related "investments" that are first-order scams (intentional fraud) and even more so second-order scams (so poorly thought out and run that the results are the same). I've said before that I think FTX was likely in that second group. What happens to your money when you invest is the same as if it were intentional fraud, and there's often criminal activity also involved, such that the results are the same and they should be treated the same. Regardless of the type of scam, there have been roughly two crypto-currencies that seem well-regarded enough that they might not result in everyone losing their money (BC and Eth) and I'll note that both of them have had technical or other issues that resulted in groups of people losing their money, and that ultimately the best we can say about them is that the jury is still out on whether they succeed. Crypto has only existed at all for about a decade. If you base your investments on who can survive a decade, you should go read about the 1920s stock market.
That post was full of painful incredulity. Ignoring, e.g., that many countries using crypto are using it for its scam purposes and focusing on self-serving myths about our swell middle-class brothers in other countries using it to escape horrible financial situations.
That crypto just coincidentally (rhetorical incredulity? The worst kind) turned into an investment asset and that this fucks with its usefulness as a currency.
Honestly, though, this crypto debacle keeps shaking down the most obnoxious people in the upper middle class so keep at it.
That last part is why I'm all for crypto -- the whole forest of schemes is sopping up an enormous amount of dollars, of which we have way too many, and in some serious fraction of cases it's not only sequestering the cash, it's vaporizing it for good.
This is all good from the point of view of reducing the amout of money in circulation, which is what is needed to tamp down inflation. Also since people are sending their dollars away to be vaporized voluntarily, in exchange for promises of gigantic wealth by and by, meaning the money is coming from people who can afford it and who are willing to believe you can get Something For Nothing, or that FOMO is a viable investment strategy, or who are planning on profiting from a financial collapse that destroys everybody else -- none of which illuminate character traits that one might admire -- this is much more socially fair than brutal interest rate increases or tax hikes, which would hurt a wider class of innocents.
Unless someone loses their wallet, in which case that BC is lost and the only return value is the cash exchanged for it at its previous value. i.e. if it was purchased for $100 each, and then the wallet got lost, it's only value was the $100 each, not the five figure value that BC was later valued at but can no longer be sold for.
(Sorry for making generalizations here, please add qualifiers as needed.) A white American soccer mom, Moscow Mayor Sergey Sobyanin, Iranian dissident Badri Khamenei and a Buddhist monk in Jagadhri have some common traits due to their shared indo-european-ness: Their words for "mother" and "milk" sound alike, and they all would probably agree that there is an old thunder good with a hammer or mace. But do they have anything inherited in common in regards to daily life practices, traditions or beliefs about the world (that say a Native American, Chinese or Pygmy person wouldn't have)? Some candidates:
1. They would all agree that a bowl of yogurt is an acceptable breakfast? (But other pastoral nomads (e.g. mongols) likely agree?)
2. They would all agree that white is the most appropriate color for a horse that a king will ride? (I'm not sure they would all agree to this though.)
3. There's some kind of festival in the spring involving large bonfires? (But large fires doesn't seem specific enough.)
4. Cremation is acceptable? (but Catholics doesn't agree, and Japanese do.) Maybe "The spreading of ashes after cremation is acceptable"?
Please suggest your own ideas! But likely this is a fools errand: indo-european cultures have diffused too much and practices have been exchanged too much with e.g. China so that nothing like what I'm looking for remains.
Words similar to "milk" or "mlq" are found in languages all over the world, likewise water/aqua or words sounding vaguely like "qwa", although in some languages the words relate more to throat and swallowing. Some people think these basic words long predated Indo-European and were actually among the first onomatopeic words ever spoken by humans!
One possible answer to Medievil Cat's question is to try comparing commonly used words in various languages meaning "crane fly". I read somewhere that names for these in a lot of Indo-European languages involve "father" or "ancestor" or "man" in some form, such as "duda", the most obvious one being the English name "daddy long legs"!
Believe it or not, that is because in pagan times crane flies were believed to embody the spririts of the ancestors. One can see why, given their skeletal spindly appearance and possibly in part the buzzing noise they make as they lurch around, as if trying to communicate (although to be fair, a lot of insects buzz!)
But the clincher for our superstitious ancestors was their mass appearance at around the time of Samhain (1st Nov), and disappearance shortly thereafter, due to their short lifespans. This was the Celtic festival of the dead, at the start of winter, and was Christianised to All Souls Day, AKA Halloween, when the gates of the Underworld opened for a while and the spirits of the departed were free to wander around among the living.
I think the meaning was "Their words for 'mother' sound alike, and their words for 'milk' sound alike", not "Each of them says 'mother' with the same sounds as 'milk'."
It isn't actually true that the words for "milk" sound alike. The words are related, but you have to prove that by comparing large sets of words in each language; the words by themselves prove nothing and, as mentioned above, aren't even similar to each other. Compare Greek "gala" with Spanish "leche". Those are true cognates!
Also, the Germanic and Slavic forms represented by "milk" are not related to the Latinate or Greek forms represented by "gala"; Germanic draws from the Indo-European verb for milking an animal, while Latin and Greek draw from the Indo-European noun referring to the substance.
Indo-Iranian languages seem to use completely different words.
You may look at common colour codes - like red for king/violence/love, maybe white for purity, black for bereavement. Not even sure if this work with India though.
Ok with your edit - historically yeah. But ostensible cremation taboo has Jewish and Semitic (Egyptian) origins.
But back to the OP idea, to even mention Catholicism in quest to identify some "ur" Indo-European culture, seems off the mark (why one would even imagine that there is such a unique thing I don't understand) given that it is only 2000 years old.
Historically no, because it was adopted by non-believers as an explicit denial of the resurrection of the body: I don't believe in any of that, I don't believe I will be resurrected, so I opt for cremation as the tidier and more modern way to dispose of my remains.
Now that cremation is widespread and does not have connotations of "deliberate denial of the resurrection", it is permissible even for Catholics. A change in attitude from the 1929 Catholic encyclopaedia on the topic:
"The legislation of the Church in forbidding cremation rests on strong motives; for cremation in the majority of cases today is knit up with circumstances that make of it a public profession of irreligion and materialism. It was the Freemasons who first obtained official recognition of this practice from various governments. The campaign opened in Italy, the first attempts being made by Brunetti, at Padua, in 1873. Numerous societies were founded after this, at Dresden, Zurich, London, Paris. In the last city a crematory was established at Pere Lachaise, on the passing of the law of 1889 dealing with freedom of funeral rites. The Church has opposed from the beginning a practice which has been used chiefly by the enemies of the Christian Faith. Reasons based on the spirit of Christian charity and the plain interests of humanity have but strengthened her in her opposition. She holds it unseemly that the human body, once the living temple of God, the instrument of heavenly virtue, sanctified so often by the sacraments, should finally be subjected to a treatment that filial piety, conjugal and fraternal love, or even mere friendship seems to revolt against as inhuman. Another argument against cremation, and drawn from medico-legal sources, lies in this: That cremation destroys all signs of violence or traces of poison, and makes examination impossible, whereas a judicial autopsy is always possible after inhumation, even of some months.
...In conclusion, it must be remembered that there is nothing directly opposed to any dogma of the Church in the practice of cremation, and that, if ever the leaders of this sinister movement so far control the governments of the world as to make this custom universal, it would not be a lapse in the faith confided to her were she obliged to conform."
Some comments have the ability to summon a wild Deiseach from the limbs where they live on a diet of potatoes and Church Fathers. That's one of the best feature of this substack.
That’s always puzzled me: That the most small-c-conservative among both Catholics and Protestants are the most against cremation.
One of my fellow Protestants want on a long rant making the very same argument about “denial of the resurrection.” Were I up for an argument, I might suggest that not one molecule of Abraham’s body is intact, having long since been eaten by bacteria and crumbled to dust.
Given a belief in an omnipotent god, and in the bodily resurrection, this should present no especial difficulty as miracles go.
No, I cannot imagine a falsifiable hypothesis that could be written, nor an experiment devised to test it. That’s why it’s faith and not science.
"Were I up for an argument, I might suggest that not one molecule of Abraham’s body is intact, having long since been eaten by bacteria and crumbled to dust."
It does depend on whether you're arguing "it is as easy for the resurrection of a cremated body as one that has been conventionally buried", or "it's all bullcrap". Cremation was at one time the equivalent of those Westboro Baptist protests at funerals. It also had connections to pagan customs of disposing of the dead.
And the dignified treatment of the remains is also a concern; there have been allegations of people disposing of cremated remains in fits of pique or anger due to quarrels with family members by dumping them or flushing them down the toilet, etc. Much harder to do that with a body buried in a coffin!
Cremation now is not associated with any particular anti-Christian stance, which is probably why you find it puzzling. But there's a difference between, let us say, eating pork and bacon because you enjoy it and see nothing wrong with eating it, and trying to force others to eat it precisely because they believe it to be wrong, like the stories in 2 Maccabees:
"Eleazar, one of the foremost scribes, a man advanced in age and of noble appearance, was being forced to open his mouth to eat pork.
But preferring a glorious death to a life of defilement, he went forward of his own accord to the instrument of torture,
spitting out the meat as they should do who have the courage to reject food unlawful to taste even for love of life.
Those in charge of that unlawful sacrifice took the man aside, because of their long acquaintance with him, and privately urged him to bring his own provisions that he could legitimately eat, and only to pretend to eat the sacrificial meat prescribed by the king.
Thus he would escape death, and be treated kindly because of his old friendship with them."
"It also happened that seven brothers with their mother were arrested and tortured with whips and scourges by the king to force them to eat pork in violation of God’s law.
One of the brothers, speaking for the others, said: “What do you expect to learn by questioning us? We are ready to die rather than transgress the laws of our ancestors.”
...After the first brother had died in this manner, they brought the second to be made sport of. After tearing off the skin and hair of his head, they asked him, “Will you eat the pork rather than have your body tortured limb by limb?"
As to the molecules of Abraham's body, we find human fossil remains from hundreds of thousands of years ago, so who knows but that some bones of Abraham remain somewhere? 😁
Mine will be interred, unless I think of a better place to scatter them.
How someone chooses to deal with the remains of their ancestors is pretty low down on my list of things to worry about. Remains not existing is not a problem for the resurrection.
On the other hand, “Abraham’s body does not exist,” is a falsifiable hypothesis, so probably does fall into the “Science” bucket, though the DNA would be challenging to sort out.
For some time, Jews were opposed to cremation because of its association with the holocaust, but I think some of that has faded.
Where is the quote from? In general, it's allowable (required?) for Jews to break most of Jewish law to save a life. There are exceptions, but I'm not sure whether coercion (rather than, say, no other food available) would count.
Believe? The bits that overlap with the Apostles and Nicean. Creeds, including the Fileque clause, with a fairly important distinction that our “catholic church” isn’t capitalized.
Even within Protestantism, we will argue endlessly among ourselves about the details.
I'm not saying that indo-europeans are culturally unique. I'm saying that it would be interesting if there's some common feature that still remains (beyond the language and the old myths). The mechanism for that would be cultural inheritance over a large set of cultural trait, with maybe some trait out of thousands surviving rather unchanged in all diverging cultures largely due to chance.
It would be interesting to search for a needle in a haystack only if you were reasonably assured that there is a good reason that there might be a needle in the haystack.
Culture is broadly speaking shared symbolic thought.
It would seem that "differences" in "culture" arise from
1. geography
(available food for example, or
relationship to a place (the mountain or water or the plain) or
relationships to what might be seen for example the flora and the fauna and the constellations)
or whether farming is viable and
2. object-making technology
(e.g. the production of music making instruments,
crafts and art,
ways of cooking,
transportation,
weapons).
Language, a particular kind of symbolic activity, obviously must be shared (or of what use would it be) but it is there any evidence that syntax creates culture? I don't think so.
Vocabulary is an outgrowth of geography and the things named - what you see and experience.
Nor does vocabulary create culture; vocabulary documents reality.
Why would anything other that what universal to all of humanity persist as soon as there is movement or expansion to new geographies or encountering/developing new technologies?
Is there any evidence that language diffusion happens at the same rate as cultural diffusion.
There might also be so-called "cultural" differences due to the shared relationship to the transcendent. But in the scheme of human history that is pretty recent.
>Why would anything other that what universal to all of humanity persist as soon as there is movement or expansion to new geographies or encountering/developing new technologies?
Because people inherit culture from their ancestors, even as geography and technology change. Isn't this obvious?
An American soccer mom and a roman senator would both understand what a wedding ring is, even though they are separated by geography and technology.
I assumed that indo-european languages have been spread together with an indo-european culture, I should have been more explicit with that.
A Hungarian American is likely anglicized enough to count as indo-european for this experiment, but lets discount them as a borderline case just for the sake of it.
I'm from the same culture as the soccer mom, and bonfires strike me as more of an autumn thing than spring.
The only example I can think of of a thunder god with a hammer or mace is Thor. If this were a broader Indo-European thing, wouldn't Zeus have a hammer too? For religion more broadly, I think you'll run into problems because most Indo-Europeans outside of India have adopted Semitic religions.
Arguably Christianity is as much Indo-European as it is Semitic (if it even makes sense to talk in those terms) since it has lots of Hellenistic influences (this is arguably true for Judaism as well to a lesser degree). Paul's letters and the gospels were all written in Greek. Islam inherits from Christianity and Judaism, and the Persians have their own version of Islam anyway. So maybe some common vestigial shard remains?
It tells you specifically and solely that there are more examples of a thunder god with a hammer or mace than the one example Bullseye could come up with offhand (for which I do not blame him, I hasten to point out lest this go in some weird accusatory direction). That was the question he implicitly posed, so that was what I replied to.
Yeah just saying really for the vast majority of history the main three weapons were club/spear/bow. Eventually a sword got added once there was copperworking.
That some gods had a club seems pretty uninteresting.
Why are so many gods associated with fruit? Because people eat fruit and like it.
Pretty sure the axe and knife came long before the sword since both can be knapped out of stone, but besides that, sure. I was only trying to supply the information requested, not in this instance tie it to some wider argument.
The various Indo-European pantheons do have strong similarities, but they are pretty subtle and scattered by historical changes. Thor shares his name's Indo-European root his Gaulish equivalent Taranis, maybe the Hittite Tarhunta, and of course with the English word "thunder". IE thunder gods usually slay giant serpents: Zeus kills Typhon, Thor fights Jormungandr, Indra slays Vritra, and Tarhunt kills Illuyanka. Often they are part of a triad of gods that rule over sky (the thunderer), land/sea, and underworld, much like Zeus, Poseidon, and Hades. (Poseidon was not only god of sea but also of earthquakes, so he was also associated with land; presumably the maritime Greeks cared more about his rule over the sea.) These three gods are children of the highest Sky Father (the Sanskit Dyaus Pitar, Jupiter, Kronos, Odin All-father), who is associated with regality, laws, and oathkeeping. Note that in the oldest Roman religion, Jupiter was *not* god of thunder, and was not the equivalent of Zeus; however, the Thunderer Zeus absorbed the functions of the Sky Father as well (as reflected in the myth of Zeus overthrowing Kronos and ruling over his brothers), including his name (Dyaus = Ju- = Zeus), and later authors linked him to Jupiter.
Zeus has the thunderbolt, which is very recognisable as coming from Indra's vajra, or at least in the same family. All the three main Olympians - Zeus, Hades, and Poseidon - have an attribute of the (one pointed) thunderbolt, Hade's bident, and Poseidon's three-pronged trident.
5. It seems appropriate for a young man to go on a journey, explore and learn new stuff before settling down and starting a family? - But this is likely present in other cultures as well.
6. You can refer to alcoholic beverages as "coming from the gods" or otherwise being associated with a god or goods? But this also seems pretty common in plenty of cultures.
I don't think this is going to be a fruitful search. Too hard to distinguish things which are actual cultural holdovers from things which developed independently in many places because they just make sense given human psychology/technology/etc.
Not really? Plenty of cultural traits are arbitrary, and shouldn't develop independently in many places. E.g. I know of no independent mace-wielding thunder-gods.
Plenty of them are arbitrary, and those ones won't repeat. But I would argue many cultural practices are not arbitrary, and those ones will repeat and will foul up your ability to determine which ones are actual shared heritage, and which ones are convergent cultural evolution.
If you know of any trait that could be shared heritage or could be convergent evolution which matches my criteria, feel free to share. You have seen my attempts, you see that I haven't found a good one. So I think your worry is overblown: the problem is not "how do we separate the X things from the Y things?", it's more "Are there any things at all that could plausibly be X or Y?".
Well sure "Respect you elders". "Don't go out at night". "Strangers are dangerous". "Children are irresponsible and need minding". "The young leader was disinherited and then returned and took rightful power". "Big animals are dangerous". "Water/Rain brings life". "Bad/untrustworthy people are left handed".
There are a lot of common features, and I suspect if wood artifacts survived with any regularity we would find many more. People aren't very different overall and find common solutions to common problems. Some of those solutions are cultural.
No. Those seem to exist in e.g. Native American, Chinese and Pygmy culture and thus they don't fulfill my criteria. I want to find something that's shared in the Indo-European cluster and doesn't seem to exist outside of it. I.e. mace-wielding thunder-gods, but in other realms than mythology.
I don't think this will work, the Iranian guy in particular is injecting tons of noise into the exercise because of the 1400-year-long contact with Islam, the Indian Aryans were locked behind the Himalayas for millennia and are much closer to Central Asians, other "natives" of the Indian subcontinent and China than to their ancestors.
I think that "Indo-European" is about as useful a category to try to map back into modern categories as "African-origin humans". It's so diffuse that, even if they do share some unique thing, that thing is now spread all over the gene pool and the meme pool with no hope of ever recovering the fact that it once belonged to Indo-Europeans alone beyond reasonable doubt. Exceptions might be things like "Blonde Hair" and "Blue Eye", and I'm not even sure they are exceptions. (Africa and most of Asia is definitely out, but South America is full of blonde-haired blue-eyed people with plenty of non-european genes ?, and Egypt and the Levant are as well to a much lesser degree)
The Moscow Mayor and the White Soccer Mom from America share much more than the rest though : Both Christian, both white-colored to the same rough degree, their grandparents probably fought with or against each other in a war in Europe somewhere, their 3-centuries+ ancestors probably traded/intermarried/emigrated together in Europe, things like that.
"South America is full of blonde-haired blue-eyed people with plenty of non-european genes ?"
Are you serious? Those are 1000% descendants of European immigrants, and the fact that there are still sizable populations of blue-eyed blondes in e.g. Argentina shows you the extent of de-facto racial segregation that goes on down there.
Oh, I know of course that all blonde hair and blue eyes in South America is due to european genes, I just don't see South Americans often lumped with "Indo-Europeans", possibly because of the significant native-black intermixing. So my point was simply "Even for something as undeniably Indo-European such as blue eyes and blonde, South Americans also have it but are not usually thought of as 'Indo-European' due to massive culutral mixing"
I don't think it's this hopeless. The Iranians and Indians do share linguistic traits and mythology with the rest of the Indo-Europeans even though the factors you state apply. It doesn't seem impossible that there are other traits that are also shared.
Apart from aspects of language itself, could there be a single cultural trait which all Indo-European groups have but which Hungarians lack? I put the chances of such a trait existing at less than 10%.
The Database of Places, Language, Culture, and Environment is the best ethnographic resource I know of in which to search for possibilities: https://d-place.org/parameters
Unfortunately, there are 2,368 different variables. I wish someone would go through them all and pick out a top 100 most informative/interesting. (Or devise some other way to make it more digestible.)
I have a question for those who want US and EU (“the West”) to broker a peace deal between Ukraine and Russia that would entail a loss of Ukrainian territory or/and some other concession which Ukraine currently rejects. How, practically, is this supposed to work if Ukraine isn’t willing to play ball?
Only solution that I can think of is that the West would simply threaten Ukrainians with withholding critical aid unless they moderate their demands. Is this the plan? Or is there some other way I am not seeing?
That's a stick, I guess. Could also use carrots? Money for rebuilding seems obvious.
What I've seen from Ukrainians is that they are deathly afraid of making a peace deal now and signing a piece of paper, Russia rebuilding their military for as long as they need, and then one day Russian tanks are again speeding towards the capital city. The only thing that could alleviate this fear are hard security guarantees from the EU or the US - something like foreign troops stationed in Ukraine as a trigger (like South Korea) rather than being a piece of paper.
NATO already pledged to support Ukraine for "as long as it takes", so unless you redefine what that implies to get out on a technicality, or take a major hit to your credibility, that lever seems unavailable.
Yes but that was only the informal wording used in press reporting. Looking at the official NATO website here (topic 4, What are NATO and Allies doing to help Ukraine?):
It seems clear that "it" means preserving Ukraine's right to exist, war crime prosecution, and post-war reconstruction. This can only be understood as "until and after Russia has been fended off", because otherwise NATO couldn't do any of these things.
Also see the statement of NATO foreign ministers of 29-30 November 22,
> We remain steadfast in our commitment to Ukraine’s independence, sovereignty, and territorial integrity. We will never recognise Russia’s illegal annexations, which blatantly violate the UN Charter. We will continue and further step up political and practical support to Ukraine as it continues to defend its sovereignty and territorial integrity and our shared values against Russian aggression, and will maintain our support for as long as necessary.
Point taken, but all of this still seems pretty weaselly, at least to me. Like, "We will never recognise Russia’s illegal annexations", but what if Ukrainians hypothetically agree to legalize them under military pressure from Russia? Nowhere does it say "we will give Ukraine enough aid so it will never be in a position when surrendering territory would seem like a lesser evil to Ukrainians".
Moreover it should be noted that US does not have a much to fear from a hit to their credibility in the EU, and neither US or EU have much to fear from a hit to their credibility among Ukrainians.
In sum, while I do not personally support using this lever, I think it is very much available, even fairly likely to be used.
I disagree about your last paragraph. A real loss of credibility in th EU (something deep enough to threaten the integrity of NATO) would really put the US on the back-foot vs China or any great power aiming at challenging the US domination.
Frankly, setting aside whether US anti-China policy makes sense (I am very unsure about that), in this area EU is not currently doing much more than it has to in order to not wreck their (well, ours) relationship with the US, our main protector against Russia and any other threat. And EU is unlikely to ally with China no matter how bad our relationship with US gets.
Arguably, loss of US credibility in the EU might even increase European cooperation with US anti-China measures, since if Europeans would be more afraid that US is going to abandon us, we would try harder to be useful to them. Like Australia.
>Nowhere does it say "we will give Ukraine enough aid so it will never be in a position when surrendering territory would seem like a lesser evil to Ukrainians".
Considering the strongly worded nature of the declaration and general public opinion, dragging their heels significantly enough to affect the outcome of the war would likely be considered a breach of the promise.
>Moreover it should be noted that US does not have a much to fear from a hit to their credibility in the EU, and neither US or EU have much to fear from a hit to their credibility among Ukrainians
While the EU is certainly involved and has its own stance on the war, we are talking about a NATO statement. Very different matter. The mutual promises of the EU are much less rigid and much more negotiable compared to the core NATO promise.
>In sum, while I do not personally support using this lever, I think it is very much available, even fairly likely to be used.
After their declaration, the lever is absolutely not available any more to NATO. The whole idea of NATO hinges on the mutual, 100% certain assistance in (defensive) war. If NATO makes that promise, even to non-member Ukraine, as strongly worded as this and unilaterally walks back on it, it would have severe negative consequences for the whole treaty. If that promise is any less than 100% credible, it would invite outside aggressors which NATO is supposed to deter to test just how much below 100% the promise is now really worth, which defeats the whole point.
>it would invite outside aggressors which NATO is supposed to deter to test just how much below 100% the promise is now really worth, which defeats the whole point.
Like whom? Do you think that once Mexicans find out that NATO is not that credible and gringos thus might not get the help from the mighty Bundeswehr, they will try to reconquer what they had lost in the last US-Mexican war?
I jest, but only slightly. Obviously, e.g. Estonia is quite dependent on NATO credibility, but Estonian military industrial complex would not be able to keep Ukraine afloat. Ukrainians need the help of mighty industrial nations of the West, who practically by definition have much less to lose from the loss of NATO credibility; although Europeans do have more to lose than Americans.
I can't see NATO saying that they will stop supplying Ukraine if it doesn't sign such a deal. But I can see NATO saying, when a deal is on the table, that if Ukraine agrees to it and Russia doesn't NATO will give even more support to Ukraine than it has been giving. That would be an incentive for Ukraine to agree.
Ukrainians get sick of dying and getting their country blown up? Even if they might eventually "win" with indefinite NATO support so they reduce demands.
>Ukrainians get sick of dying and getting their country blown up?
This has never happened in world history and you would have to assign that outcome a miniscule probability. The land war is the only one that matters, and as long as the Ukrainian people believe their leaders and soldiers can throw the Russians out, they will endure.
So like what was going on in the 13th and 14th century then? Or 1667? Or say 17th and 18th century with the Russia progressively stomping out and co-opting the Cossacks/Hetman as an independent entity/force?
Then in WWII a bunch of the Ukrainian people threw in with German invaders (some for fairly good reasons), and then these were in turn pushed out by the Red Army as it eventually won.
How does that all square with your flat statement of invincibility?
I believe the commenter before me was referring to the strategic bombing being carried out by Russia against Ukrainian civilian infrastructure ("Ukrainians get sick of dying and getting their country blown up").
What I declared to be without historical precedent was such strategic bombing winning a war on its own by defeating the morale of the civilian population, as has been proposed since WW1.
Really? Are Ukranians willing to suffer and die for Crimea? Because the power hungry fool Zelensky says the war isn't over until Ukraine controls Crimea.
In case you aren't trolling, Crimea is part of Ukraine and its annexation by Russia has not been recognized internationally save for a handful of small countries, including flavourful names like Afghanistan, North Korea, and Syria.
And evidently, yes, the Ukrainian people are willing to fight and die for it, or we wouldn't be having a 10 month war. Ukraine is fighting an existential war against a genocidal regime, everyone involved knows it, and that is entirely sufficient to explain the cohesion and morale of the Ukrainian people. Ukrainians are not puppets, they are people with agency, and to imply that Zelensky, or presumably his Western handlers, are somehow controlling them against their will is at best unnecessary, at worst enemy propaganda. Yes you can force people to fight, see Russia, but the results speak for themselves.
I don't think we know whether, if the war was not going well for Ukraine and Russia withdrawing from everything except Crimea was an option, Ukraine would accept it. They say they won't, but that doesn't tell us what they would accept if they couldn't get anything better.
Indeed, until they actually invade Crimea, I think that must be considered potentially part of a deal. I think there are internal politics that make it difficult for them to say so. But in the last eight years, it was the Russia-backed insurrectionists in the Donbas that they challenged.
>Only solution that I can think of is that the West would simply threaten Ukrainians with withholding critical aid unless they moderate their demands. Is this the plan? Or is there some other way I am not seeing?
All of the western weapons and aid are what is holding the Russians back, the Ukrainians are playing with the west's ball. If the west takes its ball and goes home what would happen to the Ukrainians? I'm guessing Ukraine would eventually get steamrolled. And what would happen to the west? They'd avoid taking the huge hit from the gas bill, spend less on weapons, maybe take a bit of a credibility hit, lose a future ally vs Russia. Who holds all of the leverage in this situation?
I don’t think Ukraine is that inconsequential for “the West” as you imagine it is. Depending on the country for sure, France maybe doesn’t care that much, the US is now seemingly always one election away from withdrawing from NATO, but the Eastern Europe will absolutely lose their shit if Ukraine were to lose because the West “is tired”. Spending less on weapons is 100% not happening, Poland is already militarizing up to the gills and arguably at the safest time when Russians demolished their professional military. There would be huge humanitarian crises, geopolitical and military repercussions up to nuclear weapon proliferation, and then green light to China with regards to Taiwan. Also I suspect the democracy in the West would suffer as well, if only because Russians would be allowed to keep buying up Western elites which is currently somewhat on hold due to this war.
I think that its largely inconsequential for the decision maker in OP's question, which is the US. Even if Eastern Europe does lose its shit what are they going to do, buy more F35s? I don't see that as deterring the US to force a settlement. And I don't see nuclear proliferation in NATO countries either, I don't think the cost benefit analysis favors a country like Poland pursuing nukes. Either way it is definitely inconsequential when compared to what Ukraine might face if the aid was cut off. Which gives the US the leverage.
> There would be huge humanitarian crises, geopolitical and military repercussions up to nuclear weapon proliferation, and then green light to China with regards to Taiwan.
What geopolitical and military repercussions are you expecting to happen in the event of a US forced settlement?
And I don't see the US forcing a settlement in Ukraine as a green light to China on Taiwan either. One because I think that the US should force a settlement in Ukraine to focus its energy on China in the first place. Two because invading Ukraine has been extremely costly for Russia, who didn't have to do it via amphibious assault. I would imagine that a Chinese blockade would be similarly costly as the Ukrainians have had a lot of success against the Russian navy. And three because the west came together and basically cut off all exports from Russia. I think China can reasonably expect the same response from the West if they were to invade Taiwan, which would absolutely tank their economy.
>Also I suspect the democracy in the West would suffer as well, if only because Russians would be allowed to keep buying up Western elites which is currently somewhat on hold due to this war.
Maybe that is more of a EU / Germany thing but I don't see much evidence of that happening in the United States to begin with.
> I don't think the cost benefit analysis favors a country like Poland pursuing nukes.
I think the opposite, NATO is no longer an unconditional alliance, American public is at least okay with questioning it or okay with voting for anti-NATO politicians for some other reasons, which is their right sure but that has security implications for Poland which ultimately can be resolved only with nukes.
Same for geopolitical repercussions really, eventually I’d expect new alliances formed in Europe.
> Green light to China with regards to Taiwan
Putin launched his invasion in part because he thinks the West is weak. The West “sacrificing” Ukraine’s territory is going to be seen as confirmation of that fact. True, Russia pays dearly cost for the war but it does not jeopardize Putin’s grip on power (seemingly?) which is the lesson China can learn. Just as they learned that famines are not a problem for the communist power. Also China is way more entangled with the West economically for the economic warfare to have a deterring effect. Even with Russia, it’s not clear what the effect is. The Economist publishes articles with the opposite conclusions: sanctions don’t work/sanctions work better than you think, all the while the Europeans are complaining about the gas bills. The leverage of the West over China is questionable.
>I think the opposite, NATO is no longer an unconditional alliance, American public is at least okay with questioning it or okay with voting for anti-NATO politicians for some other reasons, which is their right sure but that has security implications for Poland which ultimately can be resolved only with nukes.
Polling I have seen for the US generally shows support for NATO. Ukraine is not and has never been a part of NATO. NATO is giving Ukraine free money and are under no treaty obligation to do so. I don't see how adding conditions to free money to someone you were not obligated to help would somehow undermine Article 5. Unless you care to explain.
As to Poland I think you are greatly underestimating the pushback they would get if they tried to get nukes. Google has told me that Poland is planning on opening its first nuclear power plant in 2033, I seriously doubt that they would be able to withstand 10+ years of the Iran or Iran lite treatment. Especially if kicked out of NATO / EU.
>Putin launched his invasion in part because he thinks the West is weak. The West “sacrificing” Ukraine’s territory is going to be seen as confirmation of that fact. True, Russia pays dearly cost for the war but it does not jeopardize Putin’s grip on power (seemingly?) which is the lesson China can learn.
No idea why you think forcing a negotiated settlement makes the West look weak. Nothing about this war makes me think the west is weak. The west comes out looking great - all of our shiny toys work great, we all came together on some sanctions (effective or not), the Russians look like a wet towel.
As far as the "sacrifice" framing - again the west and NATO and the US in particular doesn't have any defense treaties with Ukraine. We are giving them aid because we think its in our best interest and because we are being nice. I don't think its reasonable to call attaching strings to the free money we are giving as sacrificing Ukraine, even though yeah it does suck for Ukraine.
Vis a vis China - I think the Chinese (and everyone else) will also see how vulnerable their navy is to missiles and drones. I think that the Chinese will expect at least a similar level of support to Taiwan from the US (IMO should be more as ideally the US is forcing the settlement to focus on China). I think that the Chinese will see the ferocious response from Ukrainians and think twice about their chances in Taiwan. Overall I don't think Ukraine really has any strategic value to the US, at least when compared to Taiwan. So we should force and end and move our focus to China / Taiwan.
I tend to agree with those who think that Ukrainian defeat in this war makes nuclear armed Poland much more likely. It is, however, unclear whether that would be something Washington would be worried about.
I would be fairly confident that an Ukrainian defeat would have triggered some serious confidential talks with the French about the use of their force de frappe, and the ASMP specifically.
Ukraine applied for NATO membership on 22 September this year. There is now a process involving an invite to join, a protocol to be drawn up and accepted and so on, however at some point, NATO will include Ukraine - almost certainly to the contentious internationally-accepted border that includes Donbass and the Crimea. Does this compel Russia to make peace sooner - or is this the very scenario that Putin objected to last November, the "NATOisation" of Ukraine? Is there an inevitable slide to a NATO-occupied Ukraine that only a Russian victory can hold off? A reasonable compromise in these circumstances may be Russian retention of Crimea and Donbass in exchange for NATO occupation of Ukraine.
"NATO-occupied". How exactly is the country occupied if it's a voluntary member which applied to join? And if it specifically applied to join due to Russia's unjustifiable war of invasion?
That's a game that Russia apologists play all the time.
When Russia does something horrible, they will loudly object against using any word with negative connotations. For example, killing entire village cannot be referred to as "massacre" or "murder" or even "killing" or "indiscriminate killing"... you are supposed to use the proper term "zachistka" (sweep). Because if you don't use the proper term, it shows that you are "nyekulturny"; an illiterate simpleton unable to discuss the complex realities of life. You need to educate yourself first.
And trust me, Russians have more words for hurting people than Inuits have for snow.
On the other hand, anything done by opponents of Russia must be described in the most hysterical terms possible. Thus, for example, Russian army coming to your country, killing everyone who seems like they might resist, replacing your government, and staying there indefinitely, controlling your politics and draining your economy, is called "friendly help"... but if your country applies for NATO membership, and afterwards does a joint military training once in a while, that is "occupation". Opposing Russia literally makes you a "Nazi": https://www.youtube.com/watch?v=BRFcK1BDlus
There is no such thing as voluntarily doing something Russia does not approve of. That makes you simultaneously a victim that must be rescued at any cost, and a villain that must be punished, preferably by death. The idea that someone does not want to become a part of Russia is just silly. And even if they do not, who cares, and why should that be relevant in any sense?
But the silliest of all is the idea that someone could become a "Rusophobe" as a consequence of what Russians do.
You mean like how NATO calls people terrorists, but will freak out if you say anything negative about their psychotic military operations in the middle east and north africa?
Oh? Really? What has anyone offered them? Zelesnky says the war isn't over until Ukraine controls Crimea, meaning Russia is being offered literally *less than nothing*.
a) Is it "the plan" of our politicians? - I doubt it. It is mainly Kremlin-Trolls in the west. Who have been esp. active recently - on FP, Newsweek et al. . . See for competent analyses: https://slantchev.wordpress.com/ (highly recommended!)
b) Ukraine is doing demands. Fine. Russia is doing different demands. Not so fine ones. - IF both sides believe they may be able to work something out to bridge the differense: They better do so.
c) Right now, it does clearly NOT look that way.
As Putin insists on keeping all occupied territory - and then some. And would break any ceasefire the moment it pleases him. (While not winning!) - OTOH If he would offer going back to 22.2.22 then there might be room for talks: 1. Ceasefire to allow quick Russian retreat. 2. Talks about the future may start. Might work out, or not. "The west" would THEN probably offer the Ukrainians many reasons to go for peace (even without Crimea, yes). Carrots mainly. Big ones. The "stick" being: Maybe less carrots.
Ad a), to be clear, I don’t think it is the current plan of Western politicians. I am also against it. On the other hand, I don’t think it is fair to label all its supporters as Kremlin-Trolls. This sort of debasement of the discourse is unhelpful.
What is the morally right thing to do with a person who is in a permanent vegetative state? I have a close family member who has no chance of recovery and seems cruel to keep them in this state. thoughts?
Discuss DNR options as that person would have wished with other family members. Ensure the doctors know as well.
Visit them often, ensure they are being properly taken care of, and pay the care takers appropriately. Spend time with the ill person and ensure you remember their humanity, so that the staff does as well.
Remember that this is a human person and not your property to dispose of when inconvenient or distressing. Be realistic but do not fail of hope.
It is harder on you than on the ill person. Be kind to yourself as well.
By all accounts people in such states, even if they have some brain activity, do not seem to be especially unhappy. Perhaps they are dreaming, or in some state like it.
I mean if it was a close loved one like a wife or child and there was no chance of recovery I would just kill them myself regardless of whether Euthanasia was illegal or not. Society isn't determining my family's end of life choices (though we do have the right legal docs in place in our case).
We are just space dust and literally everyone dies. Keeping the husk alive at a huge cost of resources is pointless.
If it was someone else's close family then they should decide (asusming they can pay for the care).
I believe the moral imperative in these situations is to minimize human suffering; both that of the patient's (potential) ongoing suffering, and that of their survivors, because quality of life matters *infinitely* more than mere continuation of life.
So if there is little to no chance of the patient experiencing a basically good quality of life in the future, the morally correct thing to do is end the suffering of all concerned as quickly as possible. I would advocate for euthanasia if it were an option, but if not, withdrawing care is probably going to be the only course available to you.
It's the course I've directed for myself in my living will, and that literally every single person who is close to me has directed for theirs. The only people who disagree with the idea of quality of life mattering more than mere life tend to be religious people who derive their morality from irrational beliefs. Don't allow them to be your guide in this.
Out of curiosity, at what point in the spectrum of 'quality of life' do you regard that quality as insufficient to justify not killing the person, or yourself?
And as a follow-up, do you have exact same break point as all the people who also see end of life options in this way?
I think we should treat people with as much compassion and dignity as we treat pets, which is to say when someone can no longer eat, drink, and relieve themselves, and they are unresponsive or in too much pain to communicate and/or rest comfortably, and there is no reason to think they will recover enough to ever do those things again, they should be helped to die as quickly and painlessly as possible.
Of course not everyone shares my opinion on this (which you know, so no need to be snide), but there is a reason advanced directives and living wills were invented and every healthcare provider now hassles patients about getting one on record. The healthcare default is to preserve life regardless of suffering or literal cost and enough people think that's such a terrible idea that we invented a whole legal process to override it.
Well better than I've seen some pets treated. But otherwise I totally agree with you. My mum died from Alzheimer's, and my only plan is to die before then, or somehow off myself if diagnosed. Oh which brings up the question of whether I should tell my loved ones before offing myself, or after... hmm I guess, before is the obvious answer.
As far as the OP and the poor person in a coma, my rationalist answer is to find the time for other people to come out of a coma from the same cause. And when the loved one is past the 90% mark (or 95?) then give it up.
I recently caught a This American Life episode about an American with an Alzheimer's diagnosis going to Dignitas in Switzerland to die. The service and process was both practical and beautiful.
A virtual certainty that she and and others who share her view will not have the exact same break point. Likewise, people who think the income limit of about $125K that qualifies people for partial student loan cancellations is too high will not all have the exact same number in mind for what a reasonable limit would be. People who are in substantial agreement about something are not going to be in perfect agreement about every detail. Everybody knows that. So what’s the point of demonstrating that Christina does not have the exact same standards for what counts as a live worth living as others who favor euthanasia?
I personally think it is right to euthanize these living remains of a no-longer-conscious being. However, I do not want to give an extended defense of that. But here are a few practical suggestions:
Read up on these states. There are different levels and degrees w of neurovegetative states. Articles are easy to find via Google. Also research predictors of recovery, and tests that give you an idea of whether the person is at least partially conscious. Here’s one good one: https://neurosciencenews.com/consciousness-sniff-test-16445/
Once you know about this stuff and perform these tests, you may be clearer about whether the person’s mind is permanently and totally gone. If it is, and you believe you should euthanize them, you should know that in some countries this is done in an indirect way that gets around the law against euthanasia. Nourishment and water, which are being supplied by gastric tube or IV, are withdrawn, with the rationale that they are treatments from which the patient is not benefitting. Perhaps that is legal in your country. Another option would be to ask to have the relative transferred to the care of their family. The family could then withdraw nourishment and water, or possibly find some less protracted way to end the person’s
life. But of course everybody involved would have to be in agreement with this plan.
My son (7) has not really caught the bug for reading yet, but I think I’ve found his subject. I was telling him some Greek myths the other night and he was fascinated. He wants books on them for Christmas.
Anyone have any recommendations for books either about Greek myths or based on them? He reads well so they don’t need to be geared towards first graders, but they probably should have pictures. Thanks!
"Heroes, Gods and Monsters of the Greek Myths" by Evslin, Evslin, and Hoopes. I absolutely loved these when I was a child. Something about the language made them a joy to read.
The Percy Jackson series is pretty great. Aimed at slightly older kids, but likely to appeal to younger ones as well. The author, Rick Riordan, has also written series involving Egyptian and Norse gods.
I'm interested in whether people think AI is going to render a lot of professions obsolete in the next few years.
I was quite shocked to see GPT3 programing, admittedly not to a professional human standard, but it was basically competent. I could easily imagine the next generation of AIs surpassing my ability at least, and obviously it would be exponentially cheaper than hiring human coders. I can't see how I'd stay employed in that scenario.
It seems like most jobs that mainly involve manipulating information are in the same situation. Accountants, doctors, journalists, lawyers, finance, engineering, basically the entire professional middle class, might end up going the way of the hand-loom-weavers.
Maybe I'm overreacting, there's a lot of (what sounds like) hyperbole around AI at the moment, but could we genuinely be looking at an early-stages-of-the-industrial-revolution type disruption of the labour market over the next decade?
A year ago I would have laughed at the idea, but does a near-future where AI does all the cognitive work and human workers all do manual labour (or are celebrities etc.) seem plausible?
It's not just a Markov model--the whole interesting thing about it is that it keeps some kind on context beyond the last N words output, which is what allows it to be a good chatbot, and to stay on topic for multiple sentences. Markov models do not work that way--they produce sentences that sound like English but don't make any sense and can't maintain anything like a point.
Is there a reason to think that the limitations of chat GPT's quality of writing are inherent in the technology, rather than just "computers may beat humans at chess, but they'll never conquer go...oops, wait?"
I'm wondering if something like GPT-3 will become the mouthpiece for some more general AI, just as various classifiers will become the image processing/voice recognition part of the general AI. The AI knows it wants to get a particular human to wire $1.3 million dollars from account X to account Y, and so tasks GPT-3's descendant with the job of writing an email that asks the human to do that in a plausible way, provides detailed and easy-to-follow instructions for doing so, and maintain's this AI's friendly persona in the tone of the email.
I mean, there's *something* like a language module in humans, because damage to particular parts of the brain can scramble our language abilities. (Various flavors of word salad result from different brain injuries, as well as just flat losing spoken language.) And I think something like a module for face recognition, image recognition, etc. must exist. So my assumption is that all these are modules that will eventually plug into an AI, but also into any number of other automated systems.
As an example of this now, if you ever call a credit card's 800 number to get your balance or question a charge, you get a voice recognition system. That's an AI algorithm (some kind of neural net classifier, I think) that's being used as a module by the voicemail system, which is just as dumb as it ever was. You just get to say the credit card number or selection in a menu instead of hitting a button.
If you replace "few" with "many" in the sentence, I would agree. What I think is far more likely for a long time is that AI becomes a supporting tool, and that some professions will get a lot of "AI wrangling" as part of them.
Examples: An artist can use AI to do boring work with filling details, drawing repetitive items, or even creating a first image that the artist can then bring into publishable quality. The programmer can use AI to quickly create a function, even when the AI can't be trusted to make a complete program. The lawyer can use an AI to auto-create the boring routine stuff. If a truck AI can't handle chaotic city traffic and all the non-driving stuff a trucker does, maybe it can still drive a highway autonomously while the driver sleeps?
That's pretty much my thinking as well. I'm reminded of how long it took to get good compilers for third-generation languages. Back in the sixties, seventies, and even into the eighties, a good programmer really was supposed to understand the quirks of compilers and write around them, or even just skip that and drop down to writing ASM in critical sections.
Let's also remember that a good part of a programmer's work isn't just writing code to do something, but rather negotiating with humans what the system should do in the first place and determining whether the system can be made to do what it is being asked to do either at all or with a reasonable amount of effort. I expect the same can be said of any number of white-collar professions.
"Back in the sixties, seventies, and even into the eighties, a good programmer really was supposed to understand the quirks of compilers and write around them, or even just skip that and drop down to writing ASM in critical sections."
I don't think this expectation has gone away because the compilers are better so much as because compute has become unbelievably much cheaper. Assembly language is still vastly more efficient than most other code, people just don't care about efficiency anymore.
Part of the it is that, yes, but we've definitely gotten to the point where good compilers really are better at optimization than all but the best programmers in most circumstances. Also, many of the things that made compute cheaper also made architecture much more complex, so it's actually harder to manually optimize than it used to be.
Assembly language isn't really directly telling the machine what to do anymore--instead it's more like a language that's being interpreted/JIT compiled by the hardware executing it. (It's not at all clear to me why it makes sense to do the optimization at the hardware level rather than a level or two earlier in the compiler or maybe in some JIT compiler thing in software, but maybe that ship has sailed.)
Itanium (aka Itanic), or VLIW more generally, really was that ship and it really has sailed. It seems that in the end there are too many unpredictable variables wrt recent context switches, caches, and so on for optimization of code not to include a dynamic component, plus VLIW bloats code size making instruction caches less effective. As for doing it in software, the microcode is software and is used for the parts of the optimization that can be done fast enough in software.
One reason for doing out of order execution is that the amount of time required for a load operation varies greatly depending on where the value you are loading resides. On a modern machine, a load operation might look in a level 1 cache for the value being loaded, then look in a level 2 cache, then look in a level 3 cache, and if the value isn't found in any of the caches, fetch it from main memory. Fetching a value from main memory takes a *lot* longer that fetching it from the level 1 cache.
A compiler can do a good job of instruction scheduling if it knows how long each instruction will take to execute, but figuring out how long a load instruction will take to execute is hard if not impossible. (On a machine with multiple cores, the execution of a load instruction can depend on what is happening on other cores as well as what is happening on the core executing the load instruction.) On the other hand, an out of order execution engine can issue load a bunch of load instructions and then start executing instructions using the result of a load operation as soon as that load operation completes. What this means is that an out of order machine can do a good job of scheduling instructions even if it has no idea of how long a particular load instruction will take.
TransMeta designed two x86 compatible processors using Just In Time code generation, which demonstrates that the approach is feasible. However, their first model was quite slow, their second model was slower than Intel and AMD's fastest chips, and the company concluded that developing a third processor would not be profitable. Since then, out of order has gotten more attractive relative to static scheduling because transistors have gotten cheaper, so I doubt that in order execution makes sense for a modern processor unless you are strongly optimizing for low transistor count over performance. Intel's Atom line has used out of order execution since 2013. ARM uses in order for its lowest end processor (the Cortex-A510), but everything above that is out of order.
Yes this. No AI is replacing an "accountant", not any time soon. But an AI might make an accountant's job easier, and thus you need fewer of them. What an accountant will eventually become is "the person who talks to the accounting AI and knows what it is saying and how to get it to count what we want counted".
This offers a narrative for how AIs might become very powerful and largely take over the society, even if they're not skynet or paperclip maximizers. Over time, more and more important decisions in our society are taken over by AIs, with humans managing them, or overseeing them, or sometimes simply collecting a rent on being the guy who gets paid to ask the AI the question and then pass along the answer. Nobody is exactly enslaved by an AI, but you can imagine a pretty smooth transition from our world to one in which more and more of the created/designed/built world and more and more of the culture are produced by AIs, more and more of the practical decisions (Should we build this bridge? Should we open a new factory? Should we impose sanctions on that country?) end up being made mostly by AIs, with some human supervision and final oversight.
As the AIs become more and more capable, the people who basically go along with the AIs in nearly all cases will end up winning all the competitions. It'll be like chess playing, where (as I understand it) human+AI was better than AI for awhile, but at this point, even Magnus Carlson should probably just do what the AI tells him.
If the AIs continue with their tendency to make plausible-sounding stuff that is sometimes just wrong, then the human can do quality control on the AI's output. But it seems likely to me (as an outsider to the field with no particular insight) that this problem will be solved and AIs will end up less likely to produce incorrect code/text than humans.
For code I can see it (as it's a formal system that can be easily checked against), but I'm less convinced for free text. The GPT solutions tend to respond with some kind of average of their inputs, which means that it will reliably reproduce common misperceptions, for instance.
Both for art and text, I believe we will need some kind of "reality model" before this can be solved. AI art would certainly benefit from having a definition of human anatomy.
AIs (from the current training method) also tend to generate boring well understood answers to already-solved questions. Even the stuff that we don't want the AI to tell people (like how to make meth or hotwire a car) are clearly solved - there's no new chemistry or mechanics involved.
That's great for some fields where we just want the same things processed over and over again. It's terrible in any job that requires creative answers or responses to novel inputs.
I’d like to think we have a few decades left before we’re all out of jobs but considering humans will be competing for the same limited resources on earth , by the time A.I has taken over all cognitive work humans will be considered pests by our super intelligent overlords
Will never happen. The job takeover by new technologies has always been, and will always be, succeeded by the creation of many new, previously unfathomable jobs. The limitlessness of human ingenuity renders unimportant the limit of resources on Earth.
Every generation since the beginning of the industrial revolution has had some technology or other that was totally going to create permanent mass unemployment. And every time, the people who suggested otherwise were dismissed as naive Pollyannas. I'll stick to the side with the multi-century history of vindication, thank you.
My main rebuttal to that: until now automation automates specific tasks. This destroys some jobs, but creates new jobs, and the increased productivity of the more automated economy makes those jobs more productive and better paid. But AI automation does not automates specific tasks: it can automate whole classes of tasks, potentially both the task that you wanted to automate in the first place, and all the ancillary tasks that the automation generated. So no new jobs are created, or in any case far fewer than it was the case in the past.
The standard counterargument here is that we're destroying jobs for low-IQ people and only creating jobs for high-IQ people.
We've got away with this so far because of where we started. In the stone age, 99% of jobs could be done adequately with a two-digit IQ. By the 1900s maybe we're down to 80% of jobs requiring only a two-digit IQ. But we are now reaching the point where ~50% of jobs require a three-digit IQ and still only 50% of people have a three-digit IQ. What happens when 80% of jobs need a three-digit IQ and there's nothing useful for the lower orders to do?
Some people argue that AI will be at least as good at taking the middle or high-end jobs going forward. It’s probably easier to make lawyers 10X more efficient than construction workers or home health aides. I think it’s a fairly convincing point.
I agree we will find new jobs for those displaced by technology, the comment was that we should worry that a super intelligent system could see humans as a threat to survival.
This only works as long as there are still some jobs that humans can do better than machines. It won't hold if AIs and robots become better than humans at every task.
Depends. Consider comparative advantage. The CEO might do a better job cleaning the office than the janitor, but it still makes sense to have a janitor.
I think we tend to assume that AI and robots will have infinite capacity, but I’m not sure that will be true any time soon.
Agreed, at least in theory. I had almost written an aside to that effect, but I removed it to avoid confusion.
In practice, I suspect once we're able to make (for example) a good AI radiologist, it'll be very cheap to spin up more instances or increase the capacity until all x-rays can be analyzed by AI.
Probably less true for robots, but even then I'd expect mass production and technological improvements to bring the prices down. It'll take longer to scale up production, but once it does, it might only be worth it to pay humans the equivalent of $0.25/hour today (made up number) versus buying a robot janitor.
Things are true until they aren't. I don't think it's appopriate, even if this has been true for automation thus far, to have ¯\_(ツ)_/¯ as an insurance policy.
I sympathize with that declaration. But what I'm saying isn't throwing up my arms with indifference. Indifference is the opposite of what I feel; I feel strongly that no insurance policy is needed. In my mind, the only contingency surrounding the entire premise is that people must be left socially and economically free to pursue their rational interests, and in that case, there is no need to fret about AI, and certainly no need to fret limited resources or jobs. I'd wager on one of these prediction markets that if AI ever became "overlords" to man, it would answer to its own overlord who would be, unironically, a man.
I understand that you have strong feelings, but if your feelings turned out to be wrong, I could not live off your salty tears alone.
One of the things you seem to feel strongly about is that AIs could never be actually more capable than "men", as you put it. Why do you feel this way?
Lol! Maybe you could live off my tears. AI will absolutely be more capable than men at lots of things. It already is, isn’t it? It takes more than competence at tasks to become an overlord over mankind.
If the stock market did view AI as important, how would stock prices differ from their current state? Of the top five US companies by market cap, three of them have large AI efforts or would clearly benefit from a large AI sector.
This is a misunderstanding. I did not say that stock market sees AI as unimportant. But the OP’s question is whether it is plausible that AI replaces almost all cognitive labor in near the future.
So, how would the markets look if they would see that scenario as plausible? For one thing, imho there should be a lot more venture investment into AI. But, and this is more verifiable, stocks of tech companies should move on information about AI progress, instead of traditional indicators like sales etc. And we are not really seeing that. E.g. it was recently announced that Meta developed an AI sort of able to defeat humans at a board game called Diplomacy (with heavy caveats); this seems like an important breakthrough in AI development, did it help Meta stock? Admittedly I didn’t bother to look that up, but I am guessing no, it is still in the doldrums.
And if the market believed this story, Fed raising interest rates should not have disproportionately bad effects on tech stocks, which is what we are seeing. Now, this is a subtle point since higher interest rates are going to slow investment, including investment in AI, but the fact that it affected tech companies more than e.g. oil companies is an indicator that markets still see tech investment as more speculative and uncertain than other kinds of investment. Which imho does not square with the plausibility of imminent AI takeover of, well, everything – even if AI would grow strongly but not so strongly as OP suggest, tech companies still should benefit enormously.
Now, markets are often wrong, and it is possible that they are dominated by the kind of business analysis unsuitable for assessing AI impact. But it is an important data point, nevertheless.
"Exponentially cheaper" is probably not accurate. GPT3 cost billions of dollars. It's being created and shared with the world as part of a test phase towards building GPT4+. It's also significantly flawed, such that just about nobody would ever trust using its code over a human at this moment (best case scenario is to have a competent human review the code, which doesn't currently seem much, if any, cheaper).
If OpenAI had to pay for their own research based on marketable products, GPT3 would cost seven figures per access credential and be a miserable economic failure. Or they could charge very small sums and hope people check it out for the novelty still (which is what most people have been doing). It would likely still lose eight or nine-figure sums, maybe 10 figure.
That's not to say that a true general intelligence couldn't learn multiple fields if/when it exists, but current trends do not appear likely to break into any occupations in the next few years.
I don't know how much GPT-3 specifically cost, but you also have to include the costs for GPT-2, and GPT-1, and ChatGPT, and the basic research underlying the various changes, and the costs of running a company that can research these things. It all adds up, and as long as there isn't a marketable product, those costs continue to compound. I've heard from people here at ACX that the total cost of direct AI research is well into the billions. That does not include the various companies building algorithms that aren't really AI (such as Google Search or whatever) that are more limited and actually do make money.
Why stop there? We could count all the money that went into deep learning neural nets. Or all the money that went into AI since the 70s. Or all the money spent on developing computers since the 40s.
It took a lot of time and money to get to the current level of technology. Now that we're here, it only costs pennies for GPT-3 to generate pages of code.
It's not currently at a level where it can replace programmers. It's at the level where GitHub Copilot can generate code suggestions (fairly long ones, too) and a lot of programmers find it useful.
But when it does get to the level where it can replace programmers, it'll probably still be able to generate thousands of pages of code per minute for just a few cents. That's what the parent comment meant by exponentially cheaper than human coders.
Yes, I get that and agree it's the obvious goal of everyone working on AI research.
As far as adding up costs, you wouldn't want to add in any cost center that has already offered profitable products (such as computers themselves), but only the offshoot that hasn't made any marketable products.
If/when such a product is made and can be sold, then all of the research between "clearly profitable and existing products" and "this new revolutionary product based on [new research branch]" can be evaluated for total cost verses what can be made from using the AI. The amount of money potentially available could be huge, but right now it's a huge loss.
>I'm interested in whether people think AI is going to render a lot of professions obsolete in the next few years.
No, definitely not. It may well start revolutionising some in the next few years, and it may well render some obsolete in a few decades, but things like ChatGPT, while impressive, are still a long way off being usable without human oversight for any important purpose, and even when they are adoption will take time.
One of the most exposed professions right now might well be radiologist. AI:s are becoming *really* good at interpreting x-ray images, and are already superior for some conditions. Presumably there will still be human oversight, but this could cut a lot of jobs.
I disagree. Medicine is a field in which both patient satisfaction and liability insurance will keep humans in the loop *long* after it's conclusively proven that AIs could do the job better. Even with an AI evaluating the scans, there will still be a doctor involved. We didn't get rid of doctors when we invented the x-ray (or name your new technological breakthrough that makes being a doctor easier), despite the nature of the job changing.
It's not totally clear from your comment if you're aware that radiologists already generally don't interact with patients directly? They interpret a scan, and sent a report to the patient's doctor who discusses it with the patient. From the patient's perspective, the radiologist is already nothing more than a black box the vast majority of the time.
Your comment ("there will still be a doctor involved") and the comment you're replying to ("One of the most exposed professions right now might well be radiologist") can both be true.
It's the legal and governance system as a whole that requires a human in the loop to blame. Patients may not interact with the radiologist directly, but their lawyers certainly will if something goes wrong and it comes out that the hospital was using AI.
Lapras is correct. Beyond that, humans will irrationally prefer that a "real doctor" is reviewing the scans. Due to the potential of errors and mistakes, hospitals will require that a radiologist checks off on the AI's work - similar to how self driving cars have to have a human in the driver seat even if the car has a better safety record than the human assisting it.
What would really revolutionize software development is a chatbot that could spend hours at meetings, and then provide you clearly written meeting minutes. :D
Another interesting use would be a system of accumulating documentation and tacit knowledge in companies. Searching information in logs. Etc. This actually sounds like something a chatbot should be capable of doing.
I think that the most revolutionary thing to come out of GPT Chat is the natural language improvements - especially the ability to read human input and respond. I'm old enough to clearly remember search engines that only found results with exact matches (no misspellings, no abbreviations, no last name first or vice versa). To see an open-ended prompt that can result in natural language responses is actually a big deal. The idea of an Alexa/Siri-like assistant who really can provide you with correct information on a wide variety of topics (or take your meeting notes) may be pretty close to available now.
If someone invented a "go to meeting" chatbot, the meetings would expand so that reading the minutes would take as long as the old meetings. Human labor would remain constant.
The first jobs that AI would eat into significantly are the ones I least expected it to —the creative sphere.
Plenty of amateur/ entry-level creators are going to have a tough time competing with AI. The first jobs that AI is making an impact on are voice-over artistry, graphic design (DALL-E etc.) and copy writing.
From what I see around me, arts/ humanities graduates are comparatively the most tech-illiterate and evasive to tech. This only adds to the problem
Yes, that is a conversation that I've had multiple times, with people in the "creative" professions, specifically in advertisement and copy-writing.
They are quite surprised that automation came for their jobs first, and recognize that not only most jobs are not creative, but most people are not creative.
In general their problem is not that they will be obsolete, but that the mere existence of instruments such as GPT3 will devalue their jobs and force them to lower their rates: they will still need to to all the interaction with the client and deployment of the result, while saving a trivial amount of time using AI-generated copy, but the clients will try to pay them less because "I can do it myself with a machine". Sort of what happens to translators.
Artist jobs (outside of the super famous ones selling whatever they wanted to paint that day for millions) fill two functions. One is to actual paint/draw/whatever medium the actual art. The other is to determine what art their client actually wanted. Right now that second part is quite important, as it takes artists many hours to produce one art, and getting it wrong would be very bad for everyone involved. If an AI can produce hundreds or thousands of art examples with very minimal time, the balance shifts towards just asking for a new piece of art instead of explaining better what the client wanted up front. For many artistic functions, like the graphic design department, a CEO or VP of Marketing trying a handful of prompts in DALL-E probably takes less time than trying to explain what they want to the graphic designers. They still fill a lot of additional functions currently, but that's a big hit to the necessity of graphic designers in a marketing department.
Citation very much needed on "basically competent" at coding. Ability to write things that look like code? Sure. Have that code be correct for relatively simple problems? Yes, to my surprise. But in hindsight there's lots of code snippets to train on, especially on simple, puzzle-style questions. But extend this capability to novel, complex problems? I still don't see it, even for GPT-5 or whatever. The surface keeps getting more and more polished, but actual understanding is still lacking in the same ways as the previous iterations. It's getting harder to point out definite "tells", but they're still there.
Nope. The sine qua non for any worker, be it human or machine, is reliability. In general, where machines have already replaced men, that is their big selling point: they're not usually, at least at first, cheaper, they can often be finicky to set up and maintain, and they often require significant changes to production methods -- but they are rock-solid reliable, and that gives them an edge.
Same thing with even more advanced "knowledge" work. The single most important thing about a paralegal is -- he doesn't make obvious dumbass mistakes about the law. He may not be inspired, but he won't fuck up writing up an ordinary pleading. Same with the diagnostic role of a PA or NP, or the ability of a middle-manager type to schedule workers, or a junior code monkey turning out bog-standard database interfaces. They may not do the most creative work -- for which you'd hire the more senior star of the field anyway -- but it is absolutely necessary that they don't wander off the rails randomly, that when you assign them some area of responsibility they can execute it reliably and largely flawlessly.
But so far, this is exactly where modern AI falls down. It is not reliable. It sometimes gives the right answer, but other times confidentally says something obviously wrong, or out of left field. You never know exactly how well it will perform a given task, and even the designers can't guarantee its performance. It's like the 1980s-era joke about using Linux on the desktop, that it would be as if you flew an airline that used only Concordes traveling at Mach 2.1 and sold transatlantic tickets for $50 -- but where 1 out of every 50 takeoffs the plane would explode, killing everybody.
So that makes modern AI so far mostly a parlor trick, and very interesting from general principles -- but completely unsuited to slotting into any kind of work more complex than as an amanuensis to a entrepreneurial human. I can see it being useful to artists or writers to generate stuff that they can then re-use, polish, use for inspiration, whatever -- but will always have to check. But it can't be made part of any significantly large and complex work environment, because nobody knows if it will actually do the job on any given assignment, or whether it will make some horrible mistake. However brilliant its everyday work might be, the capability to randomly do something stupid kills the deal.
If you haven't encountered it yet around the rat-sphere, check out The Story of VaccinateCA, about a grassroots effort to collect and publish reliable information about vaccine availability, back when COVID-19 vaccines were just becoming available. It's a long read, but worth it.
Thank you for sharing, this was seriously impressive and heartwarming (in the sense of showcasing humanity at its best when dealing with humanity at its worst -- well, maybe not literally "worst", but you get the idea).
My meta-take is that the government might have done well to pay the pharmacies more per shot administered. That would have incentivized them to go out and round up patients, rather than just sitting passively waiting for patients to find them. That would have raised the possibility of corruption in administration, of course.
They might have also done well to not actively forbid (under heavy penalty) healthcare providers from administering vaccines to Not Officially Approved people, even when those vaccines would otherwise end up in the trash.
I can sort of see a point to the worry that not forbidding this would have bad consequences, such as unmanageable queues at the end of the day (thus generating hotspots of covid spread), or unfairness in the distribution. However, the former issue could be mitigated with pen-and-paper queues or some other implementation, and the latter seems to have happened regardless. I'm having a hard time not believing it was just a very bad decision. Is there a better argument for this rule?
My best guess is that the government internalized the notion that vulnerability to COVID is highly variable between groups. That's why it organized administration into tiers. And it was really worried that some people would find a way to jump the line, which would have compromised delivery to the most vulnerable, so it threatened dire punishments for anyone caught abetting such behavior.
For my money, the author gives short shrift to this really quite reasonable concern. But in fairness, I don't think the government understood the *flood* of shots that would become available in short order. It was acting as though the problem to solve was making the most of scarce vaccines, when the main problem was rapid and efficient administration of as many shots as possible.
Hooking onto this to recommend more of patio11, the man behind the initiative.
Among his other cool ideas and pieces, he writes a great newsletter on banking and the financial services industry - Bits about Money (https://bam.kalzumeus.com/)
I have been wondering whether to take fever-reducing medication (e.g. paracetamol) when I have flu. On the one hand, it feels good, reduces flu symptoms and so on. But I wonder if it has any effect on the duration of the sickness, preventing fever means preventing my body from fighting it, in theory extending the fight for a couple more days.
So is my conclusion correct, that if I'm willing to suffer the symptoms, it's better not to take such medications and let the body fight it off with fever?
Speaking as not-a-medical-doctor, the accepted medical wisdom is that the fever is more dangerous to you than the disease. Preventing fever doesn't mean preventing your general immune response; fever is just one reaction to an infection, and lowering the fever may or may not reduce the duration of your illness. It's just there because, apparently, humans' survival rate was better with the fever than without it, on average, in the ancestral environment. Also see this for why you wouldn't assume your immune system is optimized for the modern environment: https://slatestarcodex.com/2019/08/19/maybe-your-zoloft-stopped-working-because-a-liver-fluke-tried-to-turn-your-nth-great-grandmother-into-a-zombie/
It seems to depend on what kind of infection you have whether reducing the fever shortens or lengthens the duration of your sickness: https://pubmed.ncbi.nlm.nih.gov/11130213/ In any case, I'd err on the side of caution, and not let fever wrack your body worse than whatever you've caught.
From your pubmed link for 2 out of 3 infections tested controlling fever lengthened the disease. Roughly (I think) for flu and food poisoning you’re better off not controlling fever and for Rocky Mountain fever you are better off controlling it. For me this compiles down to “mostly don’t control fever”
I don't know how representative the studied examples are here. This is just a single paper illustrating that in some cases fever helps recovery, but not others, not an indication of whether it is, in sum, more often better to not control fever or not. Fever can be very damaging to the body so the question is if shortening the disease is worth that tradeoff.
I agree this is just one paper etc but it is the opposite of supporting "the accepted medical wisdom is that the fever is more dangerous to you than the disease." Do you have references that support that claim?
If I did, I'd have said that it was official WHO advice or something, wouldn't I? This is the advice I've repeatedly heard from doctors, no more and no less. Treat it as you would any other accepted medical wisdom, like putting a cold compress on a contusion or disinfecting a wound with iodine.
A priori, I would think that fever is adaptive against infectious diseases. Pathogens are quite temperature sensitive, which is one of the reasons animal diseases can't always jump to humans.
The caveat here is that evolution favors surviving and reproducing, which often but not always align with longevity and good health. So it might be possible that fevers are mostly useful in the worst-case scenarios, which are much less likely to arise with modern medicine.
I think someone tested it once and found a minor benefit from not reducing the fever in some minor ailments like colds. Personally I let the fever run, but that's more as a point of principle, and I don't believe there's any strong evidence for it.
That's been my gut feeling, with one exception - I try to protect my brain. My vague theory being that evolution probably hasn't had enough time to "care" about our modern uses of intelligence. So I roll with the fever as much as possible - dressing warm, staying under lots of blankets, hot showers - but I also try to keep cool packs on my head. I think it works out pretty well, but I'm not a doctor at all.
Oh wow. That's really informative. I had naively accepted the "recent advances in epigenetics demonstrate a biological basis for generational trauma" idea. I think "epigenetic generational trauma" lived in my mind as a kind of "biological determinism but somehow not racist because reasons", and I was pretty uncomfortable with that. This clears up a lot of muddle.
I think part of this is because it's hard to distinguish "SARS-CoV-2/COVID-19 HAS LEFT PERMANENT SCARS ALL OVER MY BODY!!!!!!" (Long Covid) from "random person gets COVID-19 and coincidentally gets a completely unrelated illness sometime later" (statistical noise)
"In December 2020, Congress directed the NIH to devote $1.15 billion to studying the long-term effects of COVID, including Long COVID. The NIH established the RECOVER Initiative in February 2021 and it has allocated funds provided through the American Rescue Plan (ARP) Act of 2021 (Sec. 2401) to support this effort. RECOVER awards were made to more than 30 groups, including researchers from hospitals, health centers, and other places. These researchers, along with people affected by Long COVID and representatives from advocacy organizations, worked together to develop the main protocols that serve as the study plan for RECOVER studies. To help manage the administration of the RECOVER initiative, Research Triangle Institute (RTI) International was selected to serve as the Administrative Coordinating Center (ACC)."
Hmm I think ppl got funds if they reported adverse effects post vaccine. 500m went to NYU. I don't see long Covid clinics yet being helpful. I don't think the money has really been spent yet. In the meantime the drag on economy will reach trillions. Ppl will see when staring down the barrel of a gun only.
Well nobody wants to touch pain now after the Sacklers suit, and oxy was great for pain but high dependence, but dependence can be managed by doctors but after suit and what has happened its just a territory that is not worth them getting into, leaving people in pretty bad straits. Also its a panolopy of symptoms under autonomic disregulation - strange heart functioning, diabetes on the rise, headaches/migraines and other neurological issues....
Pain as a symptom of long covid does not seem to be on the list (apart from muscle/joint pains). If you are claiming mystery pains that require opioids as treatment, of course you are going to be treated with caution by doctors. "I need oxycontin for my long covid" is going to be a huge red flag, even if the doctor does accept that you have post-covid syndrome.
pain is typically part of an inflammatory process. What I said is that pain medication isn't really being pursued after the oxy blow up - you literally made my point, people who are in pain and are looking fo solutions are being conflated with people who are drug addicts. Saying you need medication for pain should not be a red flag of any kind. And headaches/migraines are being featured a lot post covid, during covid, and with long covid. Furthermore neuropathy disorders, which can be quite painful, the list goes on and on. But yes, pain is a symptom. In some cases it is neuroplastic and in some cases indicates an inflammatory process, etc.
This is just a tragic tradeoff that the universe (or maybe our current lousy technology) imposes on us: The best painkillers we have are very addictive. We can move the pendulum toward "fewer people in untreated pain, more addicts" or toward "fewer addicts, more people in untreated pain." Both are terrible outcomes. We did the pendulum swing toward "less pain, more addicts" and the result was a huge surge of opioid addictions and a moral panic and gazillion dollar lawsuits, so probably it will be a generation or two until the pendulum swings back. And depressingly, it seems like roughly *none* of the public discourse about this, from journalists or politicians, notes the inevitability of the tradeoff. We're all in a moral panic about opioid addictions now, and that would be off-message.
As best I can tell, most of the deaths from overdose are caused by fentanyl becoming cheap and so mixed into a bunch of illegal drugs by people who have no business at all doing life-critical mixing of drugs. We'd probably have more addicts but a lot fewer overdose deaths if you could just buy a bottle of oxycodone over the counter. But again, it's a tradeoff--deciding where on the tradeoff curve we should be is ultimately a matter of values.
Is there any symptom or group of symptoms that isn't indicative of Long Covid? I'd put a small sum of money that LC is this era's mystery illness panic. It's amazing that there's anyone left alive after PVC contamination, Multiple Chemical Sensitivity, Morgellons et. al.
I also bet if you were in the world trade center you would have completely ignored directives to stay in your office and immediately knew to run down the stairs and get out---incredible how much we know about things we know nothing about!
Come down off your high horse. You have no idea who of us on here have chronic or long-term illnesses, and you are presenting demands and uncorroborated statements as if they are self-evident facts.
You may be ill yourself, I don't know, and if so then sympathies to you. But simply stating "Doctors don't care about long covid, why isn't something being done?" is not enough, particularly as there *does* seem to be on-going research.
That the list of long covid symptoms also apply in other illnesses, and that there is going to be a long time before anything can fruitfully come out of research on long covid, is unfortunate facts of reality. Making snarky remarks about the World Trade Center isn't helping your case or making you sound like anyone other than "I feel that I am sick, why won't doctors believe me?" when you have nothing actually wrong with you.
my point with the world trade center was simply to say that people tend to think very highly of their own response to things that they have not experienced.
Nah, your point was that you don't like me and wanted to communicate your dislike.
As a younger person, I would have obeyed instructions. However, my years of experience, particularly involving emergency response has taught me to trust my own observations v. what someone is reading out of a manual. And since your scenario would have been prior to the big ICS redo, those commands would have been even more ill-informed.
What, exactly, did you think this response was going to accomplish?
"Oh, I have X, therefore all purported diseases must exist, even though there is a long history of bogus disorders which have fueled innumerable books, daytime TV interview programs, websites and medical consulting practices?"
Listen, I had a horrible post viral syndrome myself 20 years ago after the flu. It lasted 3 years. I have no doubt at all that post-viral syndromes exist and wreak havoc with people’s lives. But I am moderately skeptical about a fair amount of what’s claimed about Long Covid. The situation is really hard to read , the research results are all Over the place, and it seems like everybody has an ax to grind. In a messy confusing situation like this omebody can be skeptical without being an ignorant heartless jerk.
One thing to note: post-viral syndromes are a known thing that happens sometimes. Occasionally, you get a case of the flu or some similar virus, and you end up with lingering problems for months or even years--racing heartbeat, occasional fevers, occasional fatigue, etc. I don't think this is well-understood, though. My amateur impression is that it's probably some pathological thing happening in the immune response where your body doesn't turn down the response enough after the infection is cleared or you've ended up making some cross-reactive antibodies or T-cell receptors that are causing some kind of autoimmune thing to happen.
I bet a bunch of long covid is just millions of people getting a viral infection at the same time and some fraction having a post-viral syndrome.
But covid also messes with clotting, and you can imagine ways for that to occasionally cause all kinds of nasty problems. And of course, anything that makes you *really sick* (like hospitalized for a couple weeks) is very likely to leave you with some long-term damage.
Again - even if you think the pain is being created by the brain, it is still very much real (i.e., the can pick it up on functional MRI's). We know less about the brain than we know about the galaxy, so its amazing how many deities are in the comment section, really
If it's being created by the brain, then it has nothing to do with Covid, "Long Covid" is a total misnomer for it, and trying to apply cures or treatments based on or related to Covid will not work -- just as you wouldn't try to cure hysterical paralysis with spinal nerve grafts. You cure hysterical paralysis with a little rubber hammer which proves to the patient that she's not really paralyzed, just riddled with anxiety and neuroses. (This is incidentally why you've never heard of anybody having hysterical paralysis -- it's too easily defeated now, and that's why the replacements are vaguer, more subjective ailments that doctors can never get a good bead on.) Long Covid has to be cured the same way as fibromyalgia and all the rest: by forcing the patient to accept that the symptoms are just her externalizing her anxiety as a fake physical malady. AFAIK we basically don't know how to do this reliably at the moment; the only method that seems to work in even a small percentage of cases is brutal shaming, which our present society is a bit too gelatinous to allow.
There is a group headed by bob Patterson that is able to identify certain elements like cell retention of spike protein to accurately determine who is experiencing long covid and who isn't. The treatment for it is an anti viral and statin. I mean its very easy for you to argue that its just a placebo effect -but you can literally say that about every single drug which would render this entire conversation completely moot. HPV is a virus - causes cervical cancer. A lot of viruses cause downstream problems. You are talking about virus that has been around for two years like you have the noble. prize in science and have a decade of research under your belt. You have no idea but its plausible for a viral infection to cause some problems in certain people as many viruses do. A lot of viruses kick of autoimmune issues. These things are no longer debatable. The reality is nobody has the answers yet.
"You are talking about virus that has been around for two years like you have the noble. prize in science and have a decade of research under your belt."
And you're talking about it like an absolute textbook according-to-Hoyle hypochondriac crank, but you don't see me descending to personal insults.
"These things are no longer debatable. The reality is nobody has the answers yet."
"There is a group headed by bob Patterson that is able to identify certain elements like cell retention of spike protein to accurately determine who is experiencing long covid and who is experiencing long covid and who isn't." What does that mean, exactly? So if you have all the symptoms but he can find spike protein and other element then you really *don't* have Long Covid?" What's his explanation for why those people have all the symptoms. Also, if his statin+ antiviral treatment was effective for many sufferers seems like we would have gotten the news about it by now. I have actually been hunting online for effective LC treatments, trying to help a friend of mine. From what I can see, nobody has found some central thing wrong in the bodies of people with LC, much less come up with a treatment that targets the central thing that's wrong. The LC clinics offer piecemeal treatments -- cognitive exercises for brain fog, respiratory therapy for shortness of breath, etc. But in any case, if you have a lot of confidence that bob Patterson have a treatment that addresses the core cellular thing that's wrong, why are you so worried about LC and LC patients being neglected? If Patterson's got that, it's a game changer.
"I mean its very easy for you to argue that its just a placebo effect -but you can literally say that about every single drug which would render this entire conversation completely moot." Sure you can say it about every single drug. You can call any drug a placebo, even penicillin. However, it is quite easy to test whether a drug is effective or a placebo, so calling a drug a placebo does not render conversations moot. By the way, just looked up Patterson. He has a whole treatment machine going, which is quite expensive for patients because none of it is covered by insurance, since it's an experimental treatment. However, he has not done a randomized control treatment study yet. (That's a study where you randomly give some people your drug and some people sugar pills, and look to see if the people on the drug do better). RCT's are easy to do. The fact that he hasn't done one, yet is actively treating thousands of people, and making millions doing it, is grounds for just writing this guy off, end of story.
I mean....it's not like the vaccines which were distributed to EVERYONE practically speaking, went through the most rigorous studies and we still took those. Ps I just did the patterson test.
Please direct me to a link about this claim, because it immediately smells like our "Real Medical Doctor says Ivermectin is miracle cure!" videos I've seen being touted as "don't believe the fake medical establishment, here's the 'real deal'!"
Googling it brings up a Swedish study about neurological changes to the brain. No idea if this is legit or not:
"Results Twenty-five patients (71%) had abnormalities on MRI; multiple white matter lesions were the most common finding. Sixteen patients (46%) demonstrated impaired neurocognitive function, of which 10 (29%) had severe impairment. Twenty-six patients (74%) reported clinically significant fatigue. Patients with abnormalities on MRI had a lower Visuospatial Index (p=0.031) compared with the group with normal MRI findings.
Conclusions In this group of patients selected to undergo MRI after a clinical evaluation, a majority of patients had abnormal MRI and/or neurocognitive test results. Abnormal findings were not restricted to patients with severe disease."
And an English one, which is good data because they scanned the same group of people before and after having covid, so were able to pick up on any changes:
"Discussion
This is to our knowledge the first longitudinal imaging study of SARS-CoV-2 in which the participants were initially scanned before any of them had been infected. Our longitudinal analyses revealed a significant, deleterious impact associated with SARS-CoV-2. ...Note that these structural and microstructural longitudinal significant differences are modest in size—the strongest differences in changes observed between the SARS-CoV-2-positive and control groups, corresponding to around 2% of the mean baseline IDP value. This additional loss in the infected participants of 0.7% on average across the olfactory-related brain regions—and specifically ranging from 1.3% to 1.8% for the FreeSurfer volume of the parahippocampal/perirhinal and entorhinal cortex—can be helpfully compared with, for example, the longitudinal loss per year of around 0.2% (in middle age) to 0.3% (in older age) in hippocampal volume in community-dwelling individuals. ...Our statistics also represent an average effect; not every infected participant will display longitudinal brain abnormalities."
So not everybody who gets covid will have the brain changes.
*If* this is legit, then that's your problem right there: changes to the brain can't be easily or simply undone, if at all. We're still looking for something to help with Alzheimer's, after all.
Research is ongoing, but it's going to be slow. And most doctors are probably not going to send someone off for a brain MRI simply on the grounds of "I have a cough, muscle pains, and am recovering from covid".
You sound like you want a cure right now, but that's not going to happen. It is going to be years of study to see if the brain changes are permanent, if/how they can be alleviated or even prevented, and what medications if any help improve quality of life.
Anecdata != data, but I can tell you firsthand that a psychosomatic condition caused very pronounced arrhythmia for me, to the point that cardiologists were considering heart surgery (ablation).
Mostly a coincidence. At first I assumed that it was all related to a heart condition that runs in the family. But the symptoms stopped matching very well at some point. I had constant headaches, so was eventually sent to a neurologist, who prescribed me with medication for that. That didn't help with the headache, but helped a lot with all other symptoms (the aforementioned arrythmia, fatigue, etc).
At this point I was just doing the scattershot thing of following every lead and saying yes to every possible referral to yet another specialist. That eventually led to a diagnosis. I also had a pretty good, open-minded GP who helped the process along.
I had said they can pick up pain on functional MRIs proving that its a real experience. I do not know of any functional tests for long covid, simply post moretum autopsies showing spike protein in the brain in the like., but my whole point at the start of this thread is there should be more funding for long covid.
Sure there is, there's lots fake about them. Hysterical paralysis is a fake paralysis, you're not really paralyzed. It's a real problem, but a fake paralysis. Electromagnetic hypersensitivity is a fake allergy, you're not allergic to electricity at all, just a guy with a brain problem situation. You're faking the disease as an external expression of your scrambled endocrine system, or whatever the hell part it is that's on the fritz.
mmm, I think you're using "faking" in an inconsistent way. If a person is unable to move, that's real paralysis, period. If a person feels pain, that's real pain, period.
Usually, when people say "faking", they mean someone is deliberately and deceptively presenting some symptom they don't actually have.
I disagree with this, unsurprisingly. If you're convinced that you're paralyzed (well, in this case, that one of your limbs is), but actually you're not and the doctor can demonstrate to you with a simple maneuver that your arm/leg works fine, then that's a fake paralysis. You may not be *faking* (a word I did not use) it, but it's still fake, the same way that if someone else feeds you a counterfeit bill and you don't notice that it's bogus, you're still spending fake money when you try to use it in turn, even if it's not deliberate. I don't see any inconsistency here.
Sure, we're just using words differently, so in some sense neither of us is wrong. But fake has a very clear negative connotation, and it does call up association with fakers in most people.
Can you try to empathize with how tiring and hurtful it is to be afflicted with one of these conditions, and being flippantly dismissed over and over again? Even by the doctors who are supposed to be treating us, but instead dismiss our conditions as "fake" (and thus not worthy of sick time, medication, etc). The conflation of concepts here leads to real damage.
In particular, if your convinced you're paralyzed but there's no physical problem, we'd *really* like to avoid having some neurosurgeon try operating on your spine to fix the nonexistent problem and maybe accidentally create a bunch of physical problems to go with your psychological problem.
The problem with the way you are talking is that there are at least 2 ways in which paralysis can not be real paralysis: (a) person's leg is not in fact paralyzed, but they genuinely experience it as being so; (b) person's leg is not paralyzed and they know it is not, but they are lying. Those are 2 quite different situations, and it makes sense to use different words for what's up. Also, the word you are using for both, "fake," has a negative connotation; it generally implies that the misrepresentation of something has been done knowingly. If we say a Titian is a fake, we mean that a forger painted it and claimed that Titian did. This implication seems fair enough for situation (b), but not for situation (a).
Some cases surely are psychosomatic, but I'm pretty sure some of the cases involve some actual physical thing happening that has messed the person up. But I also suspect strongly that multiple different things are getting classed into "long covid," and you can imagine ways that might make research on it pretty hard. If 40% of the cases are psychosomatic, 40% are post-viral syndrome that sucks but will eventually resolve, 10% are lung damage from severe covid, and 10% are neurological damage from small blood clots in the brain or something, it's going to be tricky to learn much until those causes are somehow separated out.
Yes, and note that the insanely high statistics like "10% of people have long covid" are a scientific criterion optimized to capture every *possible* case.
Because a lot of people are hypochondriacs, or anti-vax crazies, or just malingerers and all those people make it hard to sort the wheat from the chaff when dealing with stuff like this.
I have a lot of "long COVID" symptoms. but I would never say I have it because I thin it is way too hard to disentangle those possible symptoms from the symptoms of the weight I put on and exercise I forwent, and of life changes I made (plus I am 2 years older). Many people just latch onto whatever trendy explanation soothes their feelies best and say they have "chronic Lyme" or whatever.
Well, it might be hard yes, but indeed even if you believe the root cause is from the brain and nothing else, no lingering viral load, spike protein, or other immune suppressant, then you are left with neuroplastic pain and symptoms, which are produced by the brain, but still very real. I'm glad you can attribute your problems to other things, but if you long covid was debilitating you'd be looking for answers, I think its sad to live in a world where we assume people are just lying, for what purpose? We are talking about 25M people in the US who probably have long covid, and they are all being gaslitt. I'm literally writing a satire about it because its a zombie movie where the zombies are being called zombies and they aren't.
Because the symptoms correlate with a billion other things, and it's like other claimed illnesses like Chronic Fatigue Syndrome etc. You feel tired, lacking in energy, breathless, minor exertion knocks you out - yes, and?
I've had lingering symptoms since my bout of covid (including getting one damn respiratory infection after another in a chain). Is that 'long covid', is it post-viral syndrome, is it something to do with my general state of health and existing pre-conditions?
Here's the NHS website on long covid symptoms:
"Signs and symptoms after coronavirus can be different from person to person. The most common ones include:
cough
fatigue
breathlessness
muscle and joint pain
sleep problems
loss of smell or taste
low mood
brain fog, loss of concentration or memory issues (cognitive impairment)
anxiety"
Well let's see: cough, fatigue, breathlessness, muscle and joint pain, sleep problems, loss of smell and taste and so on are also common symptoms of colds, flu, viral infections, vitamin deficiencies, poor diet/sleep/lack of exercise.... it's not like there is a nice, neat set of discrete symptoms that can only be long covid.
That's why doctors tend to go for "yeah it takes several weeks to get over covid/viruses" when someone shows up and says "I had covid, I was clear four weeks ago, but I still don't feel better" instead of going "ah yes, you are suffering from long covid!"
"Most people's symptoms of coronavirus get better within 4 weeks. But for some people, symptoms can last longer, or new ones can develop. Symptoms can also change over time and can affect anywhere in the body.
Healthcare professionals may refer to long COVID as:
ongoing symptomatic COVID-19 (4 to 12 weeks)
post-COVID-19 syndrome (over 12 weeks)
As this is a new condition, understanding is developing all the time. Experts are learning more about how long symptoms will last, and it'll vary from person to person. There can be different symptoms which often overlap. It can't be said exactly how long coronavirus symptoms will last. The reassuring evidence is that symptoms improve over time in most cases.
Advice is based on:
evidence from research so far
experience and expert knowledge of healthcare professionals
Research is ongoing, so advice may change when its results are known."
And yeah, some people are hypochondriacs/very anxious about health problems, who will imagine they are suffering from long covid and demand specific treatments and get very angry when they feel they have been dismissed.
That's great if we reach herd immunity but we aren't...so....symptoms may improve just to get walloped again? Yes, it's a tricky business but it's not impossible to categorize long Covid into four or five specific fix presentations and research the cause. That's why it should be properly funded. There are many hypothesis to test from all the clinical evidence.
Depends on what you mean by "long covid"? By some definitions, I had it, but that was just because my first covid infection (early 2020) did a number on my lungs, and it took about 6 months to fully recover. I don't think there's much a doctor could do about that.
I'm not sure where to put this, so I'm just going to describe what I believe to be my case of longish covid.
In 2020, my sense of taste got weird. My sense of smell was unaffected or possibly a bit stronger.
Tastes were generally weaker, and what I suspect was mediocre pork tasted actively nasty. What's more, the same food could go from not very good to mediocre to very good in the course of 24 hours.
It seemed to me in a later phase that if something tasted good, it probably was good, but if it didn't taste like much, that might be a problem at my end.
There's been gradual improvement and my sense of taste is approximately back to normal except that I may have lost some capsaicin tolerance.
I did not seek medical help-- it didn't seem like much effective was available. It didn't quite match the usual accounts, but the timing is suspicious.
As for the general point that people talking about exhaustion are failing to allow for deconditioning and being a little older, the exhaustion seems too extreme for that. We're not talking about having to go from a ten mile run to a five mile walk, we're talking about barely being able to function, or not being able to do ordinary activities at all.
Yes - the idea that aging can account for even a 20% collapse of health let alone 100% or more that people are experiencing is just not science based. I'm glad that you seem to have recovered. A lot of people do seem to clear long covid within a year, but of course, some dont' and there is also the issue of reinfection.
At its worst, just after the fever broke, I was literally out of breath from getting up from the couch and walking to the kitchen. Literally gasping for breath. There is NO WAY that was normal. I slowly got better from there. Now I'm more of less back where I started.
There's nothing quite like that feeling of having a hard time bicycling up a hill, feeling down about yourself and out of shape, concocting negative stories in your head, and then realizing that your bike got stuck in a high gear. Unstick the gear, and suddenly everything becomes so much easier, and you realize that all that self-talk was bullshit.
There’s a lot of uncertainty over how much of what’s called Long Covid is in fast a post -Covid syndrome rather than a miscellany of symptoms caused by other things. Some pretty decent research suggest that’s the case, other pretty decent research suggests that it’s not. It’s also not clear if Covid is worse than thing s like flu and mono as regards how many people have a post-viral syndrome and how disabling the syndrome is. M(I myself had a syndrome exactly like long Covid except for no shortness of breath 20 years ago after the flu. It lasted
dor about 3 years, and during that period my average feeling of physical well-being, which previously had been about an 8 on a 1-10 scale, went down to about a 2. It was just miserable) there is decent research for both points of view.
So in that case, one could argue it needs funding. It's like there's a 50/50 chance on gd do better to walk through life believing than not. (Life of Pi ending)
It is not 50-50, it is a mess. It is clearly something that can be settled with research. It is not an unanswerable question. If I had followed the research more closely I might already think I had a pretty clear idea of what a fair reading is. As it is, I am inclined to think that genuine Long Covid syndromes are much rarer than the hype suggests. A recent study I read followed a bunch of people, half of whom had Covid and half of whom had another respiratory illness. 3 months out, those who had had flu, RSV or some other similar thing had slightly*more* symptoms from the Long Covid list than those who had had Covid. I didn’t note any ways the study was badly designed. Seemed like a good study to me. Seems to me that a fair minded person reading it is going to have some doubts about how large and unique a problem Long Covid
I think that study had 50 ppl in it. Classic long Covid versus post viral issues may be different things in any event but ppl are getting massively disabled either way and all I said is we should be investing more into funding (this was my original point many comments ago)
You may be right about number of subjects. I’m not in a place where I can look up that study. However, the stats they did indicated that the result could be trusted I don’t know whether you’ve had a stats course, so here’s what that means. Theoretically any study can have invalid results just because you happened to get treatment groups that were not representative of the population. However, if your groups are huge that’s very unlikely to happen by chance. If they are small, say 3 people in each group, it wouldn’t be a bit surprising, right? Statistical tests of significance take all that into account. They basically impose much stricter requirements on studies with few subjects. — you have to find much larger effects for them to count as genuine evidence. OK, so 50 is not a lot of subjects, though it’s not tiny. But apparently the results they found passed the stats test, which would have taken into account the fact that there were not all that many subjects.
Seems to me finding should go into research. Maybe that’s what you had in mind. That is much less money than would be involved in a program operating under the assumption that Long Covid is a unique, serious and widespread disease. Under that assumption we should have government funded treatment centers, tax breaks for the sick, possibly stipends for them, , requirements for employer accommodations, etc.
I also looked up a recording of the real Oppenheimer to see which American accent he had. His voice reminds me of Mr. Rogers (from an American children's show).
American here, the accent sounds good to me. Various southern accents are hard to do well, but this neutral newscaster-y one perhaps not as hard. The trick it seems like for British actors is to not overdo the flat or twang or how hard the consonants are hit, but he manages a lilting quality here while still sounding American.
"In October the SEC charged celebrity Kim Kardashian with touting digital tokens on social media without disclosing the remuneration for her endorsement. (She settled with the SEC for $1.26 million without admitting or denying its findings.)" (WSJ 12/19/22)
It would really be a tremendous service to the world if the FBI/NSA/CIA took some time off from constantly fucking the dog to do a big sweep of the whole online "grift" world and just stick a couple tens of thousands of these people in prison. Would really help society in the long run, and reinforce that there are costs to rampant fraud. It is a huge and non-productive waste of society's resources the number of people whose full time job is stealing from suckers.
Like Logan Paul should probably be tried (and convicted) for capital crime, along with that whole ecosystem. You think I am kidding, I am not.
Strongly agree with this. Getting them to shitcan the ubiquitous surveillance of American citizens unfortunately seems impossible, so they might as well do something socially positive with it.
Well, there's also the fact that the energies of visible and UV photons interact with valence electrons, which is kind of necessary if you're going to have a meat-based biology.
There is a lot more ultraviolet and infrared light in our environment than we're able to see, but it's not exactly our environment that we've evolved to see in. That would be the ocean. There's a valley in that absorption spectrum that corresponds exactly to the middle of the range of visible light, which any animal evolving eyes would have had to target to have the best chance at seeing through murky ocean water (and, to a lesser degree, eyes made of water). We seem to have inherited that spectrum even though there are many more wavelengths that can easily be picked up through the air.
If you are talking about all else being equal, not very interesting because there are few natural sources of xrays/gamma rays to begin with that could illuminate the world.
If we are talking about the radiation being abundant enough so there was an evolutionary incentive to develop vision for it, still not very interesting because that amount of ionizing radiation would inhibit the formation of complex organic molecules necessary for life.
Specifically, there are five major types of radioactive decay: alpha (helium-4 nucleus emission), beta+ (positron emission), beta- (electron emission), gamma (direct gamma ray emission), and electron capture (exactly what it says on the tin. Any given radioisotope will typically only do one or two of these, although decay chains (several successive decay steps as one radioisotope decays into another radioisotope, before eventually reaching a stable configuration) often hit most or all of them.
Gamma decay of course emits gamma rays directly, and beta+ decay emits a positron that will usually soon find an annihilate with an electronic, releasing their combined energy as a gamma ray. Alpha and beta- decay don't emit EM radiation directly, but they do release high-energy charged particles whose kinetic energy can generate x-rays when they hit something.
I don't think electron capture generates x-rays or gamma rays in any meaningful amount: just capturing the electron an releasing a neutrino that will very probably pass through ordinary matter without interacting with it (beta+ and beta- also emit neutrinos, but they're often ignored because the electrons and positrons are much more interesting). But I just found out about electron capture in this sense a few minutes ago when I looked up some stuff to confirm my explanation of the details of the above, so my confidence in that is on the lower side.
It seems to me that software development is moving in the direction opposite to "division of labor".
Twenty years ago I would naively expect that the growing field (in both budgets and complexity) of software development will allow lots of obscure specializations. But what I observe suggests otherwise.
The idea of "full stack developer" goes again separating database from business logic from web page development from design. In "scrum", team members are officially treated as replaceable, so one day you design a database table with partitions that will contain millions of records, and the next day you tweak stylesheets to move a button one pixel leftwards on a mobile screen.
With "devops", programmers and technical support become one; and with "cloud" they are further united with network administrators. The guy who partitions database tables and tweaks stylesheets is now also responsible for designing and implementing network security. And he is supposed to be simultaneously a specialist on all of that.
I can imagine the reasons why companies could prefer this. With the everything-specialists it seems like you can hire fewer people and save money. (Not really, because you need to hire multiple everything-specialists anyway. Might as well hire a few specialists instead.) Or maybe the idea is that everyone is replaceable because everyone has all skills and knowledge, and this should keep the salaries down. (But does it actually work? Companies are complaining about lack of developers, which is probably made worse by the need to know everything, and thus the salaries are growing anyway.)
This seems to go against the established wisdom about the "division of labor" which still seems generally accepted, so I wonder if any economist has noticed and studied this paradox.
Also, I am curious whether this trend is specific for software development, or whether it exists in more, maybe even all professions. (I mean, requiring people to do what was previously multiple specializations. Not because you openly need to save money, but because you explicitly promote the belief that a *true* specialist is a specialist on everything.)
You are kind of avoiding the main argument from the "cross-functional team"-people: having lots of specialists that sit in silos creates a huge communication overlay, that eats all the efficiency gained by specialization and then some. This argument is largely true in my experience.
The argument for a multifunctional *team* is sound. I have seen places where network administrators were on a different floor, and web designers in a different building, and yes the communication was extremely slow and frustrating. I understand why it would be a good idea to make them all sit in the same room and talk to each other.
But abandoning the specialization completely, that seems like going too far.
Also, in my experience, the team members naturally try to separate into roles they feel most comfortable with. If you have five "full stack dev sec ops" on the team and leave them some autonomy, you will probably soon find out that one of them makes most of the database query optimizations, another one writes most of algorithms, another configures the deployments scripts, another writes most of automated tests, etc. Apparently, this does not slow them down on a normal day. (May become a problem when one of them is on a vacation, and his part needs a quick fix.)
Developing specialization and a division of labor particular to your team is part of scrum. That's part of why you're supposed to keep scrum teams together for a long time rather than constantly rotating people from one to another, and build them with a balance of skills. You're only supposed to be good as a generalist to the extent that it's usually better for you to do something off your specialty *with occasional help from other team members* than sit completely idle while waiting for others to be done.
> That's part of why you're supposed to keep scrum teams together for a long time rather than constantly rotating people from one to another, and build them with a balance of skills.
Thanks, I didn't know that! And yes, that makes a lot of sense.
(And, as usual, many companies doing "scrum" completely ignore this part.)
Having an official definition allows us to distinguish between "this is written in the book" and "this is what my company does, but it is not written in the book". And yes, it often happens that the things that managers insist on in the name of Scrum, and that programmers usually complain about, are actually not in the book.
For example, the book says:
> Scrum Teams are cross-functional, meaning the members have all the skills necessary to create value each Sprint. They are also self-managing, meaning they internally decide who does what, when, and how.
Surprisingly or not, Scrum neither mentions the need of "full stack developers", nor does it say that each ticket should be assigned to a random team member who happens to be free at the moment. Instead, it says the opposite, that ticket assignment should be decided internally by the developers team. So if the team agrees internally that one developer does the database and another developer does stylesheets, that is actually 100% Scrum compatible.
(Note: This is not an argument for Scrum being perfect, or even efficient. Just that it is meaningful to point at a bad thing and say "this is actually not Scrum", even if your manager insists that it is, even if most managers in most companies insist that it is.)
In the admittedly special case of IT consulting, the obvious advantage is making it easy for management to assign developers to projects of uneven length. If *every* dev can write the UI, server code, database, and network infrastructure, you just plug and play.
This isn't a new issue. Brooks wrote "The Mythical Man-Month"[1] in the seventies and it is still astoundingly relevant today. Most programming tasks are not fully atomizable, and the greatest expense in scaling up your work between multiple programmers is the inefficiency in communicating tasks and delineating task/component boundaries. The more you have people working across those task boundaries, the easier it is for each individual to contextualize where and how their component fits into what Brooks called the "programming system" (what we'd just call "the product" nowadays), and therefore the easier it is to do away with that inefficiency.
Yes, in medical data and apps, I've definitely observed this. I wouldn't say there's no specialties, we definitely have guys who are better and tend to work on different things but being able to do "everything" at least moderately well is important.
I'm not 100% sure why this is, I'm appropriately wary of upper management, but I know it's important, for obvious reasons, to have clinicians and lawyers and security at least in the loop on what we build. Which means putting together a project or tool involves 5 people, only one of whom is building the actual thing. If there's two technical people involved in building the thing, it's ok, but more than that is unworkable. Honestly, more than half of our staff feels like they're building internal tools for other staff members to actually deliver value. Weird but true.
Maybe the lack of specialization is why so much software seems to be utter garbage? Just a random example but our county recently spent ~$11 million to update a very old say 2005 era scheduling system that was pretty bad and functioned poorly for users.
Somehow after $11 million they somehow settled on a release that is worse. Like EVERYONE hates it (except the people with direct ownership of the update). Its main virtue seems to be that instead of looking like it is from 2005, it looks like it is from 2009. It also has a lot of cool functionality that might be helpful if the whole thing wasn't so badly designed. But also all the cool functionality is broken (and its been out over a year, and doesn't sound like it will be fixed), so who cares anyway.
This is a project I am positive I could have provided a better product of with IDK $50-200k. How did they spend $11 million and end up in a worse place. It is like the scheduling system was a D-, they had a million meetings, a fancy developer, and came out a year later with an F.
Other industries don't have this, no--you'd never see a job posting from Boeing for "full stack engineers" to work on their new fighter jet development.
This doesn't work as a management or development strategy, software uses it because it seems to like to aggressively insulate itself from the concerns of the outside world. It's why a lot of software is so bad (eg, no webapp can get scroll positions that don't jump around or behave randomly), why there are constant data breaches even at big companies, etc.
Its result is needing far more developers than are actually required. Instead of a small number of very-skilled specialists, you have a large number of relatively unskilled "generalists" who don't know how or why (or if) anything works, but do know a lot about random toolkits and APIs of the day.
This (at least partly) explains why you'll see small startups of just a few actual developers do something really impressive, then it gets bought by a large company who throws hundreds of devs at it, and it gets worse, or at least not proportionately better.
This mostly works out okay because most software development is actually really easy, most software is actually relatively unsophisticated, and most tech companies are relatively high margin and so can afford to do things other industries would find laughable.
I think that the computer programming profession is still discovering what makes for a sensible division of labour.
Established professions had to figure these things out long ago -- for instance "carpenter" and "bricklayer" are professions but "wall builder" is not; trades professionals specialise in a particular material rather than a particular type of structure, so if you want a wooden wall you call a carpenter and if you want a brick wall you call a bricklayer.
Computer programming (must we call it "software engineering"?) is still at the stage where it's figuring out what skillsets make sense to combine into one person and what skillsets are better spread across multiple people.
There's definitely something to this. Like there's the paradox where you kind of want security to be a specialty because it requires a very particular mindset but you can't do that because that mindset needs to be applied to every line of code, and every single little thing you do as a company to be effective.
>A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects.
I detest that he cited butchering a peaceful animal as a useful skill, but this is my own belief as well. It's a tragic fact that knowledge generated by our brains so swiftly and grotesquely exceed and overwhelm them, but it's a fact that should and must be resisted, and can be locally thwarted.
It doesn't violate any separation of concerns principle in software to do so, the GUI can still be seperate from the database without necessarily requiring that a seperate brain design each of them. Good interfaces force good seperation, bad interfaces can't save you even if you had a million strong army of developers each micro-cultivating their own sub-micro-nano-pico-femto service.
My problem with companies and corporations cynically adopting the generalist ethos is that it's nakedly self-serving and not even truly generalist in practice.
The nakedly self-serving part comes from thinking a generalist developer is a general purpose 1-man team with the magical ability to do a team's job with a man's pay.
The disappointing failure to adhere to generalism comes from their expectation that your abilities are hyper-specialized to their own needs, plug-and-play, and can be turned on and off with point-and-click ease. They don't really want a db administrator, they want a guy who knows all the ins and outs of MongoDB version m.n.k that they are currently using. They don't want any old devops guy, they want a guy who knows every last flag and switch in AWS's interface. They don't want a Knuth-style algorithms-and-datastructures guy, they just want someone who knows every exact irrelevant word question asked in the leetcode interview by heart. They don't want a Heinleinen human (who necessarily will always be T-shaped to tradeoff depth for breadth), they want a million hyper-specialized insect all controlled and dispatched from the same human-shaped interface.
> Principles-wise, division of labor is bad. [...] but it's a fact that should and must be resisted, and can be locally thwarted
Why? I don't see a single argument why specialization is bad, except for a quote from Heinlein, who isn't even an authority on this matter (so it's not an appeal to authority – maybe "appeal to edgy sounding author"?).
I didn't want to dive into my motivation and reasoning for hating specialization because that wasn't the main point, my point is that what companies are doing are not really "moving away from specialization", but worse specialization.
Specialization is bad because :
(1) it's unsustainable, how many times can you divide a stick in half before you no longer can hold it in your hand ? how many times can you divide a field of knowledge into tinier and tinier pieces before it's no longer a meaningful and coherent body of knowledge but jumbled mess of footnotes-to-footnotes-to-footnotes-to-footnotes-to.... some other body ?
Henri Poincare :
>Science is built of facts the way a house is built of bricks: but an accumulation of facts is no more science than a pile of bricks is a house
This also implies that a *division* of a science (or more generally a field of knowledge) does not necessarily produce a science. (in the same way cutting a house in half does not produce 2 houses, but 0 houses) i.e. When something has a delicate structure, you can't naively cut it in half. Specialization rarely if ever follows the (or a) "natural" structure of the fields it cuts through, it's nearly always administrative bullshit or historical accidents.
(2) it imposes an artificial structure not present in the knowledge it divides. Specialization is a classification system at its core, and all classification systems are made up. Are Databases systems programming ? Network programming ? AI ? Compilers ? All of them. Specialization in the weak sense is already what generalists naturally do, which is T-shaping : taking a broad survey then diving in. But specialization is almost always taken to far more unjustified and ridiculous reaches
(3) Because, simply put, variety is good. I don't know how else to phrase this, I think that nearly all people would prefer more variety of food, more variety of people, more variety of places, more variety of languages, more variety of anything IFF all else is kept constant. Specialization is bad because it chokes the diversity and variety of knowledge available to you (again, weak kinds of specialization are the motte, and are already practiced by nearly everybody who knows being human entails cognitive limits, that's not what I'm talking about)
(4) In the context of governance and civilization-building, specialization enables the "Seeing Like A State" phenomena and a myriad other forms of oppressions, because it breeds an entire class of people so far removed from actual value-creation aka getting shit done, a class whose entire reason for existence is finding reasons for their existence. This is what Heinlein is referring to with "Specialization is for Insects" closing remark, and is supported in practice as the early rise of city state civilizations enabled unprecedented authoritarianism and alienation.
To put this in the most blunt form : You are a monkey, and monkies don't specialize. If you were an Ant or Bee, you would specialize in fighting or breeding or governing and you would be happy and content to serve/be the queen, but you're not, you're a monkey, and monkies don't specialize, not permanently, not to the grotesque degree modern hairless monkies invented. Just because your ancestors invented computers and the nuclear bomb doesn't make you not a monkey, you're still a monkey, forever bounded by your nature to love certain things and hate certain things.
The "T-shaped generalist" is a motte-and-bailey idea. Most people (including me) would approve of the idea of knowing a little bit of everything, and a lot of something.
But how much precisely is the "little bit" that you are supposed to know of everything? Enough to solve each randomly selected Jira ticket?
If you say "no", then you are *not* a literal full-stack developer. There are parts of the stack where you cannot deliver solid work without help.
If you say "yes", then exactly what is the point of knowing the "lot of something", given that you can already solve all Jira tickets *without* it?
Its not win-win. Multi skilled techies are going to.be rarer and more.expensive than monk skilled ones. The problems of getting monk skilled teams to coordinate aren't unprecedented or.insoluble, they are just routine management issues. A lot.of.things in software development are.positird.in management just not being very good.
Nevermind, I remembered what Chesterton's Fence is: "change should not be made until the reasoning behind the current state of affairs is understood"*. Then again, this principle could serve as a Fully General Counterargument against any kind of technological change, even though technological change as a whole is beneficial for society** ***.
Chesterton's fence is not a fully general counterargument against technological change as we understand well the reasoning behind the current state of affairs in many case, especially in purely technical questions.
It seems like you're not really leaving a lot of room for nuance in this situation. We can be wary of abandoning state-issued and state-controlled currencies and still think there can be some merit to experimenting with cryptocurrency. Chesterton's Fence is a call for humility, not stasis.
Yes and it is invoked constantly by a certain democraphic for stuff that actually isn't a big deal or has even been tested in many other markets/countries but nowhere to be found on crypto, which has many very obvious problematic elements. Chesterton's Fence is just my ideas are good and your ideas are bad sophistry.
In the case of crypto there basically is no nuance. It is all bad. Crypto is a tool for deregulation and for obvious reasons it empowers immoral wealthy people and harms normal people. Tax dodging, black market stuff, hiding money in a divorce, pyramid schemes and ponzi schemes.
Chesterton's Fence says you need to understand *why* something exists before you do away with it. And in the case of regulation tech bros clearly don't understand why it exists which is why they are so hyped for crypto. Typical "get the government out of the way stuff".
The same reason they don't like unions and hold other right wing economic views.
Furthermore the vast majority of people don't actually use decentralized crypto, they use stuff like FTX, because currency decentralization is too hard to manage for 99.9% of the people.
This isn't a very charitable (or accurate) reading of Scott's post. He gave specific examples of how crypto can empower regular people, and I think we'd both agree that in certain countries helping people break certain laws/make black market transactions is a pro not a con. I don't know the stats on how much of crypto usage these cases cover (my prior is it's a small percentage) but the title of Scott's post is 'Why I'm Less Than Infinitely Hostile To Cryptocurrency', not exactly the strongest endorsement of cryptocurrency.
"In the case of crypto there basically is no nuance. It is all bad."
Tell that to people living under authoritarianism or high inflation that want to protect their wealth where it can't be debased by the government or confiscated by banks.
It is incredibly ironic to watch many of the pro-Chesterton's Fence people, including Scott with a time machine, defend cryptocurrency, fiat money, the south sea company, free silver, bretton-woods, the solidus...
Like, on the one hand, yes, I think there's validity to this. If a significant chunk of the world's currency transactions shift to crypto, there's going to be a ton of unexpected side effects and people are underestimating that.
On the other hand...like, have you talked to a crypto bro? Because they are 100% confident they know exactly why that fence is there, quoting a scholarship going back 100 years to a bunch of dead Austrians. I would accuse them of many things but never insufficient research into money and monetary policy, at least not the old/hardcore Bitcoiners.
I think humility is called for but I can't accuse the Bitcoiners of not having done their homework.
I've encountered a wide range of nuance and depth from enthusiastic pro-crypto arguments. There's a decent amount of thoughtful (albeit usually deeply unorthodox) economic and political philosophy arguments of favor of crypto, but there's also a strong strain of what seems to be crypto maximalism as a tribal affinity marker, as well as unnuanced "BTC TO THE MOON!!!" cheerleading based on little besides linear extrapolation of past trends.
The last of these camps has been pretty quiet for the past several months since major cryptocurrencies have slid substantially from their late 2021 peak, leaving the former two camps much more visible by comparison. But they were very, very loud throughout 2020 and much of 2021.
If you compare the arguments of even the top 20% least bad crypto bros to the kinds of arguments Chesterton was opposing when he talked about the fence, they are substantially similar.
I think a good analogy now from student politics types, but also credentialed professors in relevant fields, is pro-puberty blocker trans activists.
Oh yeah, there's a lot of crap Bitcoin arguments, but I don't blame movements for their fanatics. I always think back to that old Steve Yegge article on Ruby vs Python: https://sites.google.com/site/steveyegge2/anti-anti-hype
"The thrust of Eckel's article appears to be that hyper-enthusiasm is diminishing the Ruby camp's message, and it's spoiling a good gentleman's argument. Those darn hyper-enthusiasts are focusing relentlessly on how cool Ruby is and how much they like it, when what's really needed here is a balanced, objective, neutral, moderated, standards-based, point-by-point, academic discussion of Python vs. Ruby, in which we can all make well-informed decisions, and may the best language win, as long as it's Python.
Python folks never really did understand marketing."
Like, I wish everyone liked rational argumentation, but they don't, and dumb cool hype with giant boobs is a necessary step in the widespread adoption of new technology.
The key is they absolutely don't understand why the various fences crypto exists to tear down were built. It is clear from their arguments. They are exactly the kind of reformers Chesterton was talking about, even if the topic differs.
I think "understand" is doing a lot of work in this critique. After a few hour long discussions over Bretton Woods, I think they know what the arguments for fiat currency are, I think they could pass a political Turing test, I just think they don't disagree with them for a variety of...debatable reasons.
Dude, I'm not the right guy to defend Bitcoin here, but you can't claim that the Bitcoiners don't understand Chesterton's Fence if you're avoiding their big theoretical argument for Bitcoin.
I'm not sure what "domestic intra country" means but it sounds like you have a highly specific critique of Bitcoin that, to be fair, they might totally be ignoring, but that doesn't mean you get to ignore their core arguments, otherwise you're just kind of arguing past each other.
What is the existing fence? Cryptocurrency is a new idea, non-governmental currency is not. Privately issued currency was common in the U.S. in the 19th century, in Scotland in the 18th.
If cryptocurrency were truly new, Chesterton's Fence would be irrelevant - the reason it was never tried before is because it was impossible to try it, not because there was some logic to the previous state of affairs.
If it's just a higher-tech version of privately-issued currency, then the Chesterton argument makes more sense, in that presumably there was time to test the privately issued currency and decide they didn't like it and switch to the current system.
"then the Chesterton argument makes more sense, in that presumably there was time to test the privately issued currency and decide they didn't like it and switch to the current system."
It's not a surprise that "they", government, didn't like competing currencies. That's not an argument why the general populace should oppose private money.
There are now widely available apps that can generate novel images of an individual using a small 'prompt' of images provided by a user. An example of this is the Lensa app. This was immediately followed by abuse, e.g. generating celebrity nude images using a prompt that combines celebrity face shots with nude shots of others. It's not hard to anticipate an ever expanding category of embarrassing deep fakes that could be generated using just publicly available face shots.
Is anyone aware of strategies to counter this technology, when your career requires making a small number of profile pictures available on the internet? Adversarial perturbations seem promising (like the barely visible noise that makes image classifiers mistake pandas for worms or whatever) but I haven't found any hits on Google using just the obvious keywords for this kind of technique.
"People can already craft fake nudes ... So I don't see any real impact"
Presumably a lot of the distinction will be when you can use AI to deepfake a celebrity's appearance and voice into a pornographic *film* and not just a still image. Then once you have a 4K clip of that, you make a "fan cut" of your favorite Marvel film crammed full of obscene, degrading sex scenes and post it on torrent sites. There's undoubtedly a market for that kind of thing, and society such as it is now is totally unprepared for something of that sort to proliferate. Think of how feminists talk about what young famous actresses go through *now*: how are they going to sound when everyone can have non-stop access to hyper-degenerate, realistic pornography featuring every famous woman, on command, personalized by AI, 24/7, and you can't possibly stop them without shutting down effectively the entire internet?
I freely confess to not being an expert, but my impression was that the quality of these fakes has hitherto been poor, and they're also restricted to still images.
Also, I wasn't suggesting gimping AI; I was suggesting that society will have to adapt to this new reality but is wholly unprepared to do so. The feminist discourse around a woman's dignity, revenge porn, Weinstein, and so on is entirely ascendant and also entirely incompatible with almost every conceivable future. My claim was not "here's what we should do", my claim was "I predict a conflagration".
The long term solution is to change cultural norms to normalize sex work and make it so that having identifiable nudes on the internet isn't career-ending, because there's no reason it should be; despite OpenAI's attempts at neutering models, this genie isn't going back in the bottle.
In the short term, it seems the best bet is to use simply reuse the same headshot or at least ones from the same angle. Most of the apps I've seen recommend getting faces from differing angles, and different photos. Limiting the number of choices seems like the best bet there.
But really, I feel like this isn't something worth wasting brain cycles worrying about. Unless you have some sort of very specific job, it's unlikely that someone is going to be particularly motivated to single you out for fakes; and if they really are motivated, there isn't much you can do.
If convincing simulated nude photos can be made of anyone you have a few clothed photos of, then the existence of a portfolio of nude photos of you will not be evidence of anything. But there's a transition time where someone generates a simulated set of incriminating photos of an enemy and uses it to convince an unsophisticated audience (or one whose motivated reasoning in the case turns off their skepticism) to fire or shun or refuse to vote for the target.
I think the counter is slowly society will go back to older methods. There was a period of time when photographs were taken very seriously, because they were very hard to doctor. Before that you had to rely on eye witness testimony and cross checking with multiple accounts and pieces of evidence.
As photoshop skill became more common people became less trusting of photos, and the same will happen to video eventually. So rather than needing to "hunt down" all the offending fake videos because they will offend people. People will (hopefully) just update their world model with the information that videos they see are very possibly doctored and not very good evidence of anything absent other pieces of evidence.
Photos have been fairly easy to doctor for most of their existence, compare things like the fairy hoaxers (and similar) or the Stalinist practice of removing people from pictures. I think it was at most a few decades between the development of "casual" highly detailed colour photography and the time photo editing caught up with it.
So I think photos were never really considered good evidence, unless there was some circumstantial evidence that would make them seem tamper proof even today (e.g. they were taken on Polaroid by a person known to have no image manipulation skills).
The ability for anyone seamlessly doctor videos (without heroic effort) is somewhat newer but has also been around for a few years.
I think deepfakes in the sexualized sense are great, the more convincing the better, because they offer plausible deniability for people whose actual nudes come out, and make it easier to turn the conversation around to be about the publisher and not the subject of the photos or "photos". I don't feel as sanguine about the social possibilities of deepfakes in the non-sexual sense because they offer the same plausible deniability to people doing things that are actually bad.
There's something a bit chilling about this idea, that the impact of a controversial event can/should be mitigated by a background level of distrust. I can see how in this case its beneficial for a victim, but this mechanism is also the basis for e.g. state-sponsored misinformation campaigns. In Russia, instead of denying a government misdeed it is more effective to circulate nine more very different accounts of the event until the first sounds just as implausible.
There may well not be a fully general one - the genie is out of the box.
I believe it's quite easy to tweak this sort of AI to add imperceptible watermarks, which is something responsible designers may do, but that just means that that section of the market will go to their irresponsible competitors.
And there will probably be an arms race between AIs that generate deepfakes and AIs that identify them; I have no idea how that will go.
But even then, identifiable deepfakes are still probably usable for a variety of nasty purposes, and they're only going to get better, easier to generate and commoner.
After toying around with ChatGPT and reading various twitter threads/articles about it I've come to realize that I am far easier to fool than I would've guessed. The replies ChatGPT gives me that I am confident are incorrect sound remarkably persuasive on first glance. Combine this with the fact that my knowledge of facts is limited (though above the Clearer-Thinking-test-taking average apparently) and I'm now second guessing many of my favourite substack/blog posts. I know Knoll's Law basically covers this, but I think this deserves more attention. So much of the focus has been on the ability of ChatGPT to be persuasive, rather than on how easily we're fooled.
Oh yeah a well crafted piece of bullshit that has a few of the right "protein receptors" to lock into our existing knowledge in a smooth way, and perhaps an enzyme or two to flatter our preconceptions generally undergoes very little scrutiny. There are a million examples.
> that has a few of the right "protein receptors" to lock into our existing knowledge in a smooth way, and perhaps an enzyme or two to flatter our preconceptions
I see. I understand the valid points. I am not going into debating the validity of the formula. ( that I even misspelled in my initial post, as I was legitimately promptly corrected.
My initial intention ( probably not clearly expressed), was to ponder on the universality attached to it by the following message of the substack motto .
" All the rest is commentary ".
Which is exactly what I am doing.
Is it axiomatic or relative (and I am not asking philosophically ( death is absolute or relative because atoms get recycled).
Just from a pure generalized ( or not ?) applicability to that.
I am not debating, nor have the intention. I am genuinely asking and gave a ( flawed ?) example.
I've always thought of Bayes Theorem as analogous to Carnot Efficiency. It's supposed to give you an idea of what's conceptually being worked towards. You probably can't draw a perfect circle, but understanding the definition of the platonic ideal of a circle (viz. the set of equidistant points around a locus) will allow you to draw an approximate circle with reasonable fidelity. artistic heuristics about drawing a circle are "optional commentary".
Also, Bayes Theorem is an identity. It's is axiomatic in the same sense that trig identities are axiomatic. not sure what you mean by the "death is absolute or relative" thing though.
The collective noun for a group of wombats is a wisdom of wombats. This is one of those supposed facts which appears nowhere other than on lists of collective nouns.
My question is: who the heck comes up with these collective nouns, and what the heck is the process behind it? I can see them arising naturally in animals which naturally hang around in groups, but these animals are the ones with the dullest collective nouns like "flock" and "herd". The animals with really exotic collective nouns like "wisdom" or "bloat" tend to be solitary or small-group animals anyway.
Okay, you led me down the rabbit hole. I had a vague notion that collective nouns like that probably came from mediaeval bestiaries, and while the period is about correct, it seems (in English at least) that it's "terms of venery" - that is, specialised terminology that the nobility and well-educated would use, to mark themselves out from the common sort.
Also following fashion. Also probably scholars inventing and distributing these kinds of lists as a word-game. And in modern times, when new animals were discovered, people making up collective nouns to fit in with the traditional ones.
"The tradition of using "terms of venery" or "nouns of assembly," collective nouns that are specific to certain kinds of animals, stems from an English hunting tradition of the Late Middle Ages. The fashion of a consciously developed hunting language came to England from France. It was marked by an extensive proliferation of specialist vocabulary, applying different names to the same feature in different animals. The elements can be shown to have already been part of French and English hunting terminology by the beginning of the 14th century. In the course of the 14th century, it became a courtly fashion to extend the vocabulary, and by the 15th century, the tendency had reached exaggerated and even satirical proportions.
The Treatise, written by Walter of Bibbesworth in the mid-1200s, is the earliest source for collective nouns of animals in any European vernacular (and also the earliest source for animal noises). The Venerie of Twiti (early 14th century) distinguished three types of droppings of animals, and three different terms for herds of animals. Gaston Phoebus (14th century) had five terms for droppings of animals, which were extended to seven in the Master of the Game (early 15th century). The focus on collective terms for groups of animals emerged in the later 15th century. Thus, a list of collective nouns in Egerton MS 1995, dated to c. 1452 under the heading of "termis of venery &c.", extends to 70 items, and the list in the Book of Saint Albans (1486) runs to 164 items, many of which, even though introduced by "the compaynys of beestys and fowlys", relate not to venery but to human groups and professions and are clearly humorous, such as "a Doctryne of doctoris", "a Sentence of Juges", "a Fightyng of beggers", "an uncredibilite of Cocoldis", "a Melody of harpers", "a Gagle of women", "a Disworship of Scottis", etc.
The Book of Saint Albans became very popular during the 16th century and was reprinted frequently. Gervase Markham edited and commented on the list in his The Gentleman's Academic, in 1595. The book's popularity had the effect of perpetuating many of these terms as part of the Standard English lexicon even if they were originally meant to be humorous and have long ceased to have any practical application.
Even in their original context of medieval venery, the terms were of the nature of kennings, intended as a mark of erudition of the gentlemen able to use them correctly rather than for practical communication. The popularity of the terms in the modern period has resulted in the addition of numerous lighthearted, humorous or facetious collective nouns."
They've also been very popular with crime novelists, how many variations on "A Murder of Crows" or "An Unkindness of Ravens" have you seen used as a murder mystery title?
I’ve read that the term “pack of wolves” originates from this game, which if true is surely the most essential term produced by it, the sociality of wolves being their defining trait. A word like “herd” just doesn’t seem appropriate for a predator. And there’s nothing fanciful or silly-sounding about “pack”.
That's invalid; if you say "a den of foxes", you will be referring to the set of foxes that live in a particular den, a family, not a collection of foxes that happen to be in physical proximity to each other.
> The animals with really exotic collective nouns like "wisdom" or "bloat" tend to be solitary or small-group animals anyway.
The exception would be "pride of lions", which has a lot of popular currency and sees a lot of legitimate use no doubt _because_ lions are social and therefore referring to a group of them makes sense to do.
There's a murder of crows-- common bird, fancy collective noun. I have no idea whether the term was ever in common use, but it's handy for mystery writers.
Hey Scott! I was playing around with DALL-e and got it to make a pretty good stained glass image of Darwin and the finch. While I somehow think you don't really want your house to be dominated by rationalist stained glass windows, you might want to have a look. It's here: https://photos.app.goo.gl/2XNbUJxh84hbFgYz5
You should know your name appears in the photos link. If you're uncomfortable with that you should consider posting it to imgur or another free image hosting site.
Yeah, but even average me came out kinda top - Maybe I was early? -If dices are not your thing, it can be frustrating. - I found much more irritating that they abuse that "test" to advertise EA and push their book. Felt like an encounter with Jehovah's witnesses.
> I found much more irritating that they abuse that "test" to advertise EA and push their book. Felt like an encounter with Jehovah's witnesses.
Yes, I complained about that in my disappeared comment. "Would you be interested in our newsletter? How about reading our book?" is not a personality inventory. It's advertising.
Well, I marked "libertarian" - as among that looooooong list there was no option "irrationalist" (Tyler Cowen). - They asked a battery of questions that were - obvious to AXC/SSC readers -arguments to join EA. Then they asked: Do you know a philosophy/movement with such ideas as suggested? What is the name? I typed: "EA (as utilitarianism is no movement)". Then there was a bit more about EA, this time explicitly, and then a large book cover popped up, no link to to buy, though; I remember it as "Doing good better" by William MacAskill - it was blue, I recall. :) So, yeah, "EA going YW" was my thought. - Btw: I only had two kind of"IQ-tasks" the spatial ones with folding dice and those verbal: 4 words, find the odd one. (I am ol'n'confused, much dumber than the results suggested; I might mix up some details of this EA-PR, but that was how the experience went for me.)
Oh man, given the current Sam Bankman-Fried imbroglio, maybe cooling it on the William MacAskill Will Change Your Life bit might be a good idea 😂
"It was his fellow Thetans who introduced SBF to EA and then to MacAskill, who was, at that point, still virtually unknown. MacAskill was visiting MIT in search of volunteers willing to sign on to his earn-to-give program. At a café table in Cambridge, Massachusetts, MacAskill laid out his idea as if it were a business plan: a strategic investment with a return measured in human lives. The opportunity was big, MacAskill argued, because, in the developing world, life was still unconscionably cheap. Just do the math: At $2,000 per life, a million dollars could save 500 people, a billion could save half a million, and, by extension, a trillion could theoretically save half a billion humans from a miserable death.
MacAskill couldn’t have hoped for a better recruit. Not only was SBF raised in the Bay Area as a utilitarian, but he’d already been inspired by Peter Singer to take moral action. During his freshman year, SBF went vegan and organized a campaign against factory farming. As a junior, he was wondering what to do with his life. And MacAskill—Singer’s philosophical heir—had the answer: The best way for him to maximize good in the world would be to maximize his wealth.
SBF listened, nodding, as MacAskill made his pitch. The earn-to-give logic was airtight. It was, SBF realized, applied utilitarianism. Knowing what he had to do, SBF simply said, “Yep. That makes sense.” But, right there, between a bright yellow sunshade and the crumb-strewn red-brick floor, SBF’s purpose in life was set: He was going to get filthy rich, for charity’s sake. All the rest was merely execution risk.
His course established, MacAskill gave SBF one last navigational nudge to set him on his way, suggesting that SBF get an internship at Jane Street that summer."
18 hours after you commented (which is when I am commenting), the average cube rotation was around 0.2, the average hidden pattern was around 0.4, synonyms around 0.42, treasure 0.42, and made-up words was around 0.7.
I'm pretty shocked at the low average for cube rotation, and the high average for made-up words. I would have guessed the average performance to be 50% across the board.
Cube rotation was 0.24 - words:odd-one-out was 0.62 (even there I "scored higher than 79.07% of the people who took this test." Which gives me a warm glowin' feelin'. Kurt Vonnegut described once how aliens take over earth by demoralizing humans with "fake facts of high averages" that make most feel like miserable failures. I suspect a sales-trick when I get yummy scores.)
The way things have being going in the Reality Department, I expect Musk to lose the poll, but then declare the entire election process corrupt, and override it.
He'll find people stuffing ballots into mailboxes in Arizona, and plead with the technicians that all he needs is 11, 000 votes. Mobs will rush the Twitter Tower, and Musk will declare himself Emperor of Twitter, Thank You Very Much. That old CEO business was highly overrated, anyway, he'll say.
You're so out of touch with what you're talking about that you missed the part where the poll had already been long closed by the time you posted this nfoolishness.
"Free speech just means the government can't arrest you; if you're...banned from an internet community, your free speech rights aren't being violated. It's just that the people listening think you're an asshole, and they're showing you the door".
Take note, Thursday Night Massacred!
So Elon can do what he likes with his own private company, including running polls to troll the users and then declaring himself Emperor. Or did you mean it only applied when done to your *enemies*, it wasn't supposed to happen to *you*?
Nearly everyone that I follow both acknowledges that Elon is 100% allowed to do all of the random bullshit he is doing. What he's being called out on is in his repeated claims that he's actually not doing the same thing he's accused everyone else of doing and saying it's the worst thing in the world for (ie content moderation), except in a less consistent/predictable manner.
I don't think, for example, that you could find any significant number of people seriously claiming that, say, Parler is "violating free speech" if they banned whatever random leftists showed up to pick fights with their users and call everyone a racist sexist douchebag. They might point out that doing so is fundamentally the same action as banning someone for saying something racist or sexist on Twitter, so they're clearly no *better* at what their marketing claims is "free speech". But, you're not going to find many (if any) people suing Parler for getting banned and claiming their 1A rights are being violated, any more than you would for getting banned from editing Wikipedia.
People are definitely annoyed that Elon both has enough money to simply "buy a platform" in the first place, as well as annoyed that he, y'know, actually did so and then did his best to recreate the entire culture of the platform to better suit his ego. Broadly speaking, most people don't think what he did was *illegal*, just, y'know, a bad thing to do. It's like the old school forum days - someone new becomes a mod, starts changing everything that's already been in place, and you're likely going to find the old regulars aren't happy. He's a bad dude, not a criminal. (Probably. He might also be a criminal for some of the workplace changes that he implemented, but almost certainly not for his policy changes, at least not within the American legal system. EU may likely be different, I know very little about their laws.)
Do you think that people suspended or banned under the previous Twitter leadership were "assholes" who were "being shown the door" as per the linked comic? Does that equally apply or not apply to people being banned or suspended now? Why or why not?
I think Musk is beclowning himself on a daily basis, and it's not good for him to do so. I also think he has a lot of support and his base reasoning for buying Twitter was the arrogant "when you are banned from Twitter it's because you're an asshole, learn to read the room" response that was clearly not an unbiased and accurate retelling of the situation. Is Musk doing something uniquely wrong by reversing course and instituting bans on his ideological opponents, or is he going tit-for-tat on previous defectors? Is it worth admitting that the previous rules were in fact a defection on norms (and often directly against the written terms of service) that were inappropriate? Should a person have to spend $44 billion in order to make that point in a way that others take notice?
I'm not a fan of Musk and his antics, but I'm unable to find a good alternative to this situation - other than to shut down Twitter and all social media, which while I'm in favor of that, is both illegal to impose and very unlikely to happen.
I'm a (certain type of) data scientist, and a boss of mine asked me about how to think about ChatGPT. In order to organize my thoughts on it, I wrote a post about it. I'd be really interested to see what folks around here think of it: https://ipsherman.substack.com/p/an-opinion-about-ai-chatgpt-and-more . P.S. I don't spend a lot of time in these Open Threads, so I apologize if self-promotion here is gauche. Please let me know if it is, and if there's a better way to solicit feedback!
> I’ve been hostile to “AI” as a term for a long time. Probably more hostile than I should be, but hostile I have been. I've changed my mind somewhat in the past several months, for reasons we'll get to, but I'll try not to bury the lede. Here’s my current opinion on ChatGPT in particular, the AI iteration that everybody's talking about as I write this:
>> It’s the first “modern AI” that I have personally found useful, but even then, I only find it limitedly useful. I use it (when its servers aren’t overloaded) as a very specific type of search engine, for the types of questions I would otherwise Google, namely esoteric, fact-based questions like “What’s the best way to dedupe a list in Python?” or “What’s the origin of the phrase ‘For all intensive purposes’?”
I don't like when explanatory essays give a bunch of caveats to a topic that hasn't been revealed. because i have to hold all the caveats in my brains while i wait for the lede to drop. If I were you, I'd reorganize this as:
> I have mixed opinions on GPT. It’s the first “modern AI” that I have personally found useful, but even then, I only find it limitedly useful. I use it (when its servers aren’t overloaded) as a very specific type of search engine, for the types of questions I would otherwise Google, namely esoteric, fact-based questions like “What’s the best way to dedupe a list in Python?” or “What’s the origin of the phrase ‘For all intensive purposes’?" But to fully grasp the ramifications on the field of AI and society, there's a lot of context to explain.
> AI - WTF Is It?
> I’ve been hostile to “AI” as a term for a long time. Probably more hostile than I should be, but hostile I have been. I've changed my mind somewhat in the past several months. One reason is that it just doesn’t really have a definition. Or maybe its definition is so broad as to not mean anything.
----
> Sure, machine learning is called machine “learning,” but what it’s doing isn’t “learning” so much as “systematically updating parameters in increasingly complicated equations to minimize loss functions.”
This still looks kinda abstract. Borderline eulering. I know it's not your intent, but your boss might not know any better. I'd change it to
> ... isn't "learning" so much as just doing linear algebra, which is a mathematician's equivalent of using a spreadsheet.
and then maybe take another paragraph to explain that ML takes pixels, characters, etc and automatically loads them into something that resembles a spreadsheet. I think this more concretely relates ML to something your boss will understand, who will therefore conclude "if ML is just a fancy version of MS Excel, it's probably not conscious".
Great feedback, I really appreciate it! The order in the beginning is a bit of a historical accident of where the idea for the post came from, and your reordering definitely does make it easier to read.
And the eulering point is a good one too. If I'm honest, I never felt that comfortable about how I phrased that part. I just couldn't think of a way to elegantly make the point that the word "learning" in "machine learning" is really different than what most people mean when they think of "learning," and that's what I ended up with. I like your phrasing (especially comparing to a spreadsheet) a lot better.
Glad you liked the essay. Thanks for reading and commenting!
don't worry, i immediately recognized what was happening in both scenarios. i've been there. I.e. at the beginning of your writing, you were trying to explain a fuzzy perspective, but you don't know where to start so you just ramble a bit until you hit your stride. And then you weren't sure how to best summarize what ML actually was doing, because you're so used to being deep in the weeds.
Random thought: I wonder if you could use the same principle of noise cancellation to create active anti-sonar tech - basically, create a perfectly out-of-phase pulse of the same amplitude as the inbound ping, thus cancelling out your echo. This would obviously be very useful for military submarines / USVs, so I imagine either it's already being investigated or there's a technical reason it wouldn't work.
Not an expert, but I doubt it would be possible to perfectly cancel the noise across all frequencies and directions. Real life "noise cancelling" headphones certainly don't.
I know modern radar systems rapidly switch frequencies (even in a single pulse!) to bypass jamming and avoid detection, so I would assume sonars do the same.
> Conventional wisdom is that intelligence-related studies replicate better than other fields, and Clearer Thinking is testing that now by trying to replicate 40 intelligence-related claims. They’re looking for experimental subjects to take their online tests; click here if you want to help.
Heads up, but I think the second example question in the "minimum average distance to red dots" section is incorrect. The test claims that the correct answer is to choose the middle, which has a total distance of 3sqrt2. However, if you choose one of the sides instead, the total distance is only 2+sqrt5, which is slightly lower.
Hmm there must be a number of different tests. I got nothing with red dot distance, nor image folding. I did fairly well at remembering flags, till my brain got a little flag fogged in the end.
* In addition to the incorrect example question for centroids, one of the paper folding questions is impossible as well (near the middle, I forget which one). There's no way to fold the first image into the second unless you flip it over first.
* Why do we need to click a link to explicitly pause the timer between sections? It should be paused automatically when you're between questions or sections!
* SAT question enforces max of 1600 rather than 2400, regardless of when you took the test. Also not clear how to answer if you took the SAT multiple times
* no option to list a double major in college
* might be nice to have examples of what is urban, suburban, highly urban, etc. for comparison
* No way to go back in the demographics section if you mess up. Only option is to "reset everything", which is rather scary if you just spent an hour doing the previous tests and don't want to lose progress
> SAT question enforces max of 1600 rather than 2400, regardless of when you took the test
I haven't taken the online assessment that Scott linked to, but here's a general trick for comparing 2400-SAT scores with 1600-SAT scores (without using percentiles). Take the scores from the two non-math sections of the 2400-SAT. Average them. Then, add that average to the math section score, for a total out of 1600.
If you want to get non-garbage results, you need to be explicit about that, or else everyone will just guess about what they're supposed to do (in my case, I just used the math and verbal section and ignored the writing section). Or better yet, fix the question so people don't have to follow confusing workarounds in the first place.
It feels to me like it's going to be much easier to write AIs that can give 80% correct solutions 80% of the time to "hard" problems than to write ones that can reliably and consistently solve "easy" problems (definitions of "hard" and "easy" intentionally not gone into).
So what are some applications for which a quite-likely-to-be-correct output that you can't rely on is valuable?
And what are good ways of using plausible-but-unreliable applications plus secondary testing to get reliable applications?
Already started with legal contract review , corporates are becoming happy to take a risk on a low value deal (under $1m, not strategically important etc) if red flags (high risk clauses) are not present when reviewed automatically.
Today high volume deals may not even be getting a real legal review , so 80% is better than 0%. Expect consumers to be even more keen to get a basic review due to the costs of law firms.
I have three more subscriptions to Razib Khan's Unsupervised Learning to give away. Reply with your email address, or email me at the address specified in my about page:
"As soon as new psychology and behavior papers come out in Nature and Science (the two most prestigious general science journals), ..."
I don't think scientists should treat Nature and Science as the two most-prestigious general science journals anymore. Both have written editorials and/or letters to the effect that science must be politicized, and articles should be judged on the desirability of their expected impact on society, in addition to their scientific merits.
On the one hand, it's true that they've succumbed completely to Lysenkoism; on the other hand, the massive political skew of academics means they retain their old prestige anyway. In fact they probably did it to retain their prestige. And doesn't that just make it that much more urgent to test their articles for replicability?
I think what you call Lysenkoism is almost entirely in the humanities. Once I went through all of the professors at a university who had signed a particular petition asking for the reinstatement of a professor who was fired for having students physically eject a journalist who'd come to report on their protest. (I don't understand the logic of staging a protest and then attacking reporters who try to cover it, but whatever.) 100% of the English department signed it, and a pretty high fraction of all humanities profs (but less than half); IIRC about 3% of profs in the hard sciences did.
Seems to me I saw some other survey with similar results: something like 40% woke in the humanities, and around 3% in the sciences. Maybe. My memory is terrible. Also, it depends on whether sociology and psychology are counted as sciences or humanities.
No, its happening in the sciences too, though it's not as far gone as in the humanities. Also 3%/40% is laughably low. 20%/80% might be credible. Source: my lived experience as an academic scientist.
They're still the most prestigious, even if they aren't the most reliable or rigorous. Even apart from politics, the desire for flashy surprising results means that their choose articles that are more likely to be wrong than usual.
As we can see, there is an entire niche here for AI chatbots to exploit when it comes to spam and scam messaging.
Yeah I am 90% (more?) sure I got COVID at a big state high school sporting tournament where ~100,000 people attend over a few days the first week of MAR where at the time supposedly there was just 1 or 3 or 5 or something people with COVID in the entire state (and they were all Chinese nationals or recent visitors).
Not sure there was a "cover-up". But it definitely went here from "only worry if you have direct contact with people who were recently in China", to "OMG it is everywhere", in about 3 days +/-. Which makes me think they were not really giving accurate descriptions to the press at the time.
They did not gracefully transition from "be normal" to "OMG panic" as the evidence was likely suggesting, and instead clung to normalcy as long as possible until it was untenable.
For what it’s worth, I remember lots of people saying exactly this about getting COVID prior to ~March 2020 and then when the antibody tests became available I saw a lot of posts says “I guess I actually never had COVID”. I can’t put the numbers to that that you would need to make any sense of it, but I’m pretty skeptical of any specific claims. It’s pretty hard to distinguish COVID from other infections since the range of symptoms across individuals is pretty big (I never had smell loss when I had it and that’s the most identifying symptom). That said, there were probably early undetected outbreaks somewhere and your anecdote is only slightly earlier than official numbers and we all know testing wasn’t being done.
I know some people who strongly believe they got COVID November of 2019, which doesn't seem possible unless there was a massive coordinated coverup between both Chinese and Western medicine to hide that it spread earlier than thought. Certainly by now we would have some people admit that it spread earlier than expected. The most recent that I heard, there was some potential updates to how early it was breaking out in China, to November, but that there was no way it was spreading widely anywhere else - especially random continental US.
To figure out whether a local outbreak of COVID-19 happened before health authorities said a Pandemic was in the air, maybe you should look at these factors:
A. Is this an area with any amount of international travel incoming, or outgoing?
B. When local officials began officially noticing COVID cases, how large was the first surge of known-cases?
I was trying to figure out whether a large surge of cases or a small surge of cases, post-Declaration-of-Pandemic, would indicate wide circulation pre-Declaration-of-Pandemic. It actually depends on the timing: if the Declaration-of-Pandemic happened midway through the first surge of cases, then the case-count might go from single-digit to many-thousands within a week or so. On the other hand, if the first surge of cases came and went pre-Declaration-of-Pandemic, then the case-count after the Declaration-of-Pandemic would look like a splutter.
This might also be confounded by the age-curve of the first wave. Though it was not fully appreciated at the time of Declaration-of-Pandemic, sufferers of COVID in the age bracket of elderly-retirees had much higher death rates than sufferers of COVID who were not in that age bracket. Thus, an area with a large number of international students might see lots of cases of COVID, and very few deaths-due-to-COVID.
Do you think that COVID was introduced in your area sometime before December 2019, or during January 2020?
Was there an increase in rates-of-death in nursing homes before the official Declaration-of-Pandemic?
Eric Lortie writes:
"The initial batch of cases were tiny and we had low case counts throughout the entire pandemic. It was often commented on how the East Coast of Canada was a pocket of safety, when in actuality they failed completely to protect us and didn't even notice. And then when they finally did notice they didn't tell us."
I am also in Halifax, Nova Scotia, and it was my personal experience, and that of my social circle, that Eastern Canada was a significantly safer place for Covid than the rest of Canada, and even more so when compared to the US.
This was obvious when looking at the official statistics, and also anecdotally...I didn't know anyone who got it...no one...until the Omicron wave. (We've probably started to revert to the mean in the last year or so, but I have not been tracking the data as closely recently)
So I would be really interested to hear what evidence you have for these accusations of a cover up.
I'd be interested in seeing these company sick records, if you can point me to them.
In the absence of any hard data, this all sounds like wild speculation to me.
Two questions about this:
- is there excess mortality data? The US has this: https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm. It shows excess mortality jumping in late March 2020 as you'd expect, and later waves roughly when people thought there were waves. I am very willing to believe there were isolated outbreaks of COVID in the West before March 2020, but also would want more evidence than anecdotal.
- what exactly is the "anti-mandate" movement in the context of the present day in Western countries? Does it mean against vaccine mandates specifically, or COVID rules generally? For me and people I know COVID is basically over, except for a few people who continue to be worried, and structural changes that I think persist for non-COVID reasons (e.g. work from home, outdoor dining). But of course I'm not everywhere. It's not clear to me there are a lot of mandates in the US at least right now, though I got my vaccine the first day I could so I never had to worry about it and never paid that close attention. Not sure the situation in Canada.
What’s the motive for a coverup supposed to be again?
Anecdotally: My wife came back from a trip to San Francisco in late January with what seemed to be a case of the flu. She was pretty miserable for about a week. We both suspect she had covid then. It was officially not in the US at that point, largely because nobody was allowed to do their own covid test--they had to use the broken CDC-provided one for bureaucratic turf reasons.
And there's always a fuckton of people who actually have the (ordinary, miserable) flu in any January. Absent testing, there's no real way to know. And thanks to the CDC, we didn't have the tests when we needed them.
You might find this interesting: https://www.youtube.com/watch?v=XzPophi-9MY
Apparently covid antibodies were found in Milan in stored blood samples dating from September 2019
Elon Musk is running a Twitter poll - what else? - on whether he should step down as head of Twitter.
If you feel like weighing in, you know where to find this. 2 hours left.
It's a shame that it looks like he will step down rather than go full Caligula with his imperial decrees.
The question is: who would you want to replace him?
Picking only from the people I think Elon would plausible pick (long time superfans, essentially) Jason Calacanis is probably not too terrible. I sometimes watch his 'This Week in Startups' show and he's won me over a bit in the longer format.
Maybe somebody *could* have fixed Twitter before the acquisition, but is it even possible now with the new debt?
>In its last quarterly filing as a public company, for the period ended June 30, revenue was $1.18 billion, down slightly year-over-year.
>Twitter paid $23.3 million in interest expense in the quarter ended June 30, according to a filing.
>Now, the company will have to pay at least $9 billion in interest to banks and hedge funds over the next seven to eight years, when the $13 billion in debt matures, according to a review of Twitter’s loans by Mr. Davies, the former credit analyst.
https://www.wsj.com/articles/how-elon-musks-twitter-faces-mountain-of-debt-falling-revenue-and-surging-costs-11669042132
It appears they may be down about 5,000 employees from when Musk took over. Assuming salary and benefits per employee of about $150,000 per year (which I think is conservative), that's $750,000,000 a year - no small change! We don't know if Twitter's income has changed or if there are other expenses that changed (higher or lower), so it's not readily possible for us to estimate their ability to pay off the loans, but Musk has made some really big steps towards it.
There was some talk of socialising it , as a public space. This would be a very unusual thing for the US to do and then adds another problem - can anybody be ever banned again?
> There was some talk of socialising it , as a public space. This would be a very unusual thing for the US to do and then adds another problem - can anybody be ever banned again?
Sure, pedophiles. If you can ban them from speaking in private spaces, you can also ban them from speaking in public ones.
The logical next step is to streamline the process of getting someone a child pornography conviction when you want to curtail their speech.
Who was talking about public ownership of Twitter?
Anyway, in theory, if the US government bought Twitter, couldn't Congress then donate it to a foundation which is set up to manage it in perpetuity? Couldn't that foundation then ban users, and so on, without any First Amendment concerns?
Those loans are prior to the takeover? Seems like it was a clown show to me.
No, these loans are how they paid for the takeover. Leveraged buyout.
https://www.washingtonpost.com/business/elon-musks-twitter-deal-is-different-than-most-lbos-heres-how/2022/11/12/7c671b64-62b9-11ed-a131-e900e4a6336b_story.html
That doesn’t look rosy.
The $13 billion shouldn't be an issue if they make their payments for those 7-8 years, someone will (barring drastic interest rate changes) float them another $13 billion if they have managed those payments for 7+ years. Will they be able to make those payments? That probably depends on the structure of the debt and when those payments have to be made.
Also using quarterly data instead of annual data for revenue and then using a large lump sum for interest is disingenuous and makes it easy to misunderstand the situation.
Apologies, I just googled for the first news story with the loan numbers. Here's a good representation of the Twitter debt situation:
https://twitter.com/EnergyCredit1/status/1591873061149614106
and some corrections on the modeling, the unsecured cap was 11.75%
https://twitter.com/EnergyCredit1/status/1594802631679025153
Any predictions on what Elon will do after the poll closes?
Result: No, Elon should remain in charge of Twitter.
Elon changes his profile to "The People's CEO". He is spotted wearing a tshirt with that on it in the next week. Two months from now Elon says he needs to focus on Tesla and SpaceX. Elon appoints one of the hosts of the 'All-In Podcast' as CEO, most likely David Sacks or Jason Calacanis.
Result: Yes, Elon should step down.
Elon announces he will step away in two months and appoints one of the hosts of the 'All-In Podcast' as CEO, most likely David Sacks or Jason Calacanis.
Long Shot: Elon sells a bigger chunk of Twitter to the Saudis and actually really steps away and let's someone else deal with it. They choose some random CEO the new investors approve of. https://twitter.com/LauraLoomer/status/1604540593354735616
Edit for predications update. Completely wrong. Elon is going with the deep state rigged the vote, so he didn't lose in the first place. https://twitter.com/elonmusk/status/1604985324505030658
The poll closed this morning with 17M votes cast and it went 57-43 in favor of him stepping down. No response yet from Musk.
Tesla shares, which had dropped 11% recently as Musk sold shares, rose 5% this morning after the above Twitter poll closed.
Last night Snoop Dogg posted a poll "Should I run Twitter?" As of this morning it had 1.3 votes, split 80-20 "yes".
Calacanis posted a poll asking who should become Twitter’s next CEO: himself, Sacks, or Calacanis and Sacks together as co-CEOs. That poll's "other" option won with 39% of the 14,000 votes cast.
Predictions update? Starting to think I completely misread the situation.
I though Elon's poll was part of a plan that works no matter the result. But the poll closed 8+ hours ago and Elon hasn't said a thing. Maybe he's just hungover or whatever, but it's not feeling as much like a plan as I thought. Really no followup ready to go upon the poll ending?
Elon finally reacts. He's going with the deep state botted the vote. Looks like the plan wasn't to be voted out as an exit strategy? Seem like they might even redo the entire vote and only allow Twitter Blue users.
https://twitter.com/elonmusk/status/1604981780548767744
My prediction is that Elon will find a new project. It will be techy and ambitious, but not obviously related to anything else he's doing.
I'm curious, are there any other people who started out very pro-Musk and ended up anti? Where was your tipping point? Mine was, I think, when he banned the account that posted his jet's location, though my confidence had been eroding for a while before.
I don't know about very pro, but definitely a positive valence, which is now negative (more annoyed than disgusted or morally appalled). It's very odd (and a bit disheartening) to see his somewhat reasonable side publicly clash with his anti-left edgelord side (well, maybe not edgelord, but facebook-tier boomer meme, at least). Seems like buying Twitter was a bad impulse that he now can't escape from.
I think he has an unhealthy interaction with Twitter, and that involving himself deeply in it was a mistake for that reason. But I also don't require Musk to be a god or a demon, just a pretty smart and industrious dude who has accomplished a lot and will hopefully keep on accomplishing a lot in the future.
He's addicted to Twitter just any common social media flunkey addict. If his skill is "entrepreneurial engineering," then all the the time he spends on his silly tweets has got to distract him from his Real Productivity.
And it shows a weakness in our society that we're vulnerable to being so dominated by the autistic and eccentric whims of billionaires.
Why would someone be "pro" or "anti" Musk? He is just a guy, and as a standard bearer for an ideology or viewpoint he doesn't work very well because he is all over the place.
He is pretty clearly a genius businessman/investor, but perhaps a bit off his rocker these days. Though people were also saying that same thing 10+ years ago, and he is now one of the wealthiest people in human history so...
Blue tribe has rallied around hating him because they are tribal morons, but I assume that doesn't include you.
Anyway, like everyone else I think his recent shenanigans and Twitter acquisition are likely big mistakes and will fail, but I trust his judgement more than mine in such matters. Plus Twitter is blight on society so if some super rich guy wastes a bunch of money and Twitter gets destroyed that sounds like a win-win?
Phrasing is hard, especially having just gotten off a transcontinental flight. What I meant by "pro- and anti-Musk" is simply "in favor of, or against, him running Twitter". I'm a free-speech nutcase by today's standards (definitely not blue tribe), and I thought he was too, but now I'm thinking he's in favor of "free speech for people who agree with me" and am sad.
What I'll say about Musk is that he's intemperate. Which can be harnessed to good use when you're trying to lead and direct a team towards huge, almost preposterous achievements, like technological breakthroughs or military conquests. But it's a terribly counterproductive personality trait when you're trying to promote and referee a place for broad, thoughtful public discussions.
I think he sincerely wants Twitter to be the broadest possible place for thoughtful online discussion, but he just can't get out of his own way. By contrast, I don't think the previous regime truly wanted Twitter to be that place, but it certainly was a lot more temperate in how it went about it.
This is a wonderful description.
Even that might be better than having all big social media and traditional media companies reading off exactly the same page w.r.t. moderation, including stuff like kicking Trump off, blocking right-wing social media networks, blocking covid disinformation, etc. Having some stuff become unpublishable/invisible in US media and social media is dangerous on all kinds of levels.
The lockstep of traditional and social media post-2020 seems very similar to the lockstep of traditional media post-9/11 attacks, and seems equally unhealthy. Making sure nobody hears any debate or dissenting views about major policies (invading Afghanistan, invading Iraq) is a great way to help the country walk off a cliff, and indeed, that's pretty much what happened.
OK, that changes the meaning dramatically. I've always been pro-most-of-what-Elon-Musk-does, and admired his ability to get things done. That hasn't changed very much. I've always been strongly anti-Musk-running-Twitter, because that really isn't well suited to his skillset or his temperament and it is going to severely handicap his ability to do other things. Things that he's actually good at and that I want to see done well.
I'd also like to see Twitter either destroyed or radically transformed, but anyone but Elon Musk would be a better choice for that. Who's the #2 investor in that cabal? Mohammed bin Salman? Yeah, OK, let him run it - the mutual destruction of MBS and Twitter would be an absolute net positive.
I think banning the jet was an instance where a button got pushed with him. His kid was in danger, the mother of the kid was probably there, my diagnosis of that is that the mother of the child has an “Elon obey me” button that still works even though they’re divorced - this is not that uncommon. He probably got an awful phone call in which it was demanded that he Do Something, so he did. I think it did not have much to do with his thoughts or ideas on free speech, probably bypassed all of it. Is it a little upsetting that there’s an Obey button? Kind of, but I don’t think it comes up that often. Probably only a few people have one on him. The question becomes who has them & when are they pushed. Leading Twitter does mean he’ll be surrounded by people all trying to get to the buttons. I think people are trying to use him here and the sooner he leaves the better for him and Tesla. I see him as someone who gets things done where others don’t - but projects and goals. Not evolving balances of political rhetoric. Knowing that Twitter has been used as a tool of governments is helpful. Making that stop will be very hard and boring for him, plus he’ll have to admit the power brokers don’t like or respect him, there’s no mutual engineer esteem, it’s manipulation.
I think he should appoint a thirteen-person council and bail ASAP. I still respect him.
Yeah, from what I understand, this jet-tracking account was going for years, but then some rando went after Musk's kid so Dad booted it all off Twitter. Understandable reaction, whatever your opinion on doxxing or jet tracking or whatever.
I thought the Thursday Night Massacre was much more fun (exactly how full of yourself do you have to be, to invent a grandiose title for "we got suspended for two days"?)
I mean, to be fair, the name was probably coined before it was known whether the bans would be indefinite, like the case of the people who made Musk parody accounts (and only Musk parody accounts, I don't recall hearing about a single other impersonator account being banned) after Musk said comedy was legal on twitter.
It is still rather self-important, though; how many people have been banned permanently from Twitter or other social media sites? The guy I saw lamenting about it opened up with "I and eight other journalists" - stop the world, nine journalists whose names possibly nobody would recognise have (one of) their social media account suspended! Truly Democracy Dies In Darkness!
Meanwhile, if Mr Thursday Night was fired by his employer and lost access to the Twitter account that goes with the job - would we know or care? Would journalism grind to a halt? Does anyone, apart from journalists, really follow what someone says on Twitter as their main/only source of information?
I suppose some people do, but if Twitter imploded in the morning, some other site would take over from it. Everyone would go to Mastodon or somewhere. Or a new gossip website would spring up, as TikTok took over from Vine.
> this jet-tracking account was going for years, but then some rando went after Musk's kid so Dad booted it all off Twitter. Understandable reaction, whatever your opinion on doxxing or jet tracking or whatever.
Reminds me of the how a certain ex-president's account was going for years until one day a bunch of randos went after congress and the VP at his urging.
I'm a father myself, twice over, and agree about that being an understandable reaction....but it doesn't seem clear that the incident, whatever it was, actually had anything to do with the @ElonJet Twitter account. From the Washington Post:
"A confrontation between a member of Elon Musk’s security team and an alleged stalker that Musk blamed on a Twitter account that tracked his jet took place at a gas station 26 miles from Los Angeles International Airport and 23 hours after the @ElonJet account had last located the jet’s whereabouts....Police have said little about the incident but say they’ve yet to find a link between the confrontation and the jet-tracking account....
The incident occurred in South Pasadena, a Los Angeles suburb, on Tuesday at about 9:45 p.m. South Pasadena police were called to the gas station, according to the business’s manager, but made no arrests. South Pasadena police have not responded to requests for comment.
The Los Angeles Police Department said in a statement Thursday that its Threat Management Unit was in contact with Musk’s representatives and security team but that no crime reports had been filed...."
I'm not even "anti" but I still voted for him to step down. The (now reversed) policy change banning advertising accounts on other platforms was completely out of touch, and the whole rationalization of his actions at the helm of the platform went down for me. I still agree with like 90%-95% of the changes, but I have to wonder if it was mostly luck, or if I am still rationalizing away bad choices.
If anything, I totally understand and support the banning of the jet account / new doxxing policy, you cannot create an actual marketplace of ideas if it's too easy to go back down to ad-hominems/meatspace threats.
It's actually the reverse for me, I started very anti-Musk in about 2015-2016, less "angry" per se and more annoyed, utterly exasperated and eye-roll-strained with the incessant "oMg guys he's jUsT like ToNy StaRK" screams, which completely misses the 2 points of
(1) Tony Stark started as an immature and immoral character stuck at 16 years old mental age, he gradually knew better and grew out of the "I'm amazing I can do anything I want" mentality by bitter experience. Iron Man is about how Tony Stark gradually learnt to not be Tony Stark.
(2) Iron Man is a hilarious/infuriating misdepiction of how technology and engineering work, at some point in the first movie Tony is like "oh yeah ? the suit freezes at high altitudes ? here's an alloy I just pulled out of my ass*, start coating the armor with it" just fucking no, this is not how any of this works. Real science and Real technology, like Real mathematics and Real programming and Real every knowledge field that matters, long ago stopped fitting inside a single person's head, no matter who that person is. This seems to bleed into Elon Musk fanboyism, where some fan boys think he's personally writing every line of code and welding every piece of metal.
I started softening on him a little bit when he began toying with Blue Tribe and flirting with all of its traditional taboos, I'm a spiteful guy and sometimes the fastest route to my heart is to be the enemy of my enemy. It was, I admit, highly cathartic to watch the people who memorized "It's a private company it can do whatever it wants" by heart suddenly discover that Free Speech is a generally desirable value and that Corporations are no less bound by it than States are, even if our current, very flawed, laws don't always reflect that. I didn't want those people permanently booted off twitter, just banned and toyed with a couple of days to make them learn their lesson.
That said, it didn't last long. I don't remember when exactly did Elon outstay my tentative second chance, but it was before the Jet thing, probably when he reinvented shadow bans and then acted like this is a genius solution that is better than outright banning somehow. The Jet thing is a camel straw. His internet-celebrity-style polls aka "dear followers give me a crumb of attention" are also very see-through and cringe.
But regardless, it's always funny and good to destroy a social media giant, so Elon got all of us to laugh a good laugh *and* made the internet's idea pool better by an immeasurably tiny amount, all at the cost of 44$ slightly-fictional billions. I call that a good deal.
* : The dialogue tried to justify this by implying that his company used the alloy on a satellite before if I remember correctly, but this is no better. Stark industries make thousands upon thousands of weapons, gadgets and systems, so how did Stark manage to remember the one alloy that will solve his problem ? and how does the top company scientists that the villian father-figure later commissioned to solve the same problem didn't remember or discover the same thing ? and how does an alloy made to work on an unmanned satellite in orbit just magically worked on a manned in-atmosphere flight system ?
I started off disliking Musk, particularly the techno-worship around him. I don't think he's much of anything, except I have to give it to him that he's been very, very successful at business.
But all the fraught reactions (Thursday Night Massacre, you know, like Kristallnacht?) are making me very, *very* begrudgingly like him.
It's the same way J.K. Rowling has fully leaned into the TERF thing (which has become an accusation like 'racist'): oh, so you say I'm a TERF? Okay then, I'm a TERF!
https://twitter.com/jk_rowling/status/1604180531155017731
Slightly off topic, but I recently saw someone referring to a clearly anti-feminist conservative that said something anti-trans as a "TERF", and my reaction was "ah, we've reached the point where that term is starting to lose all semantic value, got it."
(For those following along at home, TERF stands for Trans-Exclusionary Radical *Feminist*. The last part is still relevant.)
You'd think the "radical" part would also automatically exclude conservatives.
Technically, "radical feminists" are those who believe that there are no differences between men and women... other than the obvious anatomical ones. (Which is why they logically have to reject the entire concept of "trans", because there can be no such thing as "female brain in male body" if there is no such thing as "female brain" in the first place.)
Of course, this should also exclude conservatives.
That trend started a very *very* long time ago, 5 or 7 years at least, where have you been all this time ?
As an anti-feminist, it's annoying as hell to me too. Feminists do not suddenly become ok when they oppose one particular wing of the wider ideology around them, if anything, TERFy feminists seem to double down on their misandry harder. Labelling them with the same label potential allies are labelled with is just more work for me to sift through.
The meaning has changed, and no longer reflects the word's etymology. That doesn't mean it's become meaningless. It now means transphobe.
Right, but my point is that its value as a distinct term was that it captured the concept of "virtue signals like a progressive but irrationally hates trans people". Without that, it's just another leftie slur to add to the pile of thought-terminating cliches.
Then why do people still say "TERFs and transphobes"? Are you suggesting it's just one of those emphatic pleonasms, like "rogues and vagabonds" or "far and away"?
Every term that denotes a philosophical position (doubly so for a term relevant to a currently-waged human rights struggle) becomes woolly and imprecise in its actual usage, but my perception is that the correct use of the term, referring to self-proclaimed feminists, still predominates. Maybe the process is beginning, as you say, but I haven't seen much of it.
I'm sure it depends on which echo chamber you're most familiar with. If you grew up reading magazines in your college's womyns resource center you'll use the original meaning. If you discovered the acronym on twitter, you'll never know it as meaning anything other than "irreedemable bad person who should be cancelled immediately."
TERF was originally an autonym, much like Social Justice Warrior. But it seems (and I'm sure someone can find counterexamples) that it's very effective to take an autonym and turn it into a smear.
I think Radfem hub is still active. And of course, the philosophy of "Feminism is a leftist class struggle political movement" goes back more than a century.
If you mean regarding the Twitter buyout, then no, I was opposed to Musk's plan from the start. (Though I also don't care about Twitter and think all the predictions about how the buyout will "change social media forever!" or "put democracy itself at risk!" are absurdly overblown. I also think Twitter itself is bad, has always been bad, and can't be made not-bad due to its very design, so on the off chance that Musk actually does cause Twitter to collapse entirely, I'll consider it a win, albeit one that was unintentional on his part.)
In a more general sense, though, I used to have a very positive opinion of Musk and his ventures, which has gradually gotten more negative over the years. The first time I started having doubts about him was during that fiasco with the trapped miners, when he accused one of the rescuers of being a pedophile, though I was give him the benefit of the doubt and write it off as a one-time oddity. Since then, various other things made my opinion of him lower considerably, though up until just a few months ago it was still more positive than negative.
I'd say the big turning point was when he started backing pro-Trump Republican candidates and parroting right-wing Culture War arguments. The Twitter buyout and the subsequent fiasco that followed were just the cherry on the cake.
Yeah, I suspect Twitter is very bad for Musk (and many other people)--it amplifies his worst qualities, gives him instant feedback from his fanboys cheering everything he says and does, and sort of builds a feedback-reward system in his brain for saying stuff that gets cheers. But Musk is actually a genius businessman. Training himself to follow what his fanboys want to hear him say is a terrible idea--he's smarter than almost all of them in the domains in which he has the power to act.
In my opinion, Elon Musk should keep doing random changes to Twitter, because that results in a more entertaining future.
(If as a consequence he loses money, or Twitter burns down, that's not my problem.)
I think that just because he is running this poll, it doesn't mean he is serious about stepping down, regardless of the results.
He seems to like messing with people on Twitter who get all high horse with him. "Do you want me to quit?" poll seems like a perfect opportunity to get all the preachy types who finger-wag about him being a baddie going, and then crushing their little hopes with a "ha ha, fooled you!" afterwards.
If that was the intention, than he made it unnecessarily difficult for himself by using the following exact phrasing: "Should I step down as head of Twitter? I will abide by the results of this poll."
Is that a legally binding offer, though? I might say "I will abide by the results - sorry, changed my mind about that, you're stuck with me".
It just means that, if he doesn't abide by the poll, he'll look bad. And looking bad seems to be his thing lately, so he probably won't abide by it.
Sure, he still can change his mind. Although he has run several polls lately, and this is the first one he explicitely added the second half. It *would* have been much easier, to say, oh, haha, changed my mind or even: sorry, something really important new has come up and changed my mind, without this second sentence. So I assume, he at least planned to abide by it, when posting this.
He could in effect also step down, and then sooo many people ask him to come back, or something else happens, and then he steps back in. :)
Final poll results:
Should I step down as head of Twitter? I will abide by the results of this poll.
Yes - 57.5%
No - 42.5%
Jeez, people are taking this seriously? In my imaginings, he just took another vape hit and achieved a cannabis level 7.2 perspective, a perspective from which running this poll looks like a deeply clever idea. In a while he'll slide down to level 5.8 or zoom up to 10.5, where instead he will think of a dick joke or a change to Twitter rules that looks awesomely awesome. And so on.
Well... if he'll no longer be the *"head"* of Twitter, what body part *will* he be?
*rimshot*
Heh, good one.
I'd say he'd be a part of the brain at the center of the cannabis-recepterXcolonic chiasma.
Does he not even know how to Sir Humphrey a poll?
Twitter's financial woes are largely--I don't know the percentage, but it's significant--due to being boycotted by companies who oppose Twitter becoming politically neutral.
Why is there no counter-movement by people who support Musk, to boycott the companies boycotting Twitter?
Generally speaking, the companies that oppose "Twitter becoming politically neutral" are those that don't want their ads to be seen alongside far right content.
People who are actually politically neutral (ie don't care about politics very much) seem to either fall in the camp of "I don't care what politics are being shown" (and thus, don't care that far right content is being allowed back onto the platform) or "I don't like seeing any type of politics" (and thus, only care about far right content being allowed back on the platform inasmuch as it increases the chances that they have to see anything political at all, and would likely mildly oppose Musk's changes, if anything).
Those who were previously banned and are now allowed back on the platform, if they were the type of people who would participate in a boycott (ie willing to forgo some level of personal convenience to send a message; I think this is wildly less common than simply saying you participate in boycotts and then not actually doing so, regardless of political leanings) are likely already doing so due to advertisers pulling out of actually right-leaning platforms (Fox News/Breitbart/etc), and so if you're already ideologically not buying an Audi or General Mills products, then you can't feasibly "boycott more" when they also pull out of advertising on Twitter.
I did not understand that.
Companies that oppose Twitter becoming politically neutral, and companies that oppose Twitter becoming unmoderated, or close enough as makes no difference. "Unmoderated" is a sort of political neutrality, but not the only sort. If I'm in the business of selling bicycles or baby food or whatever, I probably don't want my advertisements showing up next to a thousand variations of "I hate you [expletive deleted] [deleted expletives] and I want you all to die!".
Elon's comments and behavior on taking over Twitter would give a reasonable businessman cause to wonder whether or not all that fun stuff would be going unchecked on the grounds that "well, he only said he *wanted* all those people to die, he didn't say he was going to *kill* them, so it's not a death threat so Yay Free Speech!" The sensible plan would be to put your advertising budget elsewhere for a few months and see how it plays out.
Also, an unmoderated free speech zone probably shouldn't expect to support itself by advertising.
But that should have applied to Twitter for all these years already. There were plenty of people calling for death on Twitter who didn't lose their accounts because they were smiled on by the left, like Ayatollah Ali Khamenei, who used Twitter to renew the fatwah against Salman Rushdie, and whose account wasn't suspended after Salman Rushdie was consequently nearly killed; or like Taliban members and supporters.
In any case, it doesn't address what I meant, which was, Why don't individuals boycott the companies boycotting Twitter? There's been a devastating and persistent refusal of the victims of Twitter mobs to retaliate against Twitter mobs, which I would like people to recognize is unproductive. This is the same thing on a larger scale.
He definitely should. The best Elon Musk company - SpaceX - is the one where he got someone talented to do operational control of it for him while he focused on "big picture" design goals and promotional stuff for it.
About a year ago, archaeologists digging at a site called Sayburc in south east Turkey uncovered an 11,000 year old wall containing some carvings of animals and people:
It is pretty easy to search and find articles on this, along with images. But for the reader's convenience, here are a couple:
https://arkeonews.net/a-relief-of-a-man-holding-his-phallus-was-found-in-sayburc-one-of-the-tas-tepeler/
https://doi.org/10.15184/aqy.2022.125
Any guesses what these carvings were meant to convey? After so many thousands of years I would have thought there's next to no chance of being able to clearly relate them to any recorded myth or religion. But all the same, it's fun to try and guess.
For what it is worth, I think they might have been meant to represent rulership and justice, and the area where they were found was a sort of council chamber where the tribal leader and his senior counsellors or village elders met.
There are two apparent "frames". The first shows a standing man sholding his penis, and surrounded on either side by a pair of leopards facing him in a rather menacing manner. Perhaps the leopards were meant to represent his royalty, a symbol used down the ages for this, and the penis his fecundity. A king in primitive societies was expected to be fertile in his offspring, and to bestow fertility on the land, and if he failed in either, if for example a drought caused a crop failure, then the king was soon for the chop, literally!
Also, as one of the papers cited above points out, the figure between the leopards is wearing a two-braid necklace. Assuming that could be a crude form of regalia, this seems further evidence that the figure is supposed to be a chief.
The other frame, to the left of the leopards shows another man holding up what looks like a snake and facing a bull. I think in primitive societies a snake often signifies cunning and wisdom (c.f. the snake tempter in the Biblical creation story), and presumably a bull is meant to represent the opposite, i.e. bovine obstinate stupidity combined with mindless violence. So I suspect that frame is meant to represent justice contending with barbarism and primitive kneejerk responses.
We should remember that this site was one of the very first where communities had settled to start farming. So the inhabitants would have been fairly wild nomadic types and quite a handful for the ruler, and regulating them would have been fairly challenging, and thus a major if not the main preoccupation of whoever was running the place.
All that said, perhaps this symbolism is a bit advanced for those times, and the carvings just show some myth, long forgotten or transformed unrecognisably in the millennia since. For example, is it obvious the figure between the leopards is a man and not a monkey? He seems to have quite an ape-like face.
Hehe! There's plenty of stone age female "porn" though, some far older than 10,000 BC, those strange smooth looking carvings of ultra-fat women. FWIW, I reckon most of those, the smaller ones anyway, were used as dilators in female coming of age ceremonies. After all, child birth is pretty dangerous, especially that of a first born, and considering the young age girls those days often started bearing children.
+1
Look upon my schlong, ye mighty, and despair!
LOL
Those weren't *legs* of stone...
Maybe the guy carving or the guy commissioning the carving just liked leopards and snakes and bulls and thought they were cool.
As far as the guy holding his dick, guys like to hold their dicks. Maybe it was just art veritas!
Message: Don't beat off in the woods if there are tigers nearby. (Sorry I know you meant this seriously. But I really have no idea.)
The central figure reads to me as a sort of god, not a human ruler. After all, the human ruler is physically present, presumably very nearby, so you can just look at him with no need for carvings.
Notably, the "Bull" in the carving is either a recently domesticated cattle or a soon-to-be-domesticated wild auroch. This was carved around the time and place that humans first invented cows. I like to see the bull as an auroch, as wild and dangerous as the leopards (also present in the time and place, as far as I can tell)
So the mural depicts a local diety standing against the wild animals that would threaten a newly-sedentary society. This god has nothing to fear from leopards and aurochs. The phallus feels to me like a symbol of power. This god is virile and healthy.
I imagine whatever ceremonies or gatherings took place here were intended to encourage this diety to protect the village from wild animals.
Maybe whoever tamed the first aurochs was imagining this god when he did so, and gave thanks to him afterwards. Or maybe it already happened and the carvers believed that this god was responsible, and hoped he would do the same with the leopards
> Maybe whoever tamed the first aurochs was imagining this god when he did so
Domestication was probably gradual enough that people didn't realize they were doing things differently from their great-great-grandparents.
> Or maybe it already happened and the carvers believed that this god was responsible
I'll buy this one.
I thought a god, too, but this only because of the Egyptian creation myth which sometimes features a god masturbating as the act of creation.
Without context, who knows, but I do prefer more literal interpretations and so prefer to see it as a depiction of a story or purported event or act.
That all sounds pretty plausible, but in reading it I had a random thought: Do bulls, like some other animals, have an instinctive fear of snakes? If so then perhaps holding up a snake and wiggling it menacingly would have been a useful technique in controlling them: A furious wild two ton bull could be charging towards a guy, who would only have to hold up a dead cobra for the bull to screech to a halt in a cloud of dust and start hoofing it away in the opposite direction! :-)
Just a brief note: the snake and the bull are particularly associated with the Eteocretans or Minoans, pre-Greek inhabitants of Crete. Maybe some connection there; I think genetic sampling implies the Eteocretans came from Anatolia.
Snake and bull would also be Shiva/Rudra as Pashupati, "Lord of Animals" (the snake is associated in various ways with Shiva and the bull is Nandi, his mount):
https://en.wikipedia.org/wiki/Pashupati
There are a lot of "Lord of Animals/Mistress of Animals" motifs throughout Indo-European and indeed all mythologies. The famous seal from the Indus Valley is one:
https://en.wikipedia.org/wiki/Pashupati_seal
As is the Gundestrup Cauldron:
https://en.wikipedia.org/wiki/Gundestrup_cauldron
https://en.wikipedia.org/wiki/Gundestrup_cauldron#/media/File:Gundestrupkedlen-_00054_(cropped).jpg
Can I be the one to say that I'm not convinced that he's even holding his penis? He might be holding something else and archaeologists are just a bunch of perverts. The people of eleven thousand years ago would be scandalised to find out that they spent all that time on a beautiful carving of a man holding the sacred stone of whatever and we just think it's his dick.
Perhaps the object is more obviously penisy on closer inspection than it looks in the photos, though, so I guess I'll have to take their word for it.
"The people of eleven thousand years ago would be scandalised to find out that they spent all that time on a beautiful carving of a man holding the sacred stone of whatever and we just think it's his dick."
Probably not. Bronze Age petroglyphs from the same era are full of little stick guys with A, spears and B, humongous boners. Penis shame is a fairly recent thing culturally, and unlike most things blamed on Christianity, it can probably be blamed on Christianity -- before that ol' popery came along, Romans of both sexes wore little dick-shaped bell pendants on necklaces as good luck charms (look up "fascinus"), and one of the religious tasks of the Vestals was to care for the Phallus of the Roman People.
In other words, at worst the people of eleven thousand years ago would probably just think it was hilarious that we misinterpreted the <carefully carved object> as a dick.
What do you think of Attachment theory? (https://en.m.wikipedia.org/wiki/Attachment_theory). Briefly, that attachment patterns with others are formed in early life and could persist to adulthood.
Is it true, or just an artifact of attributing nature to nurture? Is it even scientific in the sense of having testable predictions? Are there good studies about it?
The anxious/avoidanct dynamic well describes my marriage, and many others; seems likely that there’s something in the theory
In practice, it seems like astrology for people who took an undergrad psychology course (that is, the version commonly used doesn't seem to have much attachment to reality beyond "sometimes people are like this).
Possibly there's a much more carefully hedged version that academics know that makes testable predictions with nontrivial accuracy, but if so none of the attachment theory proponents I've met have managed it give an example.
Not entirely disagreeing with you here but can't this be said of any scientific (especially in psychology) theory? Almost a motte-and-bailey where science lies in a nuanced and carefully articulated motte and the popsci version of it leaks out into a field of vague assertions.
A quick scan of the wikipedia article linked me to what seems like a pretty rigorous large longitudinal study ascertaining a relationship between the attachment style during childhood and clinical mental health outcomes later in life - https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3861901/
Thanks, that's interesting (and an actual specific claim!).
(Mildly skeptical of the result in this paper based on complexity penalty (the predictive factor there seems fairly specific and with p hacking potential), but haven't read the paper through and giving it props for making a specific claim that an attempted replication could disprove).
For the popsci version though - I think the popsci claim is "people have an attachment style that reflects how their young relationship with their mother was", but since attachment styles seem to be strictly ranked (avoidant is worse than non avoidant and insecure is worse than secure), it seems to just boil down to "if you had a bad relationship with your mother your relationships in the future will also be bad [in similar ways]", which seems false-ish (partly because people's bad relationship dynamics are generally not even consistent between their own relationships).
I think a distinction your summary misses is that attachment theory would only predict that if your relationship with your mother is bad in specific ways that this would have impacts on your future relationships.
So I kind of elided this point, but I think it's a weakness of the theory that its description of the ways in which that relationship would be bad is pretty vague and flexible.
Isn’t it the case that there are many phenomena that have pretty divergent results? An auto-immune disease, for example, can show up with any number of different symptoms and pathologies. I tend to think of attachment in a similar way — that poor attachment inhibits a person’s ability to trust future relationships, and that inability to trust manifests in lots of different relationship disfunctions.
What makes you think that its 'description of the ways in which that relationship would be bad is pretty vague and flexible'?
Hmm. Wasnt there an experiment where a monkey was brought up with in a cage with a mechanical food feeder, and a soft toy or something he could cuddle. Pavlovian theory would suggest he would attach to the food giver but he cuddled the toy.
Also scientifically you could test with children sent to cold institutions - are they less socialised and happy?
As I understand it, attachment theory's claim is that the monkey with the mechanical feeder would grow up to have bad relationships later in life. Open to seeing this experiment done but I'm going to guess it doesn't pan out.
Well they were definitely unhappy monkeys, if I recall.
Yes, here it is:
https://www.psychologicalscience.org/publications/observer/obsonline/harlows-classic-studies-revealed-the-importance-of-maternal-contact.html
Huh, interesting
From the Wikipedia article on Harlow:
> In an experiment called the "open-field test", an infant [monkey] was placed in a novel environment with novel objects. When the infant's surrogate mother was present, it clung to her, but then began venturing off to explore. If frightened, the infant ran back to the surrogate mother and clung to her for a time before venturing out again. Without the surrogate mother's presence, the monkeys were paralyzed with fear, huddling in a ball and sucking their thumbs.
> Another study looked at the differentiated effects of being raised with only either a wire-mother or a cloth-mother. Both groups gained weight at equal rates, but the monkeys raised on a wire-mother had softer stool and trouble digesting the milk, frequently suffering from diarrhea. Harlow's interpretation of this behavior, which is still widely accepted, was that a lack of contact comfort is psychologically stressful to the monkeys, and the digestive problems are a physiological manifestation of that stress.
I actually see this as a minor point against the popular conception of attachment theory. It suggests that despite complete absence of parents, a teddy bear was enough to give the monkeys a secure attachment style.
Yes, the Wire Mother experiment. You link to it further down, I see.
Youtube video about the experiment:
https://www.youtube.com/watch?v=RdbnwrNbINI
Yes, but attachment theory claims much more than this, and you do not need attachment theory to explain those findings. This seems somewhat analoguous with the nature/nurture-results that yes, children who are clearly abused and treated horribly will have long-lasting negative effects, but at the same time the normal variation in child-rearing doesn't seem to affect much to any outcomes we have been able to measure. The different attachment styles and their claimed effects fall to the latter "normal variation".
An extreme experiment:
https://en.m.wikipedia.org/wiki/Romanian_orphans
It seems fairly probable that lack of secure bonding in very early childhood has long term effects.
I suppose, teasing apart, the physical deprivation from the emotional deprivation, could leave room for discussion. Given the nature of people and our general need to bond with others I don’t think it’s a stretch to imagine that one’s first experience with that negotiation can have long-term affects.
This isn't directly in response to any of your questions but thought it was interesting that a YC article about working with co-founders mentions attachment theory 11 times - https://www.ycombinator.com/library/6n-how-to-work-together
I don't know that article seems like a cloud of things that are truisms and brute tautologies about what all those words mean, but mixing in some comments and predictions about things that I suspect are false.
For example the idea that there is just one attachment seems sort of wrong, or that it is tied to high quality social contact, more than quantity.
I would say for both my boys other than nursing, the "high quality social contact" was 100% more from me from ages up to ~4. I love playing with small kids, and we spent tons of high quality time on the floor of their room. Nevertheless while we have great relationships, they are definitely both more attached to their mother. Both because of the nursing, and because she is easier on them and more "mothering" of them. I am not resentful of that or anything, it is jsut the way of biology I suspect.
I do think that as the theory goes into, kids without a mother to attach to or a real poor quality mother may attach onto others, but I doubt it is often just one, and likely is a whole host of people depending on the exact balance of socializing over time.
The idea that kids with poor parentage and poor attachments as youth develop "non-standard"/pathological ways of socializing and attaching as adults seems both true, and a pretty mundane/boring/uninteresting observation.
i certainly had a crummy childhood, and that certainly is a part of some abnormal ways I relate to others compared to most, but I don't think you need some fancy "theory" to make any connections there.
IDK that "theory" seems just as much noise as signal.
My current theory about nurture is that while childhood experiences don't seem to *determine* future outcomes, they do seem somewhat *causal* in a roundabout sense
I mentally attribute all my personality traits to my parents actions, but half of them are "I was raised this way so obviously I'm like this" and half are "I was raised this way and it always bothered me so now I compulsively do the opposite." I think this is common enough. An obvious hypothesis is that our parents have nothing to do with it and it's just a natural but false assumption. An alternative hypothesis is that humans really do naturally build our personality around our childhood experiences with our parents, but there's a 50-50 chance (perhaps determined by genetics) that any given experience will result in us forming one personality trait or its opposite. This would still result in a lot of studies showing that parental quality "doesn't matter," which would arguably be true.
Its not entirely clear to me the two hypotheses are distinct. In either case, I can see the value in talking through with a psychologist the ways in which your parents have shaped your current behaviors. It's a real and natural mental construct, and changing your understanding of it could change your personality. But parenting still isn't predictive of future personality.
My view, as a psychologist: There aren't very many things that are all nurture, and it is very unlikely that this is one of them. I am sure wiring plays a part. Many people who are later diagnosed as autistic resist cuddling as infants -- they arch they backs and struggle. And I betcha if you measured things like how much time infants log looking at faces, and how they respond to cuddling you'd find a bell curve, with autistic infants at one end. On the other hand, it seems implausible that infants whose parents are absent or just cold and distant would not be affected, even if born wiring that gives them good capacity for attachment. There was a study done that so far as I know is still well thought of called the Still Face Experiment. In it, mothers were asked to interact in their usual way with the baby for a few minutes, except that they were to keep their faces absolutely blank. Infants reacted very powerfully to this, first increasing the behaviors that usually engaged their mother, then becoming frantic and howling with distress. As I recall, the effects continued to be visible for a few hours. I'm sure a chronically tuned-out, unresponsive parent would have a substantial negative effect on their infant.
On the OTHER other hand, seems unlikely that what happens with family is the only thing that can make a long term difference. Once kids are spending a lot of time with other kids the quality of kids' relationship with peer group is enormously important. I have had many patients who had loving, involved parents but bad experiences with their peers, spending years between age 8 & 18 being bullied or just friendless. They ended up avoidant, mistrustful, full of quiet rage at most others & convinced they were repellently weird. So I believe strongly that someone's attachment patterns can be greatly affected by experiences other than and later than the parent-child relationship.
P(A|B) = [P(A)*P(B|A)]P(B)
Sacrilegious commentary ( following "the rest is commentary " line on the main Astral Codex motto ( smart and undoubtedly attractive and self filtering for a certain kind of audiences ). Deliciously subversive . Not necessarily universal, and
assumingely universal.
P(B), is in practice, harder to determine than one may assume.
It is "just" the denominator and hence merely a scaling factor.
Nevertheless, this 'hard' guessing of P(B) in most applied settings, makes Bayesian inference a relative and not absolute matter: the equal sign is typically replaced by a "proportionial to" symbol.
For example, if If A is lung cancer and B radon exposure, one may have a pretty good idea of the prevalence of lung cancer in the general population i.e. P(A).
One may also be able to estimate, from appropriate sampling the radon exposure history among those who have lung cancer i.e. P(B|A).
However, to estimate radon exposure in the general population is harder -> what is an adequate sampling frame for non lung cancer individuals?
Additional perplexing commentary:
I also never understood "steps" as a metric 😅
Your starting formula is missing a divides sign! Should be P(A|B) = [P(A)*P(B|A)] / P(B) (See the version at the top of our host's page! :-) )
Actually, it's slightly more elegant and easier to remember when expressed symmetrically:
P(A|B) P(B) = P(B|A) P(A)
By coincidence, I just spotted a paper published today on the ArXiv which generalises this to quantum probabilities! https://arxiv.org/abs/2212.08088
Thanks for the references. Mea culpa in missing a the ( obvious) sign ( I could lazily blame it on ADHD - :-) ).
Thanks for the reference!!
Elegance and symmetry is indisputable. ( that's how I actually like it too, btw). But esthetics and my "mizpel aside.
The issue I think still stands. Or ?
Probability is not my strong suite in the slightest, but in all the settings that I have seen Bayes' theorem applied, the denominator is either :
1- Handwaved away, because we're only *comparing* probabilities, so only sign and relative magnitude matters.
2- Asserted to be calculable from the normalization condition (all probailities sum/integrate to 1), this is essentially the same as (1) : if you have 2 "semi-probabilities" that are 4 and 6, you can either compare them directly or - if you really want to - normalize them to true probabilities by asserting they sum or integrate to 1, and thus the underlying true probabilities are 0.4 and 0.6. It's simply a change of units.
3- Asserted to be calculable by the total probability theorem : P(B) = P(B|A)*P(A) + P(B|~A)*P(~A), where ~A is the event "not A".
Of course, for option 2, we need to be sure that the two options are the only options: if in addition to the 4 and the 6 we have a 90 that we've forgotten about, then we get 0.04 and 0.06 instead when we normalize.
(A detail, but the sort of detail that the devil proverbially resides in.)
Some more thoughts about why the crypto post bothered me so much: debates over the object-level correctness of its arguments aside, it basically pattern-matches to the arguments crypto scammers (which Scott admits are ubiquitous) make for getting money from people. I don't think Scott himself is a crypto scammer, but he does have a lot of reputation as a reasonable guy and it seems likely that him putting his weight behind arguments like this helps make people more vulnerable to them.
There's a counterpoint here that we should judge arguments based on their object-level merits, not whether they pattern match onto stuff bad people say (and while I do have some object level disagreements, I admit I'm not making them here and don't consider them overwhelmingly convincing). I think this is wrong: In cases where you know there are ubiquitous successful scammers, you should avoid trusting your object-level intuition and knee-jerk reject things that match to them if you don't have good reason to think you're better at detecting scams than they are at convincing you of scams (if you can't reliably outperform the rock with a sticker saying "crypto is scams", you're better off just listening to the rock). And as the FTX stuff showed, both Scott and a lot of other people in these parts are not reliably good at detecting crypto scams - probably good enough to detect the obvious ones, but crypto has a lot of really talented and hardworking scammers, and you're still more likely to be convinced by them than to chance into a non-scam based on object level reasoning.
"And as the FTX stuff showed, both Scott and a lot of other people in these parts are not reliably good at detecting crypto scams"
I didn't weigh in on the other thread because I got to the post too late, but this was my main takeaway -- that Scott has apparently not updated nearly enough on seeing FTX collapse to shit.
What part of the FTX debacle was due to the obviously crazy practice of just funneling FTX monies into Alameda and not setting up really anything like a robust exchange and what part was due to "crypto"? I personally weight the scam part of FTX pretty high (>50%) in terms of why it all collapsed. The fact that it was crypto was a pretty small part of it. Maybe crypto attracts this behavior?
A huge part. The Alameda money shuffle was based on:
1) FTX minting their own crypto token, then keeping the vast majority of available tokens for themselves
2) Scamming regular people out of their real money by trading it for their worthless tokens (this is how regular people lost money on the crash)
3) Taking that real money to fund Alameda
4) Selling a bunch of their trashtokens to Alameda in exchange for real money
All of this shit served to pump the apparent liquidity of both companies, which would not and could not have happened without the fakery of a cryptocurrency, inventing false money out of thin air. Crypto is the cornerstone of both these castles in the air.
(This is a very, very brief encapsulation of course, so most details and specifics are left out.)
This seems like an inaccurate summary.
2) Regular people just deposited their money with FTX, thinking that it would operate like a responsible bank (except with very high "no risk" return because wishful thinking). These customers didn't actually get tokens in return for their deposits, but rather simply a balance in their favor. Even if they did get tokens, it doesn't seem like the heart of the scam.
4) The main scam wasn't selling tokens to Alameda. FTX simply took money it was supposedly safekeeping on behalf of others and ran away with it. Yes, Alameda got a bunch of tokens. But the purpose of this was inflating its assets on paper, so that FTX could then pretend to take that inflated number at face value and give Alameda money based on this imaginary "collateral". But Alameda doesn't seem instrumental here. FTX wasn't regulated as a bank, so SBF could have just taken off carrying a huge sack with "$$$" written on it, and it would have been the same result.
So yes, there was a castle in the air. But this was used for nothing more sophisticated than tricking gullible people into thinking FTX was solvent / trustworthy. Any other fake source of legitimacy would have fulfilled the same purpose. It's a classic con (= confidence trick).
This is also consistent with the fact that FTX spent large amounts of money propping up some of their bad investments. Again, all about projecting an image of being successful.
As much as I think crypto is a huge scam and the correct market price of bitcoin is 0, the FTX scandal is only tangentially related.
Seems that you're right about 2) and I accept the correction unreservedly. The high "no risk" return is the big blinking four-story-tall neon sign that says SCAM if you just bother to read it, though.
4) I don't think is correct. The money/tokens switcheroo was a way to inflate *both* companies' value plus a way to convert consumers' money from theirs to FTX's without it just looking like naked theft.
"But Alameda doesn't seem instrumental here. FTX wasn't regulated as a bank, so SBF could have just taken off carrying a huge sack with "$$$" written on it, and it would have been the same result."
In that case, why did he bother with the whole Alameda setup? A burlap sack can't be more expensive than an LLC. Why complicate things needlessly?
No, I think it's clear that the money shuffling at the very least allowed SBF and his cronies to hold façade and collect more money from the gullible while living it up in the Bahamas.
"Any other fake source of legitimacy would have fulfilled the same purpose."
I don't think there's currently any fake source of legitimacy equivalent to crypto. It's the South Sea Bubble of our time, you can't just switch to a West Sea Bubble or some shit.
It's possible it was one of those "slip into becoming a scammer without meaning to" scenarios. Going from genuinely running FTX to "I just need to take this money here and put it all on black, that'll save my hedge fund and no-one needs to know", etc, all the while thinking themselves too special to worry about things being "technically illegal" because after all "they mean well".
From my very superficial impression, SBF sounds exactly like that. He's a psychopath. Low empathy, low impulse control. Good at saying all the right nice phrases. And unlike in TV shows, very little self-insight. Even after the whole thing collapsed, he was going on interviews and saying "Oh shucks, I'm trying to raise money to help out our customers as much as possible" as if people would somehow forget that the only reason the customers need to be "helped out" is the crimes he committed.
FTX's scam was not honoring deposits. Any other "fintech" company, say Paypal, can pull the same scam. No difference between spending customer deposits on blow and gambling on cryptos and blow and gambling in Vegas. The only difference is that crypto scammers somehow where succesful in normalizing trusting hundred-thousand-plus dolars deposits in the exchanges when people feel iffy about keeping 100$ in their fintech app of choice that's actually regulated or at least backed by companies with actual assets to go after.
I believe that's why the scam needed the fake inflated crypto to work. Putting your money in a company worth a few million feels less safe than putting your money in a company worth billions that's making so much money they don't know how to spend it all. If anyone noticed that a significant amount of the value of both FTX and Alameda were solely their own made up coins, that confidence would have dropped a *lot* and people would have pulled their money/never put it in to start.
It's like scamming 101 here, it was not a Nigerian farmer trying to give you free money, but a Nigerian Prince! He doesn't really need the money, so he's willing to help you out through his largesse! That's why you can trust him, when you would normally be concerned about giving out your bank account information to a stranger...
I don't think it matters to my argument - my main case is that crypto attracts scammers, so things related to crypto tend to be scams.
This analogizes perfectly to the internet during the 01 bubble -- companies attached “dot com” to their names to pump their stock, the field was full of scams. Good thing we abandoned that area
I don't think the rates are comparable, this is like comparing alcohol and fentanyl. Dotcom era and crypto both have scammers, but dotcom had a significant chunk of companies with sound underlying business structure (and thus was able to have a much larger fraction of people who genuinely, if wrongly, believed that their underlying business was sound or soon would be). Since crypto basically has no way to be profitable except hype, it's almost pure hype people.
What evidence do you have for the claim, "crypto has no way to be profitable except hype"? Why do you believe that? Ethereum, for example, generated about 200M of revenue in Q3 2022.
A huge amount of the FTX fraud relied specific on the mechanisms of crypto, the whole industry's culture (which is full of scammers and pump and dump, and pyramid schemes, and get rich quick suckers), and the lack of regulation in the area. All of which are issues which carry over to other projects.
Scott seems to be trying to define FTX and the collapse to a different category from "crypto" and is actively updating a different set of domains instead. I agree that this is a type of failure mode for him.
Although I agree that FTX (and MtGox before) are not exactly the same thing as "crypto," there is a lot of overlap in people and practices, and that neither of those scams/collapses was possible without crypto.
But what about 2008 crisis? It was worse, and no crypto there. Just derivatives, standard finance
That would be looking at a different set of financial categories. Indeed we should evaluate all investments based on how often or likely they are to fail. The answer in 2008 was "more likely than many thought" but not the apparent rates of fraud and loss in crypto, which seems to me closer to "you knew what you were getting into here, don't come crying to me!"
I'm seeing enormous amount of financial scam ads in Gmail and YouTube. This is probably related to the fact I write and consume content in Russian, and Google stopped caring about it.
Still, these ads are not crypto ads, but they are all scam. Banking, investment. Like 90%.
So, it would be hard to compare arbitrarily. There is a lot of scam out there. I even fell to scammy US entry documents submission website once.
I don't really understand the whole "Crypto Scammers" angle. Scammers infest a whole lot of things, I subscribe to just 2 or 3 "Reverse-Scammers" youtube channels (people who scam scammers) and its astonishing how much they post. Are there people against reddit or facebook ? Those 2 together probably host something like 50x the scammers crypto can ever hope to attract.
I'm not convinced crypto is a game changer for scammer, at most it's just an "all in one" solution that does the same things as gift cards or international bank transfers, just (very slightly) better. The best "Crypto is for bad people to do bad things" argument for me is Ransomware.
I would absolutely tell anyone to avoid a "get rich" or even just investment businesses they found marketing to them over reddit or Facebook like the plague. So yes it is the same.
What Blank said. Remember that whole Gamestop thing? Those guys weren't even scammers, just trolls who had expanded to trolling the financial system, and there were *still* a whole bunch of imbeciles hopping on that train thinking they'd get rich. I did not have a palm big enough to bury my face in. Don't take investment advice from fucking Reddit. Especially if they tell you to ride Dogecoin to the moon.
(Now imagine if Scott and other seemingly sensible people had been talking about how HODLing was actually a viable business strategy and actually Gamestop really was criminally undervalued, etc. etc. I would try to say something then, too.)
But a short squeeze is absolutely a legitimate way to make money, large Banks do it all the time whenever they notice short squeezable circumstances, and the people shorting GameStop were absolutely on the hook for more gamestop stock than existed.
Is your argument that random people on Reddit are incapable of coordinating well enough to pull off a short squeeze? Or is your argument that short squeezes themselves do not earn money?
My argument is that the Redditors consisted essentially of two cohorts: one that was coordinating to short-squeeze, not to make money but to ratfuck the hedge fund guys who'd done a Producers on the stock out of some quasi-leftist eat-the-rich sentiment (although that one guy did get rich and then investigated by the SEC and all that; never really paid enough attention to find out what ultimately happened to that guy, or even if he's gotten out of the Modern Times-esque machine of grinding gears yet), and another one that couldn't even spell short-squeeze to save their mamas' lives and who thought the stock would just keep going up indefinitely and they should hang on to it way past the date when the original shorters needed to be covered. This second group was boosting and encouraging one another like idiots and did not have anything resembling a sensible strategy; they were just the regular rube hoping that this time, *this time* he'd gotten in on the ground floor of the get-rich-quick scheme.
This is a good explanation, thanks.
Investing in complicated cryptocurrency stuff is also an entirely legitimate way to make money that can allow you to make a bunch of money, but can also help you lose your shirt.
Yes, there are tons of scams all over the place. I have a standing rule never to buy anything (or share personal information, etc.) to anyone who reaches out to me - phone, email, whatever - rather than me reaching out to them. If I get a call from a company that I recognize and should be open to dealing with but don't know the person I am talking to (picture VISA or my bank calling about a recent purchase), I hang up and call their regular number. So yeah, there are lots of scams going around. We should *not* update to think only crypto is full of scammers, but that crypto is *also* full of scammers.
That said, there's also something to the percent of crypto-related "investments" that are first-order scams (intentional fraud) and even more so second-order scams (so poorly thought out and run that the results are the same). I've said before that I think FTX was likely in that second group. What happens to your money when you invest is the same as if it were intentional fraud, and there's often criminal activity also involved, such that the results are the same and they should be treated the same. Regardless of the type of scam, there have been roughly two crypto-currencies that seem well-regarded enough that they might not result in everyone losing their money (BC and Eth) and I'll note that both of them have had technical or other issues that resulted in groups of people losing their money, and that ultimately the best we can say about them is that the jury is still out on whether they succeed. Crypto has only existed at all for about a decade. If you base your investments on who can survive a decade, you should go read about the 1920s stock market.
That post was full of painful incredulity. Ignoring, e.g., that many countries using crypto are using it for its scam purposes and focusing on self-serving myths about our swell middle-class brothers in other countries using it to escape horrible financial situations.
That crypto just coincidentally (rhetorical incredulity? The worst kind) turned into an investment asset and that this fucks with its usefulness as a currency.
Honestly, though, this crypto debacle keeps shaking down the most obnoxious people in the upper middle class so keep at it.
That last part is why I'm all for crypto -- the whole forest of schemes is sopping up an enormous amount of dollars, of which we have way too many, and in some serious fraction of cases it's not only sequestering the cash, it's vaporizing it for good.
This is all good from the point of view of reducing the amout of money in circulation, which is what is needed to tamp down inflation. Also since people are sending their dollars away to be vaporized voluntarily, in exchange for promises of gigantic wealth by and by, meaning the money is coming from people who can afford it and who are willing to believe you can get Something For Nothing, or that FOMO is a viable investment strategy, or who are planning on profiting from a financial collapse that destroys everybody else -- none of which illuminate character traits that one might admire -- this is much more socially fair than brutal interest rate increases or tax hikes, which would hurt a wider class of innocents.
Money isn’t disappeared by Bitcoin, or any asset, rising or failing - every buyer has a seller.
Unless someone loses their wallet, in which case that BC is lost and the only return value is the cash exchanged for it at its previous value. i.e. if it was purchased for $100 each, and then the wallet got lost, it's only value was the $100 each, not the five figure value that BC was later valued at but can no longer be sold for.
(Sorry for making generalizations here, please add qualifiers as needed.) A white American soccer mom, Moscow Mayor Sergey Sobyanin, Iranian dissident Badri Khamenei and a Buddhist monk in Jagadhri have some common traits due to their shared indo-european-ness: Their words for "mother" and "milk" sound alike, and they all would probably agree that there is an old thunder good with a hammer or mace. But do they have anything inherited in common in regards to daily life practices, traditions or beliefs about the world (that say a Native American, Chinese or Pygmy person wouldn't have)? Some candidates:
1. They would all agree that a bowl of yogurt is an acceptable breakfast? (But other pastoral nomads (e.g. mongols) likely agree?)
2. They would all agree that white is the most appropriate color for a horse that a king will ride? (I'm not sure they would all agree to this though.)
3. There's some kind of festival in the spring involving large bonfires? (But large fires doesn't seem specific enough.)
4. Cremation is acceptable? (but Catholics doesn't agree, and Japanese do.) Maybe "The spreading of ashes after cremation is acceptable"?
Please suggest your own ideas! But likely this is a fools errand: indo-european cultures have diffused too much and practices have been exchanged too much with e.g. China so that nothing like what I'm looking for remains.
Words similar to "milk" or "mlq" are found in languages all over the world, likewise water/aqua or words sounding vaguely like "qwa", although in some languages the words relate more to throat and swallowing. Some people think these basic words long predated Indo-European and were actually among the first onomatopeic words ever spoken by humans!
One possible answer to Medievil Cat's question is to try comparing commonly used words in various languages meaning "crane fly". I read somewhere that names for these in a lot of Indo-European languages involve "father" or "ancestor" or "man" in some form, such as "duda", the most obvious one being the English name "daddy long legs"!
Believe it or not, that is because in pagan times crane flies were believed to embody the spririts of the ancestors. One can see why, given their skeletal spindly appearance and possibly in part the buzzing noise they make as they lurch around, as if trying to communicate (although to be fair, a lot of insects buzz!)
But the clincher for our superstitious ancestors was their mass appearance at around the time of Samhain (1st Nov), and disappearance shortly thereafter, due to their short lifespans. This was the Celtic festival of the dead, at the start of winter, and was Christianised to All Souls Day, AKA Halloween, when the gates of the Underworld opened for a while and the spirits of the departed were free to wander around among the living.
I think the meaning was "Their words for 'mother' sound alike, and their words for 'milk' sound alike", not "Each of them says 'mother' with the same sounds as 'milk'."
It isn't actually true that the words for "milk" sound alike. The words are related, but you have to prove that by comparing large sets of words in each language; the words by themselves prove nothing and, as mentioned above, aren't even similar to each other. Compare Greek "gala" with Spanish "leche". Those are true cognates!
Also, the Germanic and Slavic forms represented by "milk" are not related to the Latinate or Greek forms represented by "gala"; Germanic draws from the Indo-European verb for milking an animal, while Latin and Greek draw from the Indo-European noun referring to the substance.
Indo-Iranian languages seem to use completely different words.
You may look at common colour codes - like red for king/violence/love, maybe white for purity, black for bereavement. Not even sure if this work with India though.
Good idea. I'll look into it.
Bear worship
Cremation is ok, but spreading ashes not ok for Catholics.
Why would you even imagine that indo European might be culturally unique?
What is the mechanism for your hypothesis?
Cremation is not ok for Catholics, at least it historically was not.
It has been allowed for some decades now, although it's not preferred.
Catechism 2301 - "The Church permits cremation, provided that it does not demonstrate a denial of faith in the resurrection of the body."
Ok with your edit - historically yeah. But ostensible cremation taboo has Jewish and Semitic (Egyptian) origins.
But back to the OP idea, to even mention Catholicism in quest to identify some "ur" Indo-European culture, seems off the mark (why one would even imagine that there is such a unique thing I don't understand) given that it is only 2000 years old.
Historically no, because it was adopted by non-believers as an explicit denial of the resurrection of the body: I don't believe in any of that, I don't believe I will be resurrected, so I opt for cremation as the tidier and more modern way to dispose of my remains.
Now that cremation is widespread and does not have connotations of "deliberate denial of the resurrection", it is permissible even for Catholics. A change in attitude from the 1929 Catholic encyclopaedia on the topic:
https://www.newadvent.org/cathen/04481c.htm
"The legislation of the Church in forbidding cremation rests on strong motives; for cremation in the majority of cases today is knit up with circumstances that make of it a public profession of irreligion and materialism. It was the Freemasons who first obtained official recognition of this practice from various governments. The campaign opened in Italy, the first attempts being made by Brunetti, at Padua, in 1873. Numerous societies were founded after this, at Dresden, Zurich, London, Paris. In the last city a crematory was established at Pere Lachaise, on the passing of the law of 1889 dealing with freedom of funeral rites. The Church has opposed from the beginning a practice which has been used chiefly by the enemies of the Christian Faith. Reasons based on the spirit of Christian charity and the plain interests of humanity have but strengthened her in her opposition. She holds it unseemly that the human body, once the living temple of God, the instrument of heavenly virtue, sanctified so often by the sacraments, should finally be subjected to a treatment that filial piety, conjugal and fraternal love, or even mere friendship seems to revolt against as inhuman. Another argument against cremation, and drawn from medico-legal sources, lies in this: That cremation destroys all signs of violence or traces of poison, and makes examination impossible, whereas a judicial autopsy is always possible after inhumation, even of some months.
...In conclusion, it must be remembered that there is nothing directly opposed to any dogma of the Church in the practice of cremation, and that, if ever the leaders of this sinister movement so far control the governments of the world as to make this custom universal, it would not be a lapse in the faith confided to her were she obliged to conform."
Rules for cremation:
https://cremationinstitute.com/can-catholics-be-cremated/
So, all you adherents to the sinister movement, go right ahead! 😁
Some comments have the ability to summon a wild Deiseach from the limbs where they live on a diet of potatoes and Church Fathers. That's one of the best feature of this substack.
Hey, I wasn't one of the bearers of the statue of Our Lady for the May Procession in Sixth Class for nothing, you know! 😁
That’s always puzzled me: That the most small-c-conservative among both Catholics and Protestants are the most against cremation.
One of my fellow Protestants want on a long rant making the very same argument about “denial of the resurrection.” Were I up for an argument, I might suggest that not one molecule of Abraham’s body is intact, having long since been eaten by bacteria and crumbled to dust.
Given a belief in an omnipotent god, and in the bodily resurrection, this should present no especial difficulty as miracles go.
No, I cannot imagine a falsifiable hypothesis that could be written, nor an experiment devised to test it. That’s why it’s faith and not science.
"Were I up for an argument, I might suggest that not one molecule of Abraham’s body is intact, having long since been eaten by bacteria and crumbled to dust."
It does depend on whether you're arguing "it is as easy for the resurrection of a cremated body as one that has been conventionally buried", or "it's all bullcrap". Cremation was at one time the equivalent of those Westboro Baptist protests at funerals. It also had connections to pagan customs of disposing of the dead.
And the dignified treatment of the remains is also a concern; there have been allegations of people disposing of cremated remains in fits of pique or anger due to quarrels with family members by dumping them or flushing them down the toilet, etc. Much harder to do that with a body buried in a coffin!
Cremation now is not associated with any particular anti-Christian stance, which is probably why you find it puzzling. But there's a difference between, let us say, eating pork and bacon because you enjoy it and see nothing wrong with eating it, and trying to force others to eat it precisely because they believe it to be wrong, like the stories in 2 Maccabees:
"Eleazar, one of the foremost scribes, a man advanced in age and of noble appearance, was being forced to open his mouth to eat pork.
But preferring a glorious death to a life of defilement, he went forward of his own accord to the instrument of torture,
spitting out the meat as they should do who have the courage to reject food unlawful to taste even for love of life.
Those in charge of that unlawful sacrifice took the man aside, because of their long acquaintance with him, and privately urged him to bring his own provisions that he could legitimately eat, and only to pretend to eat the sacrificial meat prescribed by the king.
Thus he would escape death, and be treated kindly because of his old friendship with them."
"It also happened that seven brothers with their mother were arrested and tortured with whips and scourges by the king to force them to eat pork in violation of God’s law.
One of the brothers, speaking for the others, said: “What do you expect to learn by questioning us? We are ready to die rather than transgress the laws of our ancestors.”
...After the first brother had died in this manner, they brought the second to be made sport of. After tearing off the skin and hair of his head, they asked him, “Will you eat the pork rather than have your body tortured limb by limb?"
As to the molecules of Abraham's body, we find human fossil remains from hundreds of thousands of years ago, so who knows but that some bones of Abraham remain somewhere? 😁
https://www.smithsonianmag.com/science-nature/east-africas-oldest-modern-human-fossil-is-way-older-than-previously-thought-180979384/
Mine will be interred, unless I think of a better place to scatter them.
How someone chooses to deal with the remains of their ancestors is pretty low down on my list of things to worry about. Remains not existing is not a problem for the resurrection.
On the other hand, “Abraham’s body does not exist,” is a falsifiable hypothesis, so probably does fall into the “Science” bucket, though the DNA would be challenging to sort out.
For some time, Jews were opposed to cremation because of its association with the holocaust, but I think some of that has faded.
Where is the quote from? In general, it's allowable (required?) for Jews to break most of Jewish law to save a life. There are exceptions, but I'm not sure whether coercion (rather than, say, no other food available) would count.
How much of Catholic doctrine do you believe?
I don't believe in Judaism, I just know a moderate amount about it and oppose anti-semities.
Believe? The bits that overlap with the Apostles and Nicean. Creeds, including the Fileque clause, with a fairly important distinction that our “catholic church” isn’t capitalized.
Even within Protestantism, we will argue endlessly among ourselves about the details.
I'm not saying that indo-europeans are culturally unique. I'm saying that it would be interesting if there's some common feature that still remains (beyond the language and the old myths). The mechanism for that would be cultural inheritance over a large set of cultural trait, with maybe some trait out of thousands surviving rather unchanged in all diverging cultures largely due to chance.
It would be interesting to search for a needle in a haystack only if you were reasonably assured that there is a good reason that there might be a needle in the haystack.
Culture is broadly speaking shared symbolic thought.
It would seem that "differences" in "culture" arise from
1. geography
(available food for example, or
relationship to a place (the mountain or water or the plain) or
relationships to what might be seen for example the flora and the fauna and the constellations)
or whether farming is viable and
2. object-making technology
(e.g. the production of music making instruments,
crafts and art,
ways of cooking,
transportation,
weapons).
Language, a particular kind of symbolic activity, obviously must be shared (or of what use would it be) but it is there any evidence that syntax creates culture? I don't think so.
Vocabulary is an outgrowth of geography and the things named - what you see and experience.
Nor does vocabulary create culture; vocabulary documents reality.
Why would anything other that what universal to all of humanity persist as soon as there is movement or expansion to new geographies or encountering/developing new technologies?
Is there any evidence that language diffusion happens at the same rate as cultural diffusion.
There might also be so-called "cultural" differences due to the shared relationship to the transcendent. But in the scheme of human history that is pretty recent.
>Why would anything other that what universal to all of humanity persist as soon as there is movement or expansion to new geographies or encountering/developing new technologies?
Because people inherit culture from their ancestors, even as geography and technology change. Isn't this obvious?
An American soccer mom and a roman senator would both understand what a wedding ring is, even though they are separated by geography and technology.
But the wedding ring is not related to language.
The premise of your inquiry as I understand it is between culture and and proto-Indo European speaker.
An "American" soccer mom is Indo-European? Maybe but necessarily.
How about if you're a Hungarian American? The other cases I'll let you think about.
I assumed that indo-european languages have been spread together with an indo-european culture, I should have been more explicit with that.
A Hungarian American is likely anglicized enough to count as indo-european for this experiment, but lets discount them as a borderline case just for the sake of it.
I'm from the same culture as the soccer mom, and bonfires strike me as more of an autumn thing than spring.
The only example I can think of of a thunder god with a hammer or mace is Thor. If this were a broader Indo-European thing, wouldn't Zeus have a hammer too? For religion more broadly, I think you'll run into problems because most Indo-Europeans outside of India have adopted Semitic religions.
Arguably Christianity is as much Indo-European as it is Semitic (if it even makes sense to talk in those terms) since it has lots of Hellenistic influences (this is arguably true for Judaism as well to a lesser degree). Paul's letters and the gospels were all written in Greek. Islam inherits from Christianity and Judaism, and the Persians have their own version of Islam anyway. So maybe some common vestigial shard remains?
Good point that Zeus has lost his mace though.
Zeus used to have a mace. Indra has the vajra, a club or mace.
I am not sure what that tells you, the club is a super old/basic weapon.
It tells you specifically and solely that there are more examples of a thunder god with a hammer or mace than the one example Bullseye could come up with offhand (for which I do not blame him, I hasten to point out lest this go in some weird accusatory direction). That was the question he implicitly posed, so that was what I replied to.
Yeah just saying really for the vast majority of history the main three weapons were club/spear/bow. Eventually a sword got added once there was copperworking.
That some gods had a club seems pretty uninteresting.
Why are so many gods associated with fruit? Because people eat fruit and like it.
Pretty sure the axe and knife came long before the sword since both can be knapped out of stone, but besides that, sure. I was only trying to supply the information requested, not in this instance tie it to some wider argument.
The various Indo-European pantheons do have strong similarities, but they are pretty subtle and scattered by historical changes. Thor shares his name's Indo-European root his Gaulish equivalent Taranis, maybe the Hittite Tarhunta, and of course with the English word "thunder". IE thunder gods usually slay giant serpents: Zeus kills Typhon, Thor fights Jormungandr, Indra slays Vritra, and Tarhunt kills Illuyanka. Often they are part of a triad of gods that rule over sky (the thunderer), land/sea, and underworld, much like Zeus, Poseidon, and Hades. (Poseidon was not only god of sea but also of earthquakes, so he was also associated with land; presumably the maritime Greeks cared more about his rule over the sea.) These three gods are children of the highest Sky Father (the Sanskit Dyaus Pitar, Jupiter, Kronos, Odin All-father), who is associated with regality, laws, and oathkeeping. Note that in the oldest Roman religion, Jupiter was *not* god of thunder, and was not the equivalent of Zeus; however, the Thunderer Zeus absorbed the functions of the Sky Father as well (as reflected in the myth of Zeus overthrowing Kronos and ruling over his brothers), including his name (Dyaus = Ju- = Zeus), and later authors linked him to Jupiter.
> Often they are part of a triad of gods that rule over sky (the thunderer), land/sea, and underworld
This is a classic example of a feature that can easily be the same everywhere due to independent invention everywhere.
Zeus has the thunderbolt, which is very recognisable as coming from Indra's vajra, or at least in the same family. All the three main Olympians - Zeus, Hades, and Poseidon - have an attribute of the (one pointed) thunderbolt, Hade's bident, and Poseidon's three-pronged trident.
See here for bident:
https://en.wikipedia.org/wiki/Bident
The Devil having a pitchfork (which gets depicted as three-pronged, rather than two-pronged) ultimately descends from Hades and the bident.
Vajra:
https://en.wikipedia.org/wiki/Vajra
Compare with depictions of Zeus' thunderbolt on coinage:
https://en.wikipedia.org/wiki/Thunderbolt
If nothing else the wood in autumn is far more suitable for bonfires, in every way.
5. It seems appropriate for a young man to go on a journey, explore and learn new stuff before settling down and starting a family? - But this is likely present in other cultures as well.
6. You can refer to alcoholic beverages as "coming from the gods" or otherwise being associated with a god or goods? But this also seems pretty common in plenty of cultures.
I don't think this is going to be a fruitful search. Too hard to distinguish things which are actual cultural holdovers from things which developed independently in many places because they just make sense given human psychology/technology/etc.
Not really? Plenty of cultural traits are arbitrary, and shouldn't develop independently in many places. E.g. I know of no independent mace-wielding thunder-gods.
Plenty of them are arbitrary, and those ones won't repeat. But I would argue many cultural practices are not arbitrary, and those ones will repeat and will foul up your ability to determine which ones are actual shared heritage, and which ones are convergent cultural evolution.
If you know of any trait that could be shared heritage or could be convergent evolution which matches my criteria, feel free to share. You have seen my attempts, you see that I haven't found a good one. So I think your worry is overblown: the problem is not "how do we separate the X things from the Y things?", it's more "Are there any things at all that could plausibly be X or Y?".
Well sure "Respect you elders". "Don't go out at night". "Strangers are dangerous". "Children are irresponsible and need minding". "The young leader was disinherited and then returned and took rightful power". "Big animals are dangerous". "Water/Rain brings life". "Bad/untrustworthy people are left handed".
There are a lot of common features, and I suspect if wood artifacts survived with any regularity we would find many more. People aren't very different overall and find common solutions to common problems. Some of those solutions are cultural.
No. Those seem to exist in e.g. Native American, Chinese and Pygmy culture and thus they don't fulfill my criteria. I want to find something that's shared in the Indo-European cluster and doesn't seem to exist outside of it. I.e. mace-wielding thunder-gods, but in other realms than mythology.
I don't think this will work, the Iranian guy in particular is injecting tons of noise into the exercise because of the 1400-year-long contact with Islam, the Indian Aryans were locked behind the Himalayas for millennia and are much closer to Central Asians, other "natives" of the Indian subcontinent and China than to their ancestors.
I think that "Indo-European" is about as useful a category to try to map back into modern categories as "African-origin humans". It's so diffuse that, even if they do share some unique thing, that thing is now spread all over the gene pool and the meme pool with no hope of ever recovering the fact that it once belonged to Indo-Europeans alone beyond reasonable doubt. Exceptions might be things like "Blonde Hair" and "Blue Eye", and I'm not even sure they are exceptions. (Africa and most of Asia is definitely out, but South America is full of blonde-haired blue-eyed people with plenty of non-european genes ?, and Egypt and the Levant are as well to a much lesser degree)
The Moscow Mayor and the White Soccer Mom from America share much more than the rest though : Both Christian, both white-colored to the same rough degree, their grandparents probably fought with or against each other in a war in Europe somewhere, their 3-centuries+ ancestors probably traded/intermarried/emigrated together in Europe, things like that.
"South America is full of blonde-haired blue-eyed people with plenty of non-european genes ?"
Are you serious? Those are 1000% descendants of European immigrants, and the fact that there are still sizable populations of blue-eyed blondes in e.g. Argentina shows you the extent of de-facto racial segregation that goes on down there.
Oh, I know of course that all blonde hair and blue eyes in South America is due to european genes, I just don't see South Americans often lumped with "Indo-Europeans", possibly because of the significant native-black intermixing. So my point was simply "Even for something as undeniably Indo-European such as blue eyes and blonde, South Americans also have it but are not usually thought of as 'Indo-European' due to massive culutral mixing"
I don't think it's this hopeless. The Iranians and Indians do share linguistic traits and mythology with the rest of the Indo-Europeans even though the factors you state apply. It doesn't seem impossible that there are other traits that are also shared.
What's hopeless is establishing that a shared trait is shared through common inheritance rather than for any of several other reasons.
>There's some kind of festival in the spring involving large bonfires? (But large fires doesn't seem specific enough.)
Huh ? We do that in the autumn/fall in the UK.
Perhaps something like:
Sibling/cousin is a primary grouping, with age difference and sex secondary groupings.
(Whereas in much of East and Southeast Asia, an older male sibling is more like an older male cousin than like a younger male sibling.)
Apart from aspects of language itself, could there be a single cultural trait which all Indo-European groups have but which Hungarians lack? I put the chances of such a trait existing at less than 10%.
The Database of Places, Language, Culture, and Environment is the best ethnographic resource I know of in which to search for possibilities: https://d-place.org/parameters
Unfortunately, there are 2,368 different variables. I wish someone would go through them all and pick out a top 100 most informative/interesting. (Or devise some other way to make it more digestible.)
I have a question for those who want US and EU (“the West”) to broker a peace deal between Ukraine and Russia that would entail a loss of Ukrainian territory or/and some other concession which Ukraine currently rejects. How, practically, is this supposed to work if Ukraine isn’t willing to play ball?
Only solution that I can think of is that the West would simply threaten Ukrainians with withholding critical aid unless they moderate their demands. Is this the plan? Or is there some other way I am not seeing?
That's a stick, I guess. Could also use carrots? Money for rebuilding seems obvious.
What I've seen from Ukrainians is that they are deathly afraid of making a peace deal now and signing a piece of paper, Russia rebuilding their military for as long as they need, and then one day Russian tanks are again speeding towards the capital city. The only thing that could alleviate this fear are hard security guarantees from the EU or the US - something like foreign troops stationed in Ukraine as a trigger (like South Korea) rather than being a piece of paper.
Israel has done this for years and years now.
Yeh it’s a bit catch 22. It could be a UN controlled zone where any invasion is an attack on the UN, but then there’s a veto.
> something like foreign troops stationed in Ukraine as a trigger
If Russia doesn't want Ukraine to have NATO membership, imagine what they'd think of Ukraine hosting NATO military bases.
NATO already pledged to support Ukraine for "as long as it takes", so unless you redefine what that implies to get out on a technicality, or take a major hit to your credibility, that lever seems unavailable.
I mean, "it" in that pledge seems somewhat underdefined
Yes but that was only the informal wording used in press reporting. Looking at the official NATO website here (topic 4, What are NATO and Allies doing to help Ukraine?):
https://www.nato.int/cps/en/natohq/topics_192648.htm
It seems clear that "it" means preserving Ukraine's right to exist, war crime prosecution, and post-war reconstruction. This can only be understood as "until and after Russia has been fended off", because otherwise NATO couldn't do any of these things.
Also see the statement of NATO foreign ministers of 29-30 November 22,
https://www.nato.int/cps/en/natohq/official_texts_209531.htm
with lines such as
> We remain steadfast in our commitment to Ukraine’s independence, sovereignty, and territorial integrity. We will never recognise Russia’s illegal annexations, which blatantly violate the UN Charter. We will continue and further step up political and practical support to Ukraine as it continues to defend its sovereignty and territorial integrity and our shared values against Russian aggression, and will maintain our support for as long as necessary.
Point taken, but all of this still seems pretty weaselly, at least to me. Like, "We will never recognise Russia’s illegal annexations", but what if Ukrainians hypothetically agree to legalize them under military pressure from Russia? Nowhere does it say "we will give Ukraine enough aid so it will never be in a position when surrendering territory would seem like a lesser evil to Ukrainians".
Moreover it should be noted that US does not have a much to fear from a hit to their credibility in the EU, and neither US or EU have much to fear from a hit to their credibility among Ukrainians.
In sum, while I do not personally support using this lever, I think it is very much available, even fairly likely to be used.
I disagree about your last paragraph. A real loss of credibility in th EU (something deep enough to threaten the integrity of NATO) would really put the US on the back-foot vs China or any great power aiming at challenging the US domination.
Frankly, setting aside whether US anti-China policy makes sense (I am very unsure about that), in this area EU is not currently doing much more than it has to in order to not wreck their (well, ours) relationship with the US, our main protector against Russia and any other threat. And EU is unlikely to ally with China no matter how bad our relationship with US gets.
Arguably, loss of US credibility in the EU might even increase European cooperation with US anti-China measures, since if Europeans would be more afraid that US is going to abandon us, we would try harder to be useful to them. Like Australia.
>Nowhere does it say "we will give Ukraine enough aid so it will never be in a position when surrendering territory would seem like a lesser evil to Ukrainians".
Considering the strongly worded nature of the declaration and general public opinion, dragging their heels significantly enough to affect the outcome of the war would likely be considered a breach of the promise.
>Moreover it should be noted that US does not have a much to fear from a hit to their credibility in the EU, and neither US or EU have much to fear from a hit to their credibility among Ukrainians
While the EU is certainly involved and has its own stance on the war, we are talking about a NATO statement. Very different matter. The mutual promises of the EU are much less rigid and much more negotiable compared to the core NATO promise.
>In sum, while I do not personally support using this lever, I think it is very much available, even fairly likely to be used.
After their declaration, the lever is absolutely not available any more to NATO. The whole idea of NATO hinges on the mutual, 100% certain assistance in (defensive) war. If NATO makes that promise, even to non-member Ukraine, as strongly worded as this and unilaterally walks back on it, it would have severe negative consequences for the whole treaty. If that promise is any less than 100% credible, it would invite outside aggressors which NATO is supposed to deter to test just how much below 100% the promise is now really worth, which defeats the whole point.
>it would invite outside aggressors which NATO is supposed to deter to test just how much below 100% the promise is now really worth, which defeats the whole point.
Like whom? Do you think that once Mexicans find out that NATO is not that credible and gringos thus might not get the help from the mighty Bundeswehr, they will try to reconquer what they had lost in the last US-Mexican war?
I jest, but only slightly. Obviously, e.g. Estonia is quite dependent on NATO credibility, but Estonian military industrial complex would not be able to keep Ukraine afloat. Ukrainians need the help of mighty industrial nations of the West, who practically by definition have much less to lose from the loss of NATO credibility; although Europeans do have more to lose than Americans.
I can't see NATO saying that they will stop supplying Ukraine if it doesn't sign such a deal. But I can see NATO saying, when a deal is on the table, that if Ukraine agrees to it and Russia doesn't NATO will give even more support to Ukraine than it has been giving. That would be an incentive for Ukraine to agree.
Ukrainians get sick of dying and getting their country blown up? Even if they might eventually "win" with indefinite NATO support so they reduce demands.
That would however be a different scenario than that which people who want active Western role in brokering peace have in mind, imho
>Ukrainians get sick of dying and getting their country blown up?
This has never happened in world history and you would have to assign that outcome a miniscule probability. The land war is the only one that matters, and as long as the Ukrainian people believe their leaders and soldiers can throw the Russians out, they will endure.
>This has never happened in world history
??????????????????????????
>??????????????????????????
!!!!!!!!!!!!!!!!!!!!!!
So like what was going on in the 13th and 14th century then? Or 1667? Or say 17th and 18th century with the Russia progressively stomping out and co-opting the Cossacks/Hetman as an independent entity/force?
Then in WWII a bunch of the Ukrainian people threw in with German invaders (some for fairly good reasons), and then these were in turn pushed out by the Red Army as it eventually won.
How does that all square with your flat statement of invincibility?
I believe the commenter before me was referring to the strategic bombing being carried out by Russia against Ukrainian civilian infrastructure ("Ukrainians get sick of dying and getting their country blown up").
What I declared to be without historical precedent was such strategic bombing winning a war on its own by defeating the morale of the civilian population, as has been proposed since WW1.
Really? Are Ukranians willing to suffer and die for Crimea? Because the power hungry fool Zelensky says the war isn't over until Ukraine controls Crimea.
In case you aren't trolling, Crimea is part of Ukraine and its annexation by Russia has not been recognized internationally save for a handful of small countries, including flavourful names like Afghanistan, North Korea, and Syria.
And evidently, yes, the Ukrainian people are willing to fight and die for it, or we wouldn't be having a 10 month war. Ukraine is fighting an existential war against a genocidal regime, everyone involved knows it, and that is entirely sufficient to explain the cohesion and morale of the Ukrainian people. Ukrainians are not puppets, they are people with agency, and to imply that Zelensky, or presumably his Western handlers, are somehow controlling them against their will is at best unnecessary, at worst enemy propaganda. Yes you can force people to fight, see Russia, but the results speak for themselves.
I don't think we know whether, if the war was not going well for Ukraine and Russia withdrawing from everything except Crimea was an option, Ukraine would accept it. They say they won't, but that doesn't tell us what they would accept if they couldn't get anything better.
Indeed, until they actually invade Crimea, I think that must be considered potentially part of a deal. I think there are internal politics that make it difficult for them to say so. But in the last eight years, it was the Russia-backed insurrectionists in the Donbas that they challenged.
Yeah, with that I agree. It would depend on the circumstances
Ukrainians absolutely are ready to suffer and die for Crimea. You sound like someone who never met an Ukrainian.
>Only solution that I can think of is that the West would simply threaten Ukrainians with withholding critical aid unless they moderate their demands. Is this the plan? Or is there some other way I am not seeing?
All of the western weapons and aid are what is holding the Russians back, the Ukrainians are playing with the west's ball. If the west takes its ball and goes home what would happen to the Ukrainians? I'm guessing Ukraine would eventually get steamrolled. And what would happen to the west? They'd avoid taking the huge hit from the gas bill, spend less on weapons, maybe take a bit of a credibility hit, lose a future ally vs Russia. Who holds all of the leverage in this situation?
I don’t think Ukraine is that inconsequential for “the West” as you imagine it is. Depending on the country for sure, France maybe doesn’t care that much, the US is now seemingly always one election away from withdrawing from NATO, but the Eastern Europe will absolutely lose their shit if Ukraine were to lose because the West “is tired”. Spending less on weapons is 100% not happening, Poland is already militarizing up to the gills and arguably at the safest time when Russians demolished their professional military. There would be huge humanitarian crises, geopolitical and military repercussions up to nuclear weapon proliferation, and then green light to China with regards to Taiwan. Also I suspect the democracy in the West would suffer as well, if only because Russians would be allowed to keep buying up Western elites which is currently somewhat on hold due to this war.
I think that its largely inconsequential for the decision maker in OP's question, which is the US. Even if Eastern Europe does lose its shit what are they going to do, buy more F35s? I don't see that as deterring the US to force a settlement. And I don't see nuclear proliferation in NATO countries either, I don't think the cost benefit analysis favors a country like Poland pursuing nukes. Either way it is definitely inconsequential when compared to what Ukraine might face if the aid was cut off. Which gives the US the leverage.
> There would be huge humanitarian crises, geopolitical and military repercussions up to nuclear weapon proliferation, and then green light to China with regards to Taiwan.
What geopolitical and military repercussions are you expecting to happen in the event of a US forced settlement?
And I don't see the US forcing a settlement in Ukraine as a green light to China on Taiwan either. One because I think that the US should force a settlement in Ukraine to focus its energy on China in the first place. Two because invading Ukraine has been extremely costly for Russia, who didn't have to do it via amphibious assault. I would imagine that a Chinese blockade would be similarly costly as the Ukrainians have had a lot of success against the Russian navy. And three because the west came together and basically cut off all exports from Russia. I think China can reasonably expect the same response from the West if they were to invade Taiwan, which would absolutely tank their economy.
>Also I suspect the democracy in the West would suffer as well, if only because Russians would be allowed to keep buying up Western elites which is currently somewhat on hold due to this war.
Maybe that is more of a EU / Germany thing but I don't see much evidence of that happening in the United States to begin with.
> I don't think the cost benefit analysis favors a country like Poland pursuing nukes.
I think the opposite, NATO is no longer an unconditional alliance, American public is at least okay with questioning it or okay with voting for anti-NATO politicians for some other reasons, which is their right sure but that has security implications for Poland which ultimately can be resolved only with nukes.
Same for geopolitical repercussions really, eventually I’d expect new alliances formed in Europe.
> Green light to China with regards to Taiwan
Putin launched his invasion in part because he thinks the West is weak. The West “sacrificing” Ukraine’s territory is going to be seen as confirmation of that fact. True, Russia pays dearly cost for the war but it does not jeopardize Putin’s grip on power (seemingly?) which is the lesson China can learn. Just as they learned that famines are not a problem for the communist power. Also China is way more entangled with the West economically for the economic warfare to have a deterring effect. Even with Russia, it’s not clear what the effect is. The Economist publishes articles with the opposite conclusions: sanctions don’t work/sanctions work better than you think, all the while the Europeans are complaining about the gas bills. The leverage of the West over China is questionable.
>I think the opposite, NATO is no longer an unconditional alliance, American public is at least okay with questioning it or okay with voting for anti-NATO politicians for some other reasons, which is their right sure but that has security implications for Poland which ultimately can be resolved only with nukes.
Polling I have seen for the US generally shows support for NATO. Ukraine is not and has never been a part of NATO. NATO is giving Ukraine free money and are under no treaty obligation to do so. I don't see how adding conditions to free money to someone you were not obligated to help would somehow undermine Article 5. Unless you care to explain.
As to Poland I think you are greatly underestimating the pushback they would get if they tried to get nukes. Google has told me that Poland is planning on opening its first nuclear power plant in 2033, I seriously doubt that they would be able to withstand 10+ years of the Iran or Iran lite treatment. Especially if kicked out of NATO / EU.
>Putin launched his invasion in part because he thinks the West is weak. The West “sacrificing” Ukraine’s territory is going to be seen as confirmation of that fact. True, Russia pays dearly cost for the war but it does not jeopardize Putin’s grip on power (seemingly?) which is the lesson China can learn.
No idea why you think forcing a negotiated settlement makes the West look weak. Nothing about this war makes me think the west is weak. The west comes out looking great - all of our shiny toys work great, we all came together on some sanctions (effective or not), the Russians look like a wet towel.
As far as the "sacrifice" framing - again the west and NATO and the US in particular doesn't have any defense treaties with Ukraine. We are giving them aid because we think its in our best interest and because we are being nice. I don't think its reasonable to call attaching strings to the free money we are giving as sacrificing Ukraine, even though yeah it does suck for Ukraine.
Vis a vis China - I think the Chinese (and everyone else) will also see how vulnerable their navy is to missiles and drones. I think that the Chinese will expect at least a similar level of support to Taiwan from the US (IMO should be more as ideally the US is forcing the settlement to focus on China). I think that the Chinese will see the ferocious response from Ukrainians and think twice about their chances in Taiwan. Overall I don't think Ukraine really has any strategic value to the US, at least when compared to Taiwan. So we should force and end and move our focus to China / Taiwan.
"I don't think the cost benefit analysis favors a country like Poland pursuing nukes."
LOL. Thus speaketh a man who lives far from the Muscovite border. The Poles don't reason remotely the way you do.
I tend to agree with those who think that Ukrainian defeat in this war makes nuclear armed Poland much more likely. It is, however, unclear whether that would be something Washington would be worried about.
I would be fairly confident that an Ukrainian defeat would have triggered some serious confidential talks with the French about the use of their force de frappe, and the ASMP specifically.
Russia isn't willing to play ball, either.
Ukraine applied for NATO membership on 22 September this year. There is now a process involving an invite to join, a protocol to be drawn up and accepted and so on, however at some point, NATO will include Ukraine - almost certainly to the contentious internationally-accepted border that includes Donbass and the Crimea. Does this compel Russia to make peace sooner - or is this the very scenario that Putin objected to last November, the "NATOisation" of Ukraine? Is there an inevitable slide to a NATO-occupied Ukraine that only a Russian victory can hold off? A reasonable compromise in these circumstances may be Russian retention of Crimea and Donbass in exchange for NATO occupation of Ukraine.
"NATO-occupied". How exactly is the country occupied if it's a voluntary member which applied to join? And if it specifically applied to join due to Russia's unjustifiable war of invasion?
That's a game that Russia apologists play all the time.
When Russia does something horrible, they will loudly object against using any word with negative connotations. For example, killing entire village cannot be referred to as "massacre" or "murder" or even "killing" or "indiscriminate killing"... you are supposed to use the proper term "zachistka" (sweep). Because if you don't use the proper term, it shows that you are "nyekulturny"; an illiterate simpleton unable to discuss the complex realities of life. You need to educate yourself first.
And trust me, Russians have more words for hurting people than Inuits have for snow.
On the other hand, anything done by opponents of Russia must be described in the most hysterical terms possible. Thus, for example, Russian army coming to your country, killing everyone who seems like they might resist, replacing your government, and staying there indefinitely, controlling your politics and draining your economy, is called "friendly help"... but if your country applies for NATO membership, and afterwards does a joint military training once in a while, that is "occupation". Opposing Russia literally makes you a "Nazi": https://www.youtube.com/watch?v=BRFcK1BDlus
There is no such thing as voluntarily doing something Russia does not approve of. That makes you simultaneously a victim that must be rescued at any cost, and a villain that must be punished, preferably by death. The idea that someone does not want to become a part of Russia is just silly. And even if they do not, who cares, and why should that be relevant in any sense?
But the silliest of all is the idea that someone could become a "Rusophobe" as a consequence of what Russians do.
You mean like how NATO calls people terrorists, but will freak out if you say anything negative about their psychotic military operations in the middle east and north africa?
That's just Whataboutism. Yes, many countries do bad things. That doesn't mean that it's somehow okay for Russia to do bad things.
I have no disagreement and nothing to add. Have a fistbump.
Oh? Really? What has anyone offered them? Zelesnky says the war isn't over until Ukraine controls Crimea, meaning Russia is being offered literally *less than nothing*.
And they should take their less-than-nothingburger, eat it, and act grateful.
What is Ukraine being offered?
They are being offered an end to the sanctions.
If the war continues to be a large net negative for them, that might be enough.
Not having tens and potentially hundreds of thousands of your citizens killed is substantially more than nothing, if you are not a psychopath.
That's a big If
If you're saying the Ukrainians should make a deal to end the war, that would take trusting the Russians quite a lot.
That seems like the obvious way to apply pressure--perhaps a secondary one is to decrease sanctions on Russia.
a) Is it "the plan" of our politicians? - I doubt it. It is mainly Kremlin-Trolls in the west. Who have been esp. active recently - on FP, Newsweek et al. . . See for competent analyses: https://slantchev.wordpress.com/ (highly recommended!)
b) Ukraine is doing demands. Fine. Russia is doing different demands. Not so fine ones. - IF both sides believe they may be able to work something out to bridge the differense: They better do so.
c) Right now, it does clearly NOT look that way.
As Putin insists on keeping all occupied territory - and then some. And would break any ceasefire the moment it pleases him. (While not winning!) - OTOH If he would offer going back to 22.2.22 then there might be room for talks: 1. Ceasefire to allow quick Russian retreat. 2. Talks about the future may start. Might work out, or not. "The west" would THEN probably offer the Ukrainians many reasons to go for peace (even without Crimea, yes). Carrots mainly. Big ones. The "stick" being: Maybe less carrots.
Ad a), to be clear, I don’t think it is the current plan of Western politicians. I am also against it. On the other hand, I don’t think it is fair to label all its supporters as Kremlin-Trolls. This sort of debasement of the discourse is unhelpful.
Scott, you should add 'Links' to the title when you include links!! To let people know!
The number of Open Threads without links of any kind is vanishingly rare, I think people know to expect some links.
What is the morally right thing to do with a person who is in a permanent vegetative state? I have a close family member who has no chance of recovery and seems cruel to keep them in this state. thoughts?
Euthanasia is not legal here
Discuss DNR options as that person would have wished with other family members. Ensure the doctors know as well.
Visit them often, ensure they are being properly taken care of, and pay the care takers appropriately. Spend time with the ill person and ensure you remember their humanity, so that the staff does as well.
Remember that this is a human person and not your property to dispose of when inconvenient or distressing. Be realistic but do not fail of hope.
It is harder on you than on the ill person. Be kind to yourself as well.
"It is harder on you than on the ill person. Be kind to yourself as well."
That's a good point.
By all accounts people in such states, even if they have some brain activity, do not seem to be especially unhappy. Perhaps they are dreaming, or in some state like it.
Uh, cite your sources, please.
I mean if it was a close loved one like a wife or child and there was no chance of recovery I would just kill them myself regardless of whether Euthanasia was illegal or not. Society isn't determining my family's end of life choices (though we do have the right legal docs in place in our case).
We are just space dust and literally everyone dies. Keeping the husk alive at a huge cost of resources is pointless.
If it was someone else's close family then they should decide (asusming they can pay for the care).
I believe the moral imperative in these situations is to minimize human suffering; both that of the patient's (potential) ongoing suffering, and that of their survivors, because quality of life matters *infinitely* more than mere continuation of life.
So if there is little to no chance of the patient experiencing a basically good quality of life in the future, the morally correct thing to do is end the suffering of all concerned as quickly as possible. I would advocate for euthanasia if it were an option, but if not, withdrawing care is probably going to be the only course available to you.
It's the course I've directed for myself in my living will, and that literally every single person who is close to me has directed for theirs. The only people who disagree with the idea of quality of life mattering more than mere life tend to be religious people who derive their morality from irrational beliefs. Don't allow them to be your guide in this.
I'm so sorry that you're going through this.
Out of curiosity, at what point in the spectrum of 'quality of life' do you regard that quality as insufficient to justify not killing the person, or yourself?
And as a follow-up, do you have exact same break point as all the people who also see end of life options in this way?
I think we should treat people with as much compassion and dignity as we treat pets, which is to say when someone can no longer eat, drink, and relieve themselves, and they are unresponsive or in too much pain to communicate and/or rest comfortably, and there is no reason to think they will recover enough to ever do those things again, they should be helped to die as quickly and painlessly as possible.
Of course not everyone shares my opinion on this (which you know, so no need to be snide), but there is a reason advanced directives and living wills were invented and every healthcare provider now hassles patients about getting one on record. The healthcare default is to preserve life regardless of suffering or literal cost and enough people think that's such a terrible idea that we invented a whole legal process to override it.
Well better than I've seen some pets treated. But otherwise I totally agree with you. My mum died from Alzheimer's, and my only plan is to die before then, or somehow off myself if diagnosed. Oh which brings up the question of whether I should tell my loved ones before offing myself, or after... hmm I guess, before is the obvious answer.
As far as the OP and the poor person in a coma, my rationalist answer is to find the time for other people to come out of a coma from the same cause. And when the loved one is past the 90% mark (or 95?) then give it up.
I recently caught a This American Life episode about an American with an Alzheimer's diagnosis going to Dignitas in Switzerland to die. The service and process was both practical and beautiful.
https://www.thisamericanlife.org/779/transcript
Regarding your second question: it’s
A virtual certainty that she and and others who share her view will not have the exact same break point. Likewise, people who think the income limit of about $125K that qualifies people for partial student loan cancellations is too high will not all have the exact same number in mind for what a reasonable limit would be. People who are in substantial agreement about something are not going to be in perfect agreement about every detail. Everybody knows that. So what’s the point of demonstrating that Christina does not have the exact same standards for what counts as a live worth living as others who favor euthanasia?
I personally think it is right to euthanize these living remains of a no-longer-conscious being. However, I do not want to give an extended defense of that. But here are a few practical suggestions:
Read up on these states. There are different levels and degrees w of neurovegetative states. Articles are easy to find via Google. Also research predictors of recovery, and tests that give you an idea of whether the person is at least partially conscious. Here’s one good one: https://neurosciencenews.com/consciousness-sniff-test-16445/
Once you know about this stuff and perform these tests, you may be clearer about whether the person’s mind is permanently and totally gone. If it is, and you believe you should euthanize them, you should know that in some countries this is done in an indirect way that gets around the law against euthanasia. Nourishment and water, which are being supplied by gastric tube or IV, are withdrawn, with the rationale that they are treatments from which the patient is not benefitting. Perhaps that is legal in your country. Another option would be to ask to have the relative transferred to the care of their family. The family could then withdraw nourishment and water, or possibly find some less protracted way to end the person’s
life. But of course everybody involved would have to be in agreement with this plan.
My son (7) has not really caught the bug for reading yet, but I think I’ve found his subject. I was telling him some Greek myths the other night and he was fascinated. He wants books on them for Christmas.
Anyone have any recommendations for books either about Greek myths or based on them? He reads well so they don’t need to be geared towards first graders, but they probably should have pictures. Thanks!
Excellent - when I get home I’m putting it on the list. Thanks!
The D’Aulaires Book of Greek Myths
Looks great - thanks!!
Stephen Fry’s Mythos, Heroes and Troy? Even better in their excellently performed audiobook form!
"Heroes, Gods and Monsters of the Greek Myths" by Evslin, Evslin, and Hoopes. I absolutely loved these when I was a child. Something about the language made them a joy to read.
The Percy Jackson series is pretty great. Aimed at slightly older kids, but likely to appeal to younger ones as well. The author, Rick Riordan, has also written series involving Egyptian and Norse gods.
I'm interested in whether people think AI is going to render a lot of professions obsolete in the next few years.
I was quite shocked to see GPT3 programing, admittedly not to a professional human standard, but it was basically competent. I could easily imagine the next generation of AIs surpassing my ability at least, and obviously it would be exponentially cheaper than hiring human coders. I can't see how I'd stay employed in that scenario.
It seems like most jobs that mainly involve manipulating information are in the same situation. Accountants, doctors, journalists, lawyers, finance, engineering, basically the entire professional middle class, might end up going the way of the hand-loom-weavers.
Maybe I'm overreacting, there's a lot of (what sounds like) hyperbole around AI at the moment, but could we genuinely be looking at an early-stages-of-the-industrial-revolution type disruption of the labour market over the next decade?
A year ago I would have laughed at the idea, but does a near-future where AI does all the cognitive work and human workers all do manual labour (or are celebrities etc.) seem plausible?
It's not just a Markov model--the whole interesting thing about it is that it keeps some kind on context beyond the last N words output, which is what allows it to be a good chatbot, and to stay on topic for multiple sentences. Markov models do not work that way--they produce sentences that sound like English but don't make any sense and can't maintain anything like a point.
Is there a reason to think that the limitations of chat GPT's quality of writing are inherent in the technology, rather than just "computers may beat humans at chess, but they'll never conquer go...oops, wait?"
I'm wondering if something like GPT-3 will become the mouthpiece for some more general AI, just as various classifiers will become the image processing/voice recognition part of the general AI. The AI knows it wants to get a particular human to wire $1.3 million dollars from account X to account Y, and so tasks GPT-3's descendant with the job of writing an email that asks the human to do that in a plausible way, provides detailed and easy-to-follow instructions for doing so, and maintain's this AI's friendly persona in the tone of the email.
I mean, there's *something* like a language module in humans, because damage to particular parts of the brain can scramble our language abilities. (Various flavors of word salad result from different brain injuries, as well as just flat losing spoken language.) And I think something like a module for face recognition, image recognition, etc. must exist. So my assumption is that all these are modules that will eventually plug into an AI, but also into any number of other automated systems.
As an example of this now, if you ever call a credit card's 800 number to get your balance or question a charge, you get a voice recognition system. That's an AI algorithm (some kind of neural net classifier, I think) that's being used as a module by the voicemail system, which is just as dumb as it ever was. You just get to say the credit card number or selection in a menu instead of hitting a button.
If you replace "few" with "many" in the sentence, I would agree. What I think is far more likely for a long time is that AI becomes a supporting tool, and that some professions will get a lot of "AI wrangling" as part of them.
Examples: An artist can use AI to do boring work with filling details, drawing repetitive items, or even creating a first image that the artist can then bring into publishable quality. The programmer can use AI to quickly create a function, even when the AI can't be trusted to make a complete program. The lawyer can use an AI to auto-create the boring routine stuff. If a truck AI can't handle chaotic city traffic and all the non-driving stuff a trucker does, maybe it can still drive a highway autonomously while the driver sleeps?
That's pretty much my thinking as well. I'm reminded of how long it took to get good compilers for third-generation languages. Back in the sixties, seventies, and even into the eighties, a good programmer really was supposed to understand the quirks of compilers and write around them, or even just skip that and drop down to writing ASM in critical sections.
Let's also remember that a good part of a programmer's work isn't just writing code to do something, but rather negotiating with humans what the system should do in the first place and determining whether the system can be made to do what it is being asked to do either at all or with a reasonable amount of effort. I expect the same can be said of any number of white-collar professions.
"Back in the sixties, seventies, and even into the eighties, a good programmer really was supposed to understand the quirks of compilers and write around them, or even just skip that and drop down to writing ASM in critical sections."
I don't think this expectation has gone away because the compilers are better so much as because compute has become unbelievably much cheaper. Assembly language is still vastly more efficient than most other code, people just don't care about efficiency anymore.
Part of the it is that, yes, but we've definitely gotten to the point where good compilers really are better at optimization than all but the best programmers in most circumstances. Also, many of the things that made compute cheaper also made architecture much more complex, so it's actually harder to manually optimize than it used to be.
Assembly language isn't really directly telling the machine what to do anymore--instead it's more like a language that's being interpreted/JIT compiled by the hardware executing it. (It's not at all clear to me why it makes sense to do the optimization at the hardware level rather than a level or two earlier in the compiler or maybe in some JIT compiler thing in software, but maybe that ship has sailed.)
Itanium (aka Itanic), or VLIW more generally, really was that ship and it really has sailed. It seems that in the end there are too many unpredictable variables wrt recent context switches, caches, and so on for optimization of code not to include a dynamic component, plus VLIW bloats code size making instruction caches less effective. As for doing it in software, the microcode is software and is used for the parts of the optimization that can be done fast enough in software.
One reason for doing out of order execution is that the amount of time required for a load operation varies greatly depending on where the value you are loading resides. On a modern machine, a load operation might look in a level 1 cache for the value being loaded, then look in a level 2 cache, then look in a level 3 cache, and if the value isn't found in any of the caches, fetch it from main memory. Fetching a value from main memory takes a *lot* longer that fetching it from the level 1 cache.
A compiler can do a good job of instruction scheduling if it knows how long each instruction will take to execute, but figuring out how long a load instruction will take to execute is hard if not impossible. (On a machine with multiple cores, the execution of a load instruction can depend on what is happening on other cores as well as what is happening on the core executing the load instruction.) On the other hand, an out of order execution engine can issue load a bunch of load instructions and then start executing instructions using the result of a load operation as soon as that load operation completes. What this means is that an out of order machine can do a good job of scheduling instructions even if it has no idea of how long a particular load instruction will take.
TransMeta designed two x86 compatible processors using Just In Time code generation, which demonstrates that the approach is feasible. However, their first model was quite slow, their second model was slower than Intel and AMD's fastest chips, and the company concluded that developing a third processor would not be profitable. Since then, out of order has gotten more attractive relative to static scheduling because transistors have gotten cheaper, so I doubt that in order execution makes sense for a modern processor unless you are strongly optimizing for low transistor count over performance. Intel's Atom line has used out of order execution since 2013. ARM uses in order for its lowest end processor (the Cortex-A510), but everything above that is out of order.
Yes this. No AI is replacing an "accountant", not any time soon. But an AI might make an accountant's job easier, and thus you need fewer of them. What an accountant will eventually become is "the person who talks to the accounting AI and knows what it is saying and how to get it to count what we want counted".
This offers a narrative for how AIs might become very powerful and largely take over the society, even if they're not skynet or paperclip maximizers. Over time, more and more important decisions in our society are taken over by AIs, with humans managing them, or overseeing them, or sometimes simply collecting a rent on being the guy who gets paid to ask the AI the question and then pass along the answer. Nobody is exactly enslaved by an AI, but you can imagine a pretty smooth transition from our world to one in which more and more of the created/designed/built world and more and more of the culture are produced by AIs, more and more of the practical decisions (Should we build this bridge? Should we open a new factory? Should we impose sanctions on that country?) end up being made mostly by AIs, with some human supervision and final oversight.
As the AIs become more and more capable, the people who basically go along with the AIs in nearly all cases will end up winning all the competitions. It'll be like chess playing, where (as I understand it) human+AI was better than AI for awhile, but at this point, even Magnus Carlson should probably just do what the AI tells him.
From what I've heard from a CPA, the main function of a helpful AI might be herding bad accounting software.
If the AIs continue with their tendency to make plausible-sounding stuff that is sometimes just wrong, then the human can do quality control on the AI's output. But it seems likely to me (as an outsider to the field with no particular insight) that this problem will be solved and AIs will end up less likely to produce incorrect code/text than humans.
For code I can see it (as it's a formal system that can be easily checked against), but I'm less convinced for free text. The GPT solutions tend to respond with some kind of average of their inputs, which means that it will reliably reproduce common misperceptions, for instance.
Both for art and text, I believe we will need some kind of "reality model" before this can be solved. AI art would certainly benefit from having a definition of human anatomy.
AIs (from the current training method) also tend to generate boring well understood answers to already-solved questions. Even the stuff that we don't want the AI to tell people (like how to make meth or hotwire a car) are clearly solved - there's no new chemistry or mechanics involved.
That's great for some fields where we just want the same things processed over and over again. It's terrible in any job that requires creative answers or responses to novel inputs.
I’d like to think we have a few decades left before we’re all out of jobs but considering humans will be competing for the same limited resources on earth , by the time A.I has taken over all cognitive work humans will be considered pests by our super intelligent overlords
Will never happen. The job takeover by new technologies has always been, and will always be, succeeded by the creation of many new, previously unfathomable jobs. The limitlessness of human ingenuity renders unimportant the limit of resources on Earth.
My willingness to gamble on this hallmark economics: about zero.
Every generation since the beginning of the industrial revolution has had some technology or other that was totally going to create permanent mass unemployment. And every time, the people who suggested otherwise were dismissed as naive Pollyannas. I'll stick to the side with the multi-century history of vindication, thank you.
My main rebuttal to that: until now automation automates specific tasks. This destroys some jobs, but creates new jobs, and the increased productivity of the more automated economy makes those jobs more productive and better paid. But AI automation does not automates specific tasks: it can automate whole classes of tasks, potentially both the task that you wanted to automate in the first place, and all the ancillary tasks that the automation generated. So no new jobs are created, or in any case far fewer than it was the case in the past.
The standard counterargument here is that we're destroying jobs for low-IQ people and only creating jobs for high-IQ people.
We've got away with this so far because of where we started. In the stone age, 99% of jobs could be done adequately with a two-digit IQ. By the 1900s maybe we're down to 80% of jobs requiring only a two-digit IQ. But we are now reaching the point where ~50% of jobs require a three-digit IQ and still only 50% of people have a three-digit IQ. What happens when 80% of jobs need a three-digit IQ and there's nothing useful for the lower orders to do?
Some people argue that AI will be at least as good at taking the middle or high-end jobs going forward. It’s probably easier to make lawyers 10X more efficient than construction workers or home health aides. I think it’s a fairly convincing point.
"What happens when 80% of jobs need a three-digit IQ and there's nothing useful for the lower orders to do?"
Prostitution? Hard to imagine the robot overlords beating real humans at that any time soon.
"Introducing the new Sexbot 9000...."
"with new futanari catgirl mode!"
I agree we will find new jobs for those displaced by technology, the comment was that we should worry that a super intelligent system could see humans as a threat to survival.
The best part is that soon the AI will come up with new jobs much faster than humans could.
This only works as long as there are still some jobs that humans can do better than machines. It won't hold if AIs and robots become better than humans at every task.
Depends. Consider comparative advantage. The CEO might do a better job cleaning the office than the janitor, but it still makes sense to have a janitor.
I think we tend to assume that AI and robots will have infinite capacity, but I’m not sure that will be true any time soon.
Agreed, at least in theory. I had almost written an aside to that effect, but I removed it to avoid confusion.
In practice, I suspect once we're able to make (for example) a good AI radiologist, it'll be very cheap to spin up more instances or increase the capacity until all x-rays can be analyzed by AI.
Probably less true for robots, but even then I'd expect mass production and technological improvements to bring the prices down. It'll take longer to scale up production, but once it does, it might only be worth it to pay humans the equivalent of $0.25/hour today (made up number) versus buying a robot janitor.
Things are true until they aren't. I don't think it's appopriate, even if this has been true for automation thus far, to have ¯\_(ツ)_/¯ as an insurance policy.
I sympathize with that declaration. But what I'm saying isn't throwing up my arms with indifference. Indifference is the opposite of what I feel; I feel strongly that no insurance policy is needed. In my mind, the only contingency surrounding the entire premise is that people must be left socially and economically free to pursue their rational interests, and in that case, there is no need to fret about AI, and certainly no need to fret limited resources or jobs. I'd wager on one of these prediction markets that if AI ever became "overlords" to man, it would answer to its own overlord who would be, unironically, a man.
I understand that you have strong feelings, but if your feelings turned out to be wrong, I could not live off your salty tears alone.
One of the things you seem to feel strongly about is that AIs could never be actually more capable than "men", as you put it. Why do you feel this way?
Lol! Maybe you could live off my tears. AI will absolutely be more capable than men at lots of things. It already is, isn’t it? It takes more than competence at tasks to become an overlord over mankind.
Not an answer, just a note: stock market obviously does not think so.
If the stock market did view AI as important, how would stock prices differ from their current state? Of the top five US companies by market cap, three of them have large AI efforts or would clearly benefit from a large AI sector.
This is a misunderstanding. I did not say that stock market sees AI as unimportant. But the OP’s question is whether it is plausible that AI replaces almost all cognitive labor in near the future.
So, how would the markets look if they would see that scenario as plausible? For one thing, imho there should be a lot more venture investment into AI. But, and this is more verifiable, stocks of tech companies should move on information about AI progress, instead of traditional indicators like sales etc. And we are not really seeing that. E.g. it was recently announced that Meta developed an AI sort of able to defeat humans at a board game called Diplomacy (with heavy caveats); this seems like an important breakthrough in AI development, did it help Meta stock? Admittedly I didn’t bother to look that up, but I am guessing no, it is still in the doldrums.
And if the market believed this story, Fed raising interest rates should not have disproportionately bad effects on tech stocks, which is what we are seeing. Now, this is a subtle point since higher interest rates are going to slow investment, including investment in AI, but the fact that it affected tech companies more than e.g. oil companies is an indicator that markets still see tech investment as more speculative and uncertain than other kinds of investment. Which imho does not square with the plausibility of imminent AI takeover of, well, everything – even if AI would grow strongly but not so strongly as OP suggest, tech companies still should benefit enormously.
Now, markets are often wrong, and it is possible that they are dominated by the kind of business analysis unsuitable for assessing AI impact. But it is an important data point, nevertheless.
I see what you mean. Thank you for expanding on that point - I hadn't considered that.
"Exponentially cheaper" is probably not accurate. GPT3 cost billions of dollars. It's being created and shared with the world as part of a test phase towards building GPT4+. It's also significantly flawed, such that just about nobody would ever trust using its code over a human at this moment (best case scenario is to have a competent human review the code, which doesn't currently seem much, if any, cheaper).
If OpenAI had to pay for their own research based on marketable products, GPT3 would cost seven figures per access credential and be a miserable economic failure. Or they could charge very small sums and hope people check it out for the novelty still (which is what most people have been doing). It would likely still lose eight or nine-figure sums, maybe 10 figure.
That's not to say that a true general intelligence couldn't learn multiple fields if/when it exists, but current trends do not appear likely to break into any occupations in the next few years.
Didn't GPT-3 cost only $12 million?
I don't know how much GPT-3 specifically cost, but you also have to include the costs for GPT-2, and GPT-1, and ChatGPT, and the basic research underlying the various changes, and the costs of running a company that can research these things. It all adds up, and as long as there isn't a marketable product, those costs continue to compound. I've heard from people here at ACX that the total cost of direct AI research is well into the billions. That does not include the various companies building algorithms that aren't really AI (such as Google Search or whatever) that are more limited and actually do make money.
Why stop there? We could count all the money that went into deep learning neural nets. Or all the money that went into AI since the 70s. Or all the money spent on developing computers since the 40s.
It took a lot of time and money to get to the current level of technology. Now that we're here, it only costs pennies for GPT-3 to generate pages of code.
It's not currently at a level where it can replace programmers. It's at the level where GitHub Copilot can generate code suggestions (fairly long ones, too) and a lot of programmers find it useful.
But when it does get to the level where it can replace programmers, it'll probably still be able to generate thousands of pages of code per minute for just a few cents. That's what the parent comment meant by exponentially cheaper than human coders.
Yes, I get that and agree it's the obvious goal of everyone working on AI research.
As far as adding up costs, you wouldn't want to add in any cost center that has already offered profitable products (such as computers themselves), but only the offshoot that hasn't made any marketable products.
If/when such a product is made and can be sold, then all of the research between "clearly profitable and existing products" and "this new revolutionary product based on [new research branch]" can be evaluated for total cost verses what can be made from using the AI. The amount of money potentially available could be huge, but right now it's a huge loss.
>I'm interested in whether people think AI is going to render a lot of professions obsolete in the next few years.
No, definitely not. It may well start revolutionising some in the next few years, and it may well render some obsolete in a few decades, but things like ChatGPT, while impressive, are still a long way off being usable without human oversight for any important purpose, and even when they are adoption will take time.
The resulting code is never quite right outside simple scenarios. Fixing that takes knowledge.
One of the most exposed professions right now might well be radiologist. AI:s are becoming *really* good at interpreting x-ray images, and are already superior for some conditions. Presumably there will still be human oversight, but this could cut a lot of jobs.
I disagree. Medicine is a field in which both patient satisfaction and liability insurance will keep humans in the loop *long* after it's conclusively proven that AIs could do the job better. Even with an AI evaluating the scans, there will still be a doctor involved. We didn't get rid of doctors when we invented the x-ray (or name your new technological breakthrough that makes being a doctor easier), despite the nature of the job changing.
It's not totally clear from your comment if you're aware that radiologists already generally don't interact with patients directly? They interpret a scan, and sent a report to the patient's doctor who discusses it with the patient. From the patient's perspective, the radiologist is already nothing more than a black box the vast majority of the time.
Your comment ("there will still be a doctor involved") and the comment you're replying to ("One of the most exposed professions right now might well be radiologist") can both be true.
It's the legal and governance system as a whole that requires a human in the loop to blame. Patients may not interact with the radiologist directly, but their lawyers certainly will if something goes wrong and it comes out that the hospital was using AI.
Lapras is correct. Beyond that, humans will irrationally prefer that a "real doctor" is reviewing the scans. Due to the potential of errors and mistakes, hospitals will require that a radiologist checks off on the AI's work - similar to how self driving cars have to have a human in the driver seat even if the car has a better safety record than the human assisting it.
Also regulation. We may keep radiologists around the way some diesel trains had firemen.
What would really revolutionize software development is a chatbot that could spend hours at meetings, and then provide you clearly written meeting minutes. :D
Another interesting use would be a system of accumulating documentation and tacit knowledge in companies. Searching information in logs. Etc. This actually sounds like something a chatbot should be capable of doing.
I think that the most revolutionary thing to come out of GPT Chat is the natural language improvements - especially the ability to read human input and respond. I'm old enough to clearly remember search engines that only found results with exact matches (no misspellings, no abbreviations, no last name first or vice versa). To see an open-ended prompt that can result in natural language responses is actually a big deal. The idea of an Alexa/Siri-like assistant who really can provide you with correct information on a wide variety of topics (or take your meeting notes) may be pretty close to available now.
If someone invented a "go to meeting" chatbot, the meetings would expand so that reading the minutes would take as long as the old meetings. Human labor would remain constant.
The first jobs that AI would eat into significantly are the ones I least expected it to —the creative sphere.
Plenty of amateur/ entry-level creators are going to have a tough time competing with AI. The first jobs that AI is making an impact on are voice-over artistry, graphic design (DALL-E etc.) and copy writing.
From what I see around me, arts/ humanities graduates are comparatively the most tech-illiterate and evasive to tech. This only adds to the problem
Yes, that is a conversation that I've had multiple times, with people in the "creative" professions, specifically in advertisement and copy-writing.
They are quite surprised that automation came for their jobs first, and recognize that not only most jobs are not creative, but most people are not creative.
In general their problem is not that they will be obsolete, but that the mere existence of instruments such as GPT3 will devalue their jobs and force them to lower their rates: they will still need to to all the interaction with the client and deployment of the result, while saving a trivial amount of time using AI-generated copy, but the clients will try to pay them less because "I can do it myself with a machine". Sort of what happens to translators.
Artist jobs (outside of the super famous ones selling whatever they wanted to paint that day for millions) fill two functions. One is to actual paint/draw/whatever medium the actual art. The other is to determine what art their client actually wanted. Right now that second part is quite important, as it takes artists many hours to produce one art, and getting it wrong would be very bad for everyone involved. If an AI can produce hundreds or thousands of art examples with very minimal time, the balance shifts towards just asking for a new piece of art instead of explaining better what the client wanted up front. For many artistic functions, like the graphic design department, a CEO or VP of Marketing trying a handful of prompts in DALL-E probably takes less time than trying to explain what they want to the graphic designers. They still fill a lot of additional functions currently, but that's a big hit to the necessity of graphic designers in a marketing department.
Citation very much needed on "basically competent" at coding. Ability to write things that look like code? Sure. Have that code be correct for relatively simple problems? Yes, to my surprise. But in hindsight there's lots of code snippets to train on, especially on simple, puzzle-style questions. But extend this capability to novel, complex problems? I still don't see it, even for GPT-5 or whatever. The surface keeps getting more and more polished, but actual understanding is still lacking in the same ways as the previous iterations. It's getting harder to point out definite "tells", but they're still there.
Nope. The sine qua non for any worker, be it human or machine, is reliability. In general, where machines have already replaced men, that is their big selling point: they're not usually, at least at first, cheaper, they can often be finicky to set up and maintain, and they often require significant changes to production methods -- but they are rock-solid reliable, and that gives them an edge.
Same thing with even more advanced "knowledge" work. The single most important thing about a paralegal is -- he doesn't make obvious dumbass mistakes about the law. He may not be inspired, but he won't fuck up writing up an ordinary pleading. Same with the diagnostic role of a PA or NP, or the ability of a middle-manager type to schedule workers, or a junior code monkey turning out bog-standard database interfaces. They may not do the most creative work -- for which you'd hire the more senior star of the field anyway -- but it is absolutely necessary that they don't wander off the rails randomly, that when you assign them some area of responsibility they can execute it reliably and largely flawlessly.
But so far, this is exactly where modern AI falls down. It is not reliable. It sometimes gives the right answer, but other times confidentally says something obviously wrong, or out of left field. You never know exactly how well it will perform a given task, and even the designers can't guarantee its performance. It's like the 1980s-era joke about using Linux on the desktop, that it would be as if you flew an airline that used only Concordes traveling at Mach 2.1 and sold transatlantic tickets for $50 -- but where 1 out of every 50 takeoffs the plane would explode, killing everybody.
So that makes modern AI so far mostly a parlor trick, and very interesting from general principles -- but completely unsuited to slotting into any kind of work more complex than as an amanuensis to a entrepreneurial human. I can see it being useful to artists or writers to generate stuff that they can then re-use, polish, use for inspiration, whatever -- but will always have to check. But it can't be made part of any significantly large and complex work environment, because nobody knows if it will actually do the job on any given assignment, or whether it will make some horrible mistake. However brilliant its everyday work might be, the capability to randomly do something stupid kills the deal.
If you haven't encountered it yet around the rat-sphere, check out The Story of VaccinateCA, about a grassroots effort to collect and publish reliable information about vaccine availability, back when COVID-19 vaccines were just becoming available. It's a long read, but worth it.
https://www.worksinprogress.co/issue/the-story-of-vaccinateca/
Thank you for sharing, this was seriously impressive and heartwarming (in the sense of showcasing humanity at its best when dealing with humanity at its worst -- well, maybe not literally "worst", but you get the idea).
My meta-take is that the government might have done well to pay the pharmacies more per shot administered. That would have incentivized them to go out and round up patients, rather than just sitting passively waiting for patients to find them. That would have raised the possibility of corruption in administration, of course.
They might have also done well to not actively forbid (under heavy penalty) healthcare providers from administering vaccines to Not Officially Approved people, even when those vaccines would otherwise end up in the trash.
I can sort of see a point to the worry that not forbidding this would have bad consequences, such as unmanageable queues at the end of the day (thus generating hotspots of covid spread), or unfairness in the distribution. However, the former issue could be mitigated with pen-and-paper queues or some other implementation, and the latter seems to have happened regardless. I'm having a hard time not believing it was just a very bad decision. Is there a better argument for this rule?
This is definitely a recurring concept, and not exclusive to the gaming industry (if you skip the "funnel" label): see https://www.joelonsoftware.com/2000/04/26/designing-for-people-who-have-better-things-to-do-with-their-lives/ from back around the turn of the century, for instance, or, somewhat more recently, "Beware Trivial Inconveniences" by our host.
This is not new, though I'll grant that there are a lot of people in government (and elsewhere) who apparently have not been exposed to the idea.
My best guess is that the government internalized the notion that vulnerability to COVID is highly variable between groups. That's why it organized administration into tiers. And it was really worried that some people would find a way to jump the line, which would have compromised delivery to the most vulnerable, so it threatened dire punishments for anyone caught abetting such behavior.
For my money, the author gives short shrift to this really quite reasonable concern. But in fairness, I don't think the government understood the *flood* of shots that would become available in short order. It was acting as though the problem to solve was making the most of scarce vaccines, when the main problem was rapid and efficient administration of as many shots as possible.
Hooking onto this to recommend more of patio11, the man behind the initiative.
Among his other cool ideas and pieces, he writes a great newsletter on banking and the financial services industry - Bits about Money (https://bam.kalzumeus.com/)
https://experimentalhistory.substack.com/p/the-rise-and-fall-of-peer-review?ref=the-browser
Great article, loved the framing
I have been wondering whether to take fever-reducing medication (e.g. paracetamol) when I have flu. On the one hand, it feels good, reduces flu symptoms and so on. But I wonder if it has any effect on the duration of the sickness, preventing fever means preventing my body from fighting it, in theory extending the fight for a couple more days.
So is my conclusion correct, that if I'm willing to suffer the symptoms, it's better not to take such medications and let the body fight it off with fever?
I've wondered this as well. Are there certain illnesses for which fevers help and others for which they don't?
Speaking as not-a-medical-doctor, the accepted medical wisdom is that the fever is more dangerous to you than the disease. Preventing fever doesn't mean preventing your general immune response; fever is just one reaction to an infection, and lowering the fever may or may not reduce the duration of your illness. It's just there because, apparently, humans' survival rate was better with the fever than without it, on average, in the ancestral environment. Also see this for why you wouldn't assume your immune system is optimized for the modern environment: https://slatestarcodex.com/2019/08/19/maybe-your-zoloft-stopped-working-because-a-liver-fluke-tried-to-turn-your-nth-great-grandmother-into-a-zombie/
It seems to depend on what kind of infection you have whether reducing the fever shortens or lengthens the duration of your sickness: https://pubmed.ncbi.nlm.nih.gov/11130213/ In any case, I'd err on the side of caution, and not let fever wrack your body worse than whatever you've caught.
From your pubmed link for 2 out of 3 infections tested controlling fever lengthened the disease. Roughly (I think) for flu and food poisoning you’re better off not controlling fever and for Rocky Mountain fever you are better off controlling it. For me this compiles down to “mostly don’t control fever”
I don't know how representative the studied examples are here. This is just a single paper illustrating that in some cases fever helps recovery, but not others, not an indication of whether it is, in sum, more often better to not control fever or not. Fever can be very damaging to the body so the question is if shortening the disease is worth that tradeoff.
I agree this is just one paper etc but it is the opposite of supporting "the accepted medical wisdom is that the fever is more dangerous to you than the disease." Do you have references that support that claim?
If I did, I'd have said that it was official WHO advice or something, wouldn't I? This is the advice I've repeatedly heard from doctors, no more and no less. Treat it as you would any other accepted medical wisdom, like putting a cold compress on a contusion or disinfecting a wound with iodine.
I did some reading online and I think it’s time to update your false belief. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7717216 https://www.mayoclinic.org/diseases-conditions/fever/in-depth/fever/art-20050997
A priori, I would think that fever is adaptive against infectious diseases. Pathogens are quite temperature sensitive, which is one of the reasons animal diseases can't always jump to humans.
The caveat here is that evolution favors surviving and reproducing, which often but not always align with longevity and good health. So it might be possible that fevers are mostly useful in the worst-case scenarios, which are much less likely to arise with modern medicine.
I think someone tested it once and found a minor benefit from not reducing the fever in some minor ailments like colds. Personally I let the fever run, but that's more as a point of principle, and I don't believe there's any strong evidence for it.
That's been my gut feeling, with one exception - I try to protect my brain. My vague theory being that evolution probably hasn't had enough time to "care" about our modern uses of intelligence. So I roll with the fever as much as possible - dressing warm, staying under lots of blankets, hot showers - but I also try to keep cool packs on my head. I think it works out pretty well, but I'm not a doctor at all.
Very high fever is dangerous, but a 101 fever is just miserable.
An excellent post by Razib Khan on why "transgenerational epigenetic inheritance" isn't important for humans: https://razib.substack.com/p/you-cant-take-it-with-you-straight
It's a bit of a scoop of my own epigenetics post series (which has been delayed as an indirect consequence of FTX) but I hope to expand upon it a lot.
Oh wow. That's really informative. I had naively accepted the "recent advances in epigenetics demonstrate a biological basis for generational trauma" idea. I think "epigenetic generational trauma" lived in my mind as a kind of "biological determinism but somehow not racist because reasons", and I was pretty uncomfortable with that. This clears up a lot of muddle.
You may be interested in my second post in the series, https://denovo.substack.com/p/epigenetics-of-the-mammalian-germline
Why is Long Covid so underfunded. Why do doctors dismiss it? Do doctors dismiss it because it's underfunded?
I think part of this is because it's hard to distinguish "SARS-CoV-2/COVID-19 HAS LEFT PERMANENT SCARS ALL OVER MY BODY!!!!!!" (Long Covid) from "random person gets COVID-19 and coincidentally gets a completely unrelated illness sometime later" (statistical noise)
Is it underfunded? Congress directed the NIH to spend $1.15 billion to study the long-term effects of COVID.
https://recovercovid.org/faqs
"In December 2020, Congress directed the NIH to devote $1.15 billion to studying the long-term effects of COVID, including Long COVID. The NIH established the RECOVER Initiative in February 2021 and it has allocated funds provided through the American Rescue Plan (ARP) Act of 2021 (Sec. 2401) to support this effort. RECOVER awards were made to more than 30 groups, including researchers from hospitals, health centers, and other places. These researchers, along with people affected by Long COVID and representatives from advocacy organizations, worked together to develop the main protocols that serve as the study plan for RECOVER studies. To help manage the administration of the RECOVER initiative, Research Triangle Institute (RTI) International was selected to serve as the Administrative Coordinating Center (ACC)."
Hmm I think ppl got funds if they reported adverse effects post vaccine. 500m went to NYU. I don't see long Covid clinics yet being helpful. I don't think the money has really been spent yet. In the meantime the drag on economy will reach trillions. Ppl will see when staring down the barrel of a gun only.
In general, diseases where exhaustion and/or pain are primary symptoms get ignored.
Well nobody wants to touch pain now after the Sacklers suit, and oxy was great for pain but high dependence, but dependence can be managed by doctors but after suit and what has happened its just a territory that is not worth them getting into, leaving people in pretty bad straits. Also its a panolopy of symptoms under autonomic disregulation - strange heart functioning, diabetes on the rise, headaches/migraines and other neurological issues....
Pain as a symptom of long covid does not seem to be on the list (apart from muscle/joint pains). If you are claiming mystery pains that require opioids as treatment, of course you are going to be treated with caution by doctors. "I need oxycontin for my long covid" is going to be a huge red flag, even if the doctor does accept that you have post-covid syndrome.
pain is typically part of an inflammatory process. What I said is that pain medication isn't really being pursued after the oxy blow up - you literally made my point, people who are in pain and are looking fo solutions are being conflated with people who are drug addicts. Saying you need medication for pain should not be a red flag of any kind. And headaches/migraines are being featured a lot post covid, during covid, and with long covid. Furthermore neuropathy disorders, which can be quite painful, the list goes on and on. But yes, pain is a symptom. In some cases it is neuroplastic and in some cases indicates an inflammatory process, etc.
This is just a tragic tradeoff that the universe (or maybe our current lousy technology) imposes on us: The best painkillers we have are very addictive. We can move the pendulum toward "fewer people in untreated pain, more addicts" or toward "fewer addicts, more people in untreated pain." Both are terrible outcomes. We did the pendulum swing toward "less pain, more addicts" and the result was a huge surge of opioid addictions and a moral panic and gazillion dollar lawsuits, so probably it will be a generation or two until the pendulum swings back. And depressingly, it seems like roughly *none* of the public discourse about this, from journalists or politicians, notes the inevitability of the tradeoff. We're all in a moral panic about opioid addictions now, and that would be off-message.
As best I can tell, most of the deaths from overdose are caused by fentanyl becoming cheap and so mixed into a bunch of illegal drugs by people who have no business at all doing life-critical mixing of drugs. We'd probably have more addicts but a lot fewer overdose deaths if you could just buy a bottle of oxycodone over the counter. But again, it's a tradeoff--deciding where on the tradeoff curve we should be is ultimately a matter of values.
I didn't mean to say that pain was especially part of long covid-- that's what I meant by "in general".
Is there any symptom or group of symptoms that isn't indicative of Long Covid? I'd put a small sum of money that LC is this era's mystery illness panic. It's amazing that there's anyone left alive after PVC contamination, Multiple Chemical Sensitivity, Morgellons et. al.
Spoken by someone who probably has zero experience with chronic illness
I also bet if you were in the world trade center you would have completely ignored directives to stay in your office and immediately knew to run down the stairs and get out---incredible how much we know about things we know nothing about!
Come down off your high horse. You have no idea who of us on here have chronic or long-term illnesses, and you are presenting demands and uncorroborated statements as if they are self-evident facts.
You may be ill yourself, I don't know, and if so then sympathies to you. But simply stating "Doctors don't care about long covid, why isn't something being done?" is not enough, particularly as there *does* seem to be on-going research.
That the list of long covid symptoms also apply in other illnesses, and that there is going to be a long time before anything can fruitfully come out of research on long covid, is unfortunate facts of reality. Making snarky remarks about the World Trade Center isn't helping your case or making you sound like anyone other than "I feel that I am sick, why won't doctors believe me?" when you have nothing actually wrong with you.
my point with the world trade center was simply to say that people tend to think very highly of their own response to things that they have not experienced.
Nah, your point was that you don't like me and wanted to communicate your dislike.
As a younger person, I would have obeyed instructions. However, my years of experience, particularly involving emergency response has taught me to trust my own observations v. what someone is reading out of a manual. And since your scenario would have been prior to the big ICS redo, those commands would have been even more ill-informed.
All people do? Oh, do you do that? Or is it just the people whose opinions you dislike?
What, exactly, did you think this response was going to accomplish?
"Oh, I have X, therefore all purported diseases must exist, even though there is a long history of bogus disorders which have fueled innumerable books, daytime TV interview programs, websites and medical consulting practices?"
Listen, I had a horrible post viral syndrome myself 20 years ago after the flu. It lasted 3 years. I have no doubt at all that post-viral syndromes exist and wreak havoc with people’s lives. But I am moderately skeptical about a fair amount of what’s claimed about Long Covid. The situation is really hard to read , the research results are all Over the place, and it seems like everybody has an ax to grind. In a messy confusing situation like this omebody can be skeptical without being an ignorant heartless jerk.
One thing to note: post-viral syndromes are a known thing that happens sometimes. Occasionally, you get a case of the flu or some similar virus, and you end up with lingering problems for months or even years--racing heartbeat, occasional fevers, occasional fatigue, etc. I don't think this is well-understood, though. My amateur impression is that it's probably some pathological thing happening in the immune response where your body doesn't turn down the response enough after the infection is cleared or you've ended up making some cross-reactive antibodies or T-cell receptors that are causing some kind of autoimmune thing to happen.
I bet a bunch of long covid is just millions of people getting a viral infection at the same time and some fraction having a post-viral syndrome.
But covid also messes with clotting, and you can imagine ways for that to occasionally cause all kinds of nasty problems. And of course, anything that makes you *really sick* (like hospitalized for a couple weeks) is very likely to leave you with some long-term damage.
Again - even if you think the pain is being created by the brain, it is still very much real (i.e., the can pick it up on functional MRI's). We know less about the brain than we know about the galaxy, so its amazing how many deities are in the comment section, really
If it's being created by the brain, then it has nothing to do with Covid, "Long Covid" is a total misnomer for it, and trying to apply cures or treatments based on or related to Covid will not work -- just as you wouldn't try to cure hysterical paralysis with spinal nerve grafts. You cure hysterical paralysis with a little rubber hammer which proves to the patient that she's not really paralyzed, just riddled with anxiety and neuroses. (This is incidentally why you've never heard of anybody having hysterical paralysis -- it's too easily defeated now, and that's why the replacements are vaguer, more subjective ailments that doctors can never get a good bead on.) Long Covid has to be cured the same way as fibromyalgia and all the rest: by forcing the patient to accept that the symptoms are just her externalizing her anxiety as a fake physical malady. AFAIK we basically don't know how to do this reliably at the moment; the only method that seems to work in even a small percentage of cases is brutal shaming, which our present society is a bit too gelatinous to allow.
There is a group headed by bob Patterson that is able to identify certain elements like cell retention of spike protein to accurately determine who is experiencing long covid and who isn't. The treatment for it is an anti viral and statin. I mean its very easy for you to argue that its just a placebo effect -but you can literally say that about every single drug which would render this entire conversation completely moot. HPV is a virus - causes cervical cancer. A lot of viruses cause downstream problems. You are talking about virus that has been around for two years like you have the noble. prize in science and have a decade of research under your belt. You have no idea but its plausible for a viral infection to cause some problems in certain people as many viruses do. A lot of viruses kick of autoimmune issues. These things are no longer debatable. The reality is nobody has the answers yet.
"You are talking about virus that has been around for two years like you have the noble. prize in science and have a decade of research under your belt."
And you're talking about it like an absolute textbook according-to-Hoyle hypochondriac crank, but you don't see me descending to personal insults.
"These things are no longer debatable. The reality is nobody has the answers yet."
These two sentences contradict one another.
I haven't said a word of hypochondria or any reference to my personal experience but thanks
"There is a group headed by bob Patterson that is able to identify certain elements like cell retention of spike protein to accurately determine who is experiencing long covid and who is experiencing long covid and who isn't." What does that mean, exactly? So if you have all the symptoms but he can find spike protein and other element then you really *don't* have Long Covid?" What's his explanation for why those people have all the symptoms. Also, if his statin+ antiviral treatment was effective for many sufferers seems like we would have gotten the news about it by now. I have actually been hunting online for effective LC treatments, trying to help a friend of mine. From what I can see, nobody has found some central thing wrong in the bodies of people with LC, much less come up with a treatment that targets the central thing that's wrong. The LC clinics offer piecemeal treatments -- cognitive exercises for brain fog, respiratory therapy for shortness of breath, etc. But in any case, if you have a lot of confidence that bob Patterson have a treatment that addresses the core cellular thing that's wrong, why are you so worried about LC and LC patients being neglected? If Patterson's got that, it's a game changer.
"I mean its very easy for you to argue that its just a placebo effect -but you can literally say that about every single drug which would render this entire conversation completely moot." Sure you can say it about every single drug. You can call any drug a placebo, even penicillin. However, it is quite easy to test whether a drug is effective or a placebo, so calling a drug a placebo does not render conversations moot. By the way, just looked up Patterson. He has a whole treatment machine going, which is quite expensive for patients because none of it is covered by insurance, since it's an experimental treatment. However, he has not done a randomized control treatment study yet. (That's a study where you randomly give some people your drug and some people sugar pills, and look to see if the people on the drug do better). RCT's are easy to do. The fact that he hasn't done one, yet is actively treating thousands of people, and making millions doing it, is grounds for just writing this guy off, end of story.
I mean....it's not like the vaccines which were distributed to EVERYONE practically speaking, went through the most rigorous studies and we still took those. Ps I just did the patterson test.
"the can pick it up on functional MRI's"
Please direct me to a link about this claim, because it immediately smells like our "Real Medical Doctor says Ivermectin is miracle cure!" videos I've seen being touted as "don't believe the fake medical establishment, here's the 'real deal'!"
Googling it brings up a Swedish study about neurological changes to the brain. No idea if this is legit or not:
https://bmjopen.bmj.com/content/11/10/e055164
"Results Twenty-five patients (71%) had abnormalities on MRI; multiple white matter lesions were the most common finding. Sixteen patients (46%) demonstrated impaired neurocognitive function, of which 10 (29%) had severe impairment. Twenty-six patients (74%) reported clinically significant fatigue. Patients with abnormalities on MRI had a lower Visuospatial Index (p=0.031) compared with the group with normal MRI findings.
Conclusions In this group of patients selected to undergo MRI after a clinical evaluation, a majority of patients had abnormal MRI and/or neurocognitive test results. Abnormal findings were not restricted to patients with severe disease."
There also seems to be an American study:
https://press.rsna.org/timssnet/media/pressreleases/14_pr_target.cfm?id=2381
And an English one, which is good data because they scanned the same group of people before and after having covid, so were able to pick up on any changes:
"Discussion
This is to our knowledge the first longitudinal imaging study of SARS-CoV-2 in which the participants were initially scanned before any of them had been infected. Our longitudinal analyses revealed a significant, deleterious impact associated with SARS-CoV-2. ...Note that these structural and microstructural longitudinal significant differences are modest in size—the strongest differences in changes observed between the SARS-CoV-2-positive and control groups, corresponding to around 2% of the mean baseline IDP value. This additional loss in the infected participants of 0.7% on average across the olfactory-related brain regions—and specifically ranging from 1.3% to 1.8% for the FreeSurfer volume of the parahippocampal/perirhinal and entorhinal cortex—can be helpfully compared with, for example, the longitudinal loss per year of around 0.2% (in middle age) to 0.3% (in older age) in hippocampal volume in community-dwelling individuals. ...Our statistics also represent an average effect; not every infected participant will display longitudinal brain abnormalities."
So not everybody who gets covid will have the brain changes.
*If* this is legit, then that's your problem right there: changes to the brain can't be easily or simply undone, if at all. We're still looking for something to help with Alzheimer's, after all.
Research is ongoing, but it's going to be slow. And most doctors are probably not going to send someone off for a brain MRI simply on the grounds of "I have a cough, muscle pains, and am recovering from covid".
You sound like you want a cure right now, but that's not going to happen. It is going to be years of study to see if the brain changes are permanent, if/how they can be alleviated or even prevented, and what medications if any help improve quality of life.
Anecdata != data, but I can tell you firsthand that a psychosomatic condition caused very pronounced arrhythmia for me, to the point that cardiologists were considering heart surgery (ablation).
How did you find out what was actually going on?
Mostly a coincidence. At first I assumed that it was all related to a heart condition that runs in the family. But the symptoms stopped matching very well at some point. I had constant headaches, so was eventually sent to a neurologist, who prescribed me with medication for that. That didn't help with the headache, but helped a lot with all other symptoms (the aforementioned arrythmia, fatigue, etc).
At this point I was just doing the scattershot thing of following every lead and saying yes to every possible referral to yet another specialist. That eventually led to a diagnosis. I also had a pretty good, open-minded GP who helped the process along.
I had said they can pick up pain on functional MRIs proving that its a real experience. I do not know of any functional tests for long covid, simply post moretum autopsies showing spike protein in the brain in the like., but my whole point at the start of this thread is there should be more funding for long covid.
There's nothing fake about psychosomatic diseases. There are potentially a lot of fakers or deluded people, but that's not the same thing.
Sure there is, there's lots fake about them. Hysterical paralysis is a fake paralysis, you're not really paralyzed. It's a real problem, but a fake paralysis. Electromagnetic hypersensitivity is a fake allergy, you're not allergic to electricity at all, just a guy with a brain problem situation. You're faking the disease as an external expression of your scrambled endocrine system, or whatever the hell part it is that's on the fritz.
mmm, I think you're using "faking" in an inconsistent way. If a person is unable to move, that's real paralysis, period. If a person feels pain, that's real pain, period.
Usually, when people say "faking", they mean someone is deliberately and deceptively presenting some symptom they don't actually have.
I disagree with this, unsurprisingly. If you're convinced that you're paralyzed (well, in this case, that one of your limbs is), but actually you're not and the doctor can demonstrate to you with a simple maneuver that your arm/leg works fine, then that's a fake paralysis. You may not be *faking* (a word I did not use) it, but it's still fake, the same way that if someone else feeds you a counterfeit bill and you don't notice that it's bogus, you're still spending fake money when you try to use it in turn, even if it's not deliberate. I don't see any inconsistency here.
Sure, we're just using words differently, so in some sense neither of us is wrong. But fake has a very clear negative connotation, and it does call up association with fakers in most people.
Can you try to empathize with how tiring and hurtful it is to be afflicted with one of these conditions, and being flippantly dismissed over and over again? Even by the doctors who are supposed to be treating us, but instead dismiss our conditions as "fake" (and thus not worthy of sick time, medication, etc). The conflation of concepts here leads to real damage.
In particular, if your convinced you're paralyzed but there's no physical problem, we'd *really* like to avoid having some neurosurgeon try operating on your spine to fix the nonexistent problem and maybe accidentally create a bunch of physical problems to go with your psychological problem.
The problem with the way you are talking is that there are at least 2 ways in which paralysis can not be real paralysis: (a) person's leg is not in fact paralyzed, but they genuinely experience it as being so; (b) person's leg is not paralyzed and they know it is not, but they are lying. Those are 2 quite different situations, and it makes sense to use different words for what's up. Also, the word you are using for both, "fake," has a negative connotation; it generally implies that the misrepresentation of something has been done knowingly. If we say a Titian is a fake, we mean that a forger painted it and claimed that Titian did. This implication seems fair enough for situation (b), but not for situation (a).
thank you vitor, yeesh
Some cases surely are psychosomatic, but I'm pretty sure some of the cases involve some actual physical thing happening that has messed the person up. But I also suspect strongly that multiple different things are getting classed into "long covid," and you can imagine ways that might make research on it pretty hard. If 40% of the cases are psychosomatic, 40% are post-viral syndrome that sucks but will eventually resolve, 10% are lung damage from severe covid, and 10% are neurological damage from small blood clots in the brain or something, it's going to be tricky to learn much until those causes are somehow separated out.
Yes, and note that the insanely high statistics like "10% of people have long covid" are a scientific criterion optimized to capture every *possible* case.
>Why do doctors dismiss it?
Because a lot of people are hypochondriacs, or anti-vax crazies, or just malingerers and all those people make it hard to sort the wheat from the chaff when dealing with stuff like this.
I have a lot of "long COVID" symptoms. but I would never say I have it because I thin it is way too hard to disentangle those possible symptoms from the symptoms of the weight I put on and exercise I forwent, and of life changes I made (plus I am 2 years older). Many people just latch onto whatever trendy explanation soothes their feelies best and say they have "chronic Lyme" or whatever.
Well, it might be hard yes, but indeed even if you believe the root cause is from the brain and nothing else, no lingering viral load, spike protein, or other immune suppressant, then you are left with neuroplastic pain and symptoms, which are produced by the brain, but still very real. I'm glad you can attribute your problems to other things, but if you long covid was debilitating you'd be looking for answers, I think its sad to live in a world where we assume people are just lying, for what purpose? We are talking about 25M people in the US who probably have long covid, and they are all being gaslitt. I'm literally writing a satire about it because its a zombie movie where the zombies are being called zombies and they aren't.
Because the symptoms correlate with a billion other things, and it's like other claimed illnesses like Chronic Fatigue Syndrome etc. You feel tired, lacking in energy, breathless, minor exertion knocks you out - yes, and?
I've had lingering symptoms since my bout of covid (including getting one damn respiratory infection after another in a chain). Is that 'long covid', is it post-viral syndrome, is it something to do with my general state of health and existing pre-conditions?
Here's the NHS website on long covid symptoms:
"Signs and symptoms after coronavirus can be different from person to person. The most common ones include:
cough
fatigue
breathlessness
muscle and joint pain
sleep problems
loss of smell or taste
low mood
brain fog, loss of concentration or memory issues (cognitive impairment)
anxiety"
Well let's see: cough, fatigue, breathlessness, muscle and joint pain, sleep problems, loss of smell and taste and so on are also common symptoms of colds, flu, viral infections, vitamin deficiencies, poor diet/sleep/lack of exercise.... it's not like there is a nice, neat set of discrete symptoms that can only be long covid.
That's why doctors tend to go for "yeah it takes several weeks to get over covid/viruses" when someone shows up and says "I had covid, I was clear four weeks ago, but I still don't feel better" instead of going "ah yes, you are suffering from long covid!"
https://www.nhsinform.scot/long-term-effects-of-covid-19-long-covid/about-long-covid/what-is-long-covid
"Most people's symptoms of coronavirus get better within 4 weeks. But for some people, symptoms can last longer, or new ones can develop. Symptoms can also change over time and can affect anywhere in the body.
Healthcare professionals may refer to long COVID as:
ongoing symptomatic COVID-19 (4 to 12 weeks)
post-COVID-19 syndrome (over 12 weeks)
As this is a new condition, understanding is developing all the time. Experts are learning more about how long symptoms will last, and it'll vary from person to person. There can be different symptoms which often overlap. It can't be said exactly how long coronavirus symptoms will last. The reassuring evidence is that symptoms improve over time in most cases.
Advice is based on:
evidence from research so far
experience and expert knowledge of healthcare professionals
Research is ongoing, so advice may change when its results are known."
And yeah, some people are hypochondriacs/very anxious about health problems, who will imagine they are suffering from long covid and demand specific treatments and get very angry when they feel they have been dismissed.
That's great if we reach herd immunity but we aren't...so....symptoms may improve just to get walloped again? Yes, it's a tricky business but it's not impossible to categorize long Covid into four or five specific fix presentations and research the cause. That's why it should be properly funded. There are many hypothesis to test from all the clinical evidence.
Depends on what you mean by "long covid"? By some definitions, I had it, but that was just because my first covid infection (early 2020) did a number on my lungs, and it took about 6 months to fully recover. I don't think there's much a doctor could do about that.
I'm not sure where to put this, so I'm just going to describe what I believe to be my case of longish covid.
In 2020, my sense of taste got weird. My sense of smell was unaffected or possibly a bit stronger.
Tastes were generally weaker, and what I suspect was mediocre pork tasted actively nasty. What's more, the same food could go from not very good to mediocre to very good in the course of 24 hours.
It seemed to me in a later phase that if something tasted good, it probably was good, but if it didn't taste like much, that might be a problem at my end.
There's been gradual improvement and my sense of taste is approximately back to normal except that I may have lost some capsaicin tolerance.
I did not seek medical help-- it didn't seem like much effective was available. It didn't quite match the usual accounts, but the timing is suspicious.
As for the general point that people talking about exhaustion are failing to allow for deconditioning and being a little older, the exhaustion seems too extreme for that. We're not talking about having to go from a ten mile run to a five mile walk, we're talking about barely being able to function, or not being able to do ordinary activities at all.
Yes - the idea that aging can account for even a 20% collapse of health let alone 100% or more that people are experiencing is just not science based. I'm glad that you seem to have recovered. A lot of people do seem to clear long covid within a year, but of course, some dont' and there is also the issue of reinfection.
At its worst, just after the fever broke, I was literally out of breath from getting up from the couch and walking to the kitchen. Literally gasping for breath. There is NO WAY that was normal. I slowly got better from there. Now I'm more of less back where I started.
There's nothing quite like that feeling of having a hard time bicycling up a hill, feeling down about yourself and out of shape, concocting negative stories in your head, and then realizing that your bike got stuck in a high gear. Unstick the gear, and suddenly everything becomes so much easier, and you realize that all that self-talk was bullshit.
I've had the blaming myself for difficulties with biking, though in my case, it was tires gradually flattening.
There’s a lot of uncertainty over how much of what’s called Long Covid is in fast a post -Covid syndrome rather than a miscellany of symptoms caused by other things. Some pretty decent research suggest that’s the case, other pretty decent research suggests that it’s not. It’s also not clear if Covid is worse than thing s like flu and mono as regards how many people have a post-viral syndrome and how disabling the syndrome is. M(I myself had a syndrome exactly like long Covid except for no shortness of breath 20 years ago after the flu. It lasted
dor about 3 years, and during that period my average feeling of physical well-being, which previously had been about an 8 on a 1-10 scale, went down to about a 2. It was just miserable) there is decent research for both points of view.
So in that case, one could argue it needs funding. It's like there's a 50/50 chance on gd do better to walk through life believing than not. (Life of Pi ending)
It is not 50-50, it is a mess. It is clearly something that can be settled with research. It is not an unanswerable question. If I had followed the research more closely I might already think I had a pretty clear idea of what a fair reading is. As it is, I am inclined to think that genuine Long Covid syndromes are much rarer than the hype suggests. A recent study I read followed a bunch of people, half of whom had Covid and half of whom had another respiratory illness. 3 months out, those who had had flu, RSV or some other similar thing had slightly*more* symptoms from the Long Covid list than those who had had Covid. I didn’t note any ways the study was badly designed. Seemed like a good study to me. Seems to me that a fair minded person reading it is going to have some doubts about how large and unique a problem Long Covid
I think that study had 50 ppl in it. Classic long Covid versus post viral issues may be different things in any event but ppl are getting massively disabled either way and all I said is we should be investing more into funding (this was my original point many comments ago)
You may be right about number of subjects. I’m not in a place where I can look up that study. However, the stats they did indicated that the result could be trusted I don’t know whether you’ve had a stats course, so here’s what that means. Theoretically any study can have invalid results just because you happened to get treatment groups that were not representative of the population. However, if your groups are huge that’s very unlikely to happen by chance. If they are small, say 3 people in each group, it wouldn’t be a bit surprising, right? Statistical tests of significance take all that into account. They basically impose much stricter requirements on studies with few subjects. — you have to find much larger effects for them to count as genuine evidence. OK, so 50 is not a lot of subjects, though it’s not tiny. But apparently the results they found passed the stats test, which would have taken into account the fact that there were not all that many subjects.
Seems to me finding should go into research. Maybe that’s what you had in mind. That is much less money than would be involved in a program operating under the assumption that Long Covid is a unique, serious and widespread disease. Under that assumption we should have government funded treatment centers, tax breaks for the sick, possibly stipends for them, , requirements for employer accommodations, etc.
This was not the study I had in mind, but a similar one with similar results. It had 1000 patients. https://www.medpagetoday.com/infectiousdisease/covid19/102011
New movie to come out in 2023 about Robert Oppeheimer and the Manhattan Project, titled "Oppenheimer".
Trailer here: https://www.youtube.com/watch?v=LZq9kSRwfkg
Starring Cillian Murphy, and if the voice-over is him, I have to say I'm impressed with his American accent. Real Americans may disagree 😀
I'm American, and it sounds pretty good to me.
I also looked up a recording of the real Oppenheimer to see which American accent he had. His voice reminds me of Mr. Rogers (from an American children's show).
Cillian Murphy is great, so I will probably watch unless it gets atrocious reviews.
American here, the accent sounds good to me. Various southern accents are hard to do well, but this neutral newscaster-y one perhaps not as hard. The trick it seems like for British actors is to not overdo the flat or twang or how hard the consonants are hit, but he manages a lilting quality here while still sounding American.
Looks good. Soundtrack feels like it’s going to break into the opening theme of Star Trek but never does.
Regarding the previous ACX survey, the philosophy survey was the most interesting part of it. Did no interesting results come of that?
"In October the SEC charged celebrity Kim Kardashian with touting digital tokens on social media without disclosing the remuneration for her endorsement. (She settled with the SEC for $1.26 million without admitting or denying its findings.)" (WSJ 12/19/22)
It would really be a tremendous service to the world if the FBI/NSA/CIA took some time off from constantly fucking the dog to do a big sweep of the whole online "grift" world and just stick a couple tens of thousands of these people in prison. Would really help society in the long run, and reinforce that there are costs to rampant fraud. It is a huge and non-productive waste of society's resources the number of people whose full time job is stealing from suckers.
Like Logan Paul should probably be tried (and convicted) for capital crime, along with that whole ecosystem. You think I am kidding, I am not.
Strongly agree with this. Getting them to shitcan the ubiquitous surveillance of American citizens unfortunately seems impossible, so they might as well do something socially positive with it.
This is tacky of me, but I can just imagine a tv series about huge numbers of celebrities in prison.
What would the world look like if you could only see X-rays?
What would it look like if you could only see gamma rays?
Dim. Our eyes evolved to see the wavelengths that are most common in our environment.
Well, there's also the fact that the energies of visible and UV photons interact with valence electrons, which is kind of necessary if you're going to have a meat-based biology.
I'm gonna nitpick here because there's a really cool detail about visible light and the absorption spectrum of water: https://en.wikipedia.org/wiki/Electromagnetic_absorption_by_water#/media/File:Absorption_spectrum_of_liquid_water.png
There is a lot more ultraviolet and infrared light in our environment than we're able to see, but it's not exactly our environment that we've evolved to see in. That would be the ocean. There's a valley in that absorption spectrum that corresponds exactly to the middle of the range of visible light, which any animal evolving eyes would have had to target to have the best chance at seeing through murky ocean water (and, to a lesser degree, eyes made of water). We seem to have inherited that spectrum even though there are many more wavelengths that can easily be picked up through the air.
If you are talking about all else being equal, not very interesting because there are few natural sources of xrays/gamma rays to begin with that could illuminate the world.
If we are talking about the radiation being abundant enough so there was an evolutionary incentive to develop vision for it, still not very interesting because that amount of ionizing radiation would inhibit the formation of complex organic molecules necessary for life.
What sources of x-rays and gamma rays are present on the Earth's surface?
Any naturally occuring radioactive material produces radiation on decay, including gamma and x-rays.
Specifically, there are five major types of radioactive decay: alpha (helium-4 nucleus emission), beta+ (positron emission), beta- (electron emission), gamma (direct gamma ray emission), and electron capture (exactly what it says on the tin. Any given radioisotope will typically only do one or two of these, although decay chains (several successive decay steps as one radioisotope decays into another radioisotope, before eventually reaching a stable configuration) often hit most or all of them.
Gamma decay of course emits gamma rays directly, and beta+ decay emits a positron that will usually soon find an annihilate with an electronic, releasing their combined energy as a gamma ray. Alpha and beta- decay don't emit EM radiation directly, but they do release high-energy charged particles whose kinetic energy can generate x-rays when they hit something.
I don't think electron capture generates x-rays or gamma rays in any meaningful amount: just capturing the electron an releasing a neutrino that will very probably pass through ordinary matter without interacting with it (beta+ and beta- also emit neutrinos, but they're often ignored because the electrons and positrons are much more interesting). But I just found out about electron capture in this sense a few minutes ago when I looked up some stuff to confirm my explanation of the details of the above, so my confidence in that is on the lower side.
In addition to radioactive decay, other major terrestrial sources of x-rays include lightning strikes and dentists.
It seems to me that software development is moving in the direction opposite to "division of labor".
Twenty years ago I would naively expect that the growing field (in both budgets and complexity) of software development will allow lots of obscure specializations. But what I observe suggests otherwise.
The idea of "full stack developer" goes again separating database from business logic from web page development from design. In "scrum", team members are officially treated as replaceable, so one day you design a database table with partitions that will contain millions of records, and the next day you tweak stylesheets to move a button one pixel leftwards on a mobile screen.
With "devops", programmers and technical support become one; and with "cloud" they are further united with network administrators. The guy who partitions database tables and tweaks stylesheets is now also responsible for designing and implementing network security. And he is supposed to be simultaneously a specialist on all of that.
I can imagine the reasons why companies could prefer this. With the everything-specialists it seems like you can hire fewer people and save money. (Not really, because you need to hire multiple everything-specialists anyway. Might as well hire a few specialists instead.) Or maybe the idea is that everyone is replaceable because everyone has all skills and knowledge, and this should keep the salaries down. (But does it actually work? Companies are complaining about lack of developers, which is probably made worse by the need to know everything, and thus the salaries are growing anyway.)
This seems to go against the established wisdom about the "division of labor" which still seems generally accepted, so I wonder if any economist has noticed and studied this paradox.
Also, I am curious whether this trend is specific for software development, or whether it exists in more, maybe even all professions. (I mean, requiring people to do what was previously multiple specializations. Not because you openly need to save money, but because you explicitly promote the belief that a *true* specialist is a specialist on everything.)
You are kind of avoiding the main argument from the "cross-functional team"-people: having lots of specialists that sit in silos creates a huge communication overlay, that eats all the efficiency gained by specialization and then some. This argument is largely true in my experience.
The argument for a multifunctional *team* is sound. I have seen places where network administrators were on a different floor, and web designers in a different building, and yes the communication was extremely slow and frustrating. I understand why it would be a good idea to make them all sit in the same room and talk to each other.
But abandoning the specialization completely, that seems like going too far.
Also, in my experience, the team members naturally try to separate into roles they feel most comfortable with. If you have five "full stack dev sec ops" on the team and leave them some autonomy, you will probably soon find out that one of them makes most of the database query optimizations, another one writes most of algorithms, another configures the deployments scripts, another writes most of automated tests, etc. Apparently, this does not slow them down on a normal day. (May become a problem when one of them is on a vacation, and his part needs a quick fix.)
Developing specialization and a division of labor particular to your team is part of scrum. That's part of why you're supposed to keep scrum teams together for a long time rather than constantly rotating people from one to another, and build them with a balance of skills. You're only supposed to be good as a generalist to the extent that it's usually better for you to do something off your specialty *with occasional help from other team members* than sit completely idle while waiting for others to be done.
> That's part of why you're supposed to keep scrum teams together for a long time rather than constantly rotating people from one to another, and build them with a balance of skills.
Thanks, I didn't know that! And yes, that makes a lot of sense.
(And, as usual, many companies doing "scrum" completely ignore this part.)
Everything good is part of scrum. Everything bad is not. Everybody who doesn’t like scrum is doing it wrong.
Here is the official definition of Scrum: https://scrumguides.org/scrum-guide.html
Having an official definition allows us to distinguish between "this is written in the book" and "this is what my company does, but it is not written in the book". And yes, it often happens that the things that managers insist on in the name of Scrum, and that programmers usually complain about, are actually not in the book.
For example, the book says:
> Scrum Teams are cross-functional, meaning the members have all the skills necessary to create value each Sprint. They are also self-managing, meaning they internally decide who does what, when, and how.
Surprisingly or not, Scrum neither mentions the need of "full stack developers", nor does it say that each ticket should be assigned to a random team member who happens to be free at the moment. Instead, it says the opposite, that ticket assignment should be decided internally by the developers team. So if the team agrees internally that one developer does the database and another developer does stylesheets, that is actually 100% Scrum compatible.
(Note: This is not an argument for Scrum being perfect, or even efficient. Just that it is meaningful to point at a bad thing and say "this is actually not Scrum", even if your manager insists that it is, even if most managers in most companies insist that it is.)
In the admittedly special case of IT consulting, the obvious advantage is making it easy for management to assign developers to projects of uneven length. If *every* dev can write the UI, server code, database, and network infrastructure, you just plug and play.
This isn't a new issue. Brooks wrote "The Mythical Man-Month"[1] in the seventies and it is still astoundingly relevant today. Most programming tasks are not fully atomizable, and the greatest expense in scaling up your work between multiple programmers is the inefficiency in communicating tasks and delineating task/component boundaries. The more you have people working across those task boundaries, the easier it is for each individual to contextualize where and how their component fits into what Brooks called the "programming system" (what we'd just call "the product" nowadays), and therefore the easier it is to do away with that inefficiency.
[1] https://en.wikipedia.org/wiki/The_Mythical_Man-Month
Yes, in medical data and apps, I've definitely observed this. I wouldn't say there's no specialties, we definitely have guys who are better and tend to work on different things but being able to do "everything" at least moderately well is important.
I'm not 100% sure why this is, I'm appropriately wary of upper management, but I know it's important, for obvious reasons, to have clinicians and lawyers and security at least in the loop on what we build. Which means putting together a project or tool involves 5 people, only one of whom is building the actual thing. If there's two technical people involved in building the thing, it's ok, but more than that is unworkable. Honestly, more than half of our staff feels like they're building internal tools for other staff members to actually deliver value. Weird but true.
Maybe the lack of specialization is why so much software seems to be utter garbage? Just a random example but our county recently spent ~$11 million to update a very old say 2005 era scheduling system that was pretty bad and functioned poorly for users.
Somehow after $11 million they somehow settled on a release that is worse. Like EVERYONE hates it (except the people with direct ownership of the update). Its main virtue seems to be that instead of looking like it is from 2005, it looks like it is from 2009. It also has a lot of cool functionality that might be helpful if the whole thing wasn't so badly designed. But also all the cool functionality is broken (and its been out over a year, and doesn't sound like it will be fixed), so who cares anyway.
This is a project I am positive I could have provided a better product of with IDK $50-200k. How did they spend $11 million and end up in a worse place. It is like the scheduling system was a D-, they had a million meetings, a fancy developer, and came out a year later with an F.
Other industries don't have this, no--you'd never see a job posting from Boeing for "full stack engineers" to work on their new fighter jet development.
This doesn't work as a management or development strategy, software uses it because it seems to like to aggressively insulate itself from the concerns of the outside world. It's why a lot of software is so bad (eg, no webapp can get scroll positions that don't jump around or behave randomly), why there are constant data breaches even at big companies, etc.
Its result is needing far more developers than are actually required. Instead of a small number of very-skilled specialists, you have a large number of relatively unskilled "generalists" who don't know how or why (or if) anything works, but do know a lot about random toolkits and APIs of the day.
This (at least partly) explains why you'll see small startups of just a few actual developers do something really impressive, then it gets bought by a large company who throws hundreds of devs at it, and it gets worse, or at least not proportionately better.
This mostly works out okay because most software development is actually really easy, most software is actually relatively unsophisticated, and most tech companies are relatively high margin and so can afford to do things other industries would find laughable.
I think that the computer programming profession is still discovering what makes for a sensible division of labour.
Established professions had to figure these things out long ago -- for instance "carpenter" and "bricklayer" are professions but "wall builder" is not; trades professionals specialise in a particular material rather than a particular type of structure, so if you want a wooden wall you call a carpenter and if you want a brick wall you call a bricklayer.
Computer programming (must we call it "software engineering"?) is still at the stage where it's figuring out what skillsets make sense to combine into one person and what skillsets are better spread across multiple people.
There's definitely something to this. Like there's the paradox where you kind of want security to be a specialty because it requires a very particular mindset but you can't do that because that mindset needs to be applied to every line of code, and every single little thing you do as a company to be effective.
I think that businesses have convinced workers in this case to become better (or wider) trained for less money. It’s a clever ploy.
Principles-wise, division of labor is bad.
From the wise Heinlein (Time Enough For Love) :
>A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects.
I detest that he cited butchering a peaceful animal as a useful skill, but this is my own belief as well. It's a tragic fact that knowledge generated by our brains so swiftly and grotesquely exceed and overwhelm them, but it's a fact that should and must be resisted, and can be locally thwarted.
It doesn't violate any separation of concerns principle in software to do so, the GUI can still be seperate from the database without necessarily requiring that a seperate brain design each of them. Good interfaces force good seperation, bad interfaces can't save you even if you had a million strong army of developers each micro-cultivating their own sub-micro-nano-pico-femto service.
My problem with companies and corporations cynically adopting the generalist ethos is that it's nakedly self-serving and not even truly generalist in practice.
The nakedly self-serving part comes from thinking a generalist developer is a general purpose 1-man team with the magical ability to do a team's job with a man's pay.
The disappointing failure to adhere to generalism comes from their expectation that your abilities are hyper-specialized to their own needs, plug-and-play, and can be turned on and off with point-and-click ease. They don't really want a db administrator, they want a guy who knows all the ins and outs of MongoDB version m.n.k that they are currently using. They don't want any old devops guy, they want a guy who knows every last flag and switch in AWS's interface. They don't want a Knuth-style algorithms-and-datastructures guy, they just want someone who knows every exact irrelevant word question asked in the leetcode interview by heart. They don't want a Heinleinen human (who necessarily will always be T-shaped to tradeoff depth for breadth), they want a million hyper-specialized insect all controlled and dispatched from the same human-shaped interface.
> Principles-wise, division of labor is bad. [...] but it's a fact that should and must be resisted, and can be locally thwarted
Why? I don't see a single argument why specialization is bad, except for a quote from Heinlein, who isn't even an authority on this matter (so it's not an appeal to authority – maybe "appeal to edgy sounding author"?).
I didn't want to dive into my motivation and reasoning for hating specialization because that wasn't the main point, my point is that what companies are doing are not really "moving away from specialization", but worse specialization.
Specialization is bad because :
(1) it's unsustainable, how many times can you divide a stick in half before you no longer can hold it in your hand ? how many times can you divide a field of knowledge into tinier and tinier pieces before it's no longer a meaningful and coherent body of knowledge but jumbled mess of footnotes-to-footnotes-to-footnotes-to-footnotes-to.... some other body ?
Henri Poincare :
>Science is built of facts the way a house is built of bricks: but an accumulation of facts is no more science than a pile of bricks is a house
This also implies that a *division* of a science (or more generally a field of knowledge) does not necessarily produce a science. (in the same way cutting a house in half does not produce 2 houses, but 0 houses) i.e. When something has a delicate structure, you can't naively cut it in half. Specialization rarely if ever follows the (or a) "natural" structure of the fields it cuts through, it's nearly always administrative bullshit or historical accidents.
(2) it imposes an artificial structure not present in the knowledge it divides. Specialization is a classification system at its core, and all classification systems are made up. Are Databases systems programming ? Network programming ? AI ? Compilers ? All of them. Specialization in the weak sense is already what generalists naturally do, which is T-shaping : taking a broad survey then diving in. But specialization is almost always taken to far more unjustified and ridiculous reaches
(3) Because, simply put, variety is good. I don't know how else to phrase this, I think that nearly all people would prefer more variety of food, more variety of people, more variety of places, more variety of languages, more variety of anything IFF all else is kept constant. Specialization is bad because it chokes the diversity and variety of knowledge available to you (again, weak kinds of specialization are the motte, and are already practiced by nearly everybody who knows being human entails cognitive limits, that's not what I'm talking about)
(4) In the context of governance and civilization-building, specialization enables the "Seeing Like A State" phenomena and a myriad other forms of oppressions, because it breeds an entire class of people so far removed from actual value-creation aka getting shit done, a class whose entire reason for existence is finding reasons for their existence. This is what Heinlein is referring to with "Specialization is for Insects" closing remark, and is supported in practice as the early rise of city state civilizations enabled unprecedented authoritarianism and alienation.
To put this in the most blunt form : You are a monkey, and monkies don't specialize. If you were an Ant or Bee, you would specialize in fighting or breeding or governing and you would be happy and content to serve/be the queen, but you're not, you're a monkey, and monkies don't specialize, not permanently, not to the grotesque degree modern hairless monkies invented. Just because your ancestors invented computers and the nuclear bomb doesn't make you not a monkey, you're still a monkey, forever bounded by your nature to love certain things and hate certain things.
The "T-shaped generalist" is a motte-and-bailey idea. Most people (including me) would approve of the idea of knowing a little bit of everything, and a lot of something.
But how much precisely is the "little bit" that you are supposed to know of everything? Enough to solve each randomly selected Jira ticket?
If you say "no", then you are *not* a literal full-stack developer. There are parts of the stack where you cannot deliver solid work without help.
If you say "yes", then exactly what is the point of knowing the "lot of something", given that you can already solve all Jira tickets *without* it?
Its not win-win. Multi skilled techies are going to.be rarer and more.expensive than monk skilled ones. The problems of getting monk skilled teams to coordinate aren't unprecedented or.insoluble, they are just routine management issues. A lot.of.things in software development are.positird.in management just not being very good.
It is incredibly ironic to watch many of the pro-Chesterton's Fence people, including Scott, defend cryptocurrency. Almost unimaginably so.
H̶o̶w̶ ̶s̶o̶?̶
Nevermind, I remembered what Chesterton's Fence is: "change should not be made until the reasoning behind the current state of affairs is understood"*. Then again, this principle could serve as a Fully General Counterargument against any kind of technological change, even though technological change as a whole is beneficial for society** ***.
*Plagiarism warning: taken verbatim from https://thoughtbot.com/blog/chestertons-fence
**Although Metaculus user soritical claims that "Most major transitions in human history have been morally ambiguous"; see https://www.metaculus.com/questions/4118/will-there-be-a-positive-transition-to-a-world-with-radically-smarter-than-human-artificial-intelligence/#comment-29787
***Citing your sources is a hell of a drug. So is adding footnotes. Please do not get addicted to this.
Chesterton's fence is not a fully general counterargument against technological change as we understand well the reasoning behind the current state of affairs in many case, especially in purely technical questions.
It seems like you're not really leaving a lot of room for nuance in this situation. We can be wary of abandoning state-issued and state-controlled currencies and still think there can be some merit to experimenting with cryptocurrency. Chesterton's Fence is a call for humility, not stasis.
Yes and it is invoked constantly by a certain democraphic for stuff that actually isn't a big deal or has even been tested in many other markets/countries but nowhere to be found on crypto, which has many very obvious problematic elements. Chesterton's Fence is just my ideas are good and your ideas are bad sophistry.
In the case of crypto there basically is no nuance. It is all bad. Crypto is a tool for deregulation and for obvious reasons it empowers immoral wealthy people and harms normal people. Tax dodging, black market stuff, hiding money in a divorce, pyramid schemes and ponzi schemes.
Chesterton's Fence says you need to understand *why* something exists before you do away with it. And in the case of regulation tech bros clearly don't understand why it exists which is why they are so hyped for crypto. Typical "get the government out of the way stuff".
The same reason they don't like unions and hold other right wing economic views.
Furthermore the vast majority of people don't actually use decentralized crypto, they use stuff like FTX, because currency decentralization is too hard to manage for 99.9% of the people.
This isn't a very charitable (or accurate) reading of Scott's post. He gave specific examples of how crypto can empower regular people, and I think we'd both agree that in certain countries helping people break certain laws/make black market transactions is a pro not a con. I don't know the stats on how much of crypto usage these cases cover (my prior is it's a small percentage) but the title of Scott's post is 'Why I'm Less Than Infinitely Hostile To Cryptocurrency', not exactly the strongest endorsement of cryptocurrency.
The title of course seems balanced but Scott's behavior and other actions outside the blog post don't really support that title.
> Tax dodging, black market stuff, hiding money in a divorce
One could also imagine using cash or gold for a similar purpose.
Yes but both of those are significantly harder to move. Especially across "borders".
"In the case of crypto there basically is no nuance. It is all bad."
Tell that to people living under authoritarianism or high inflation that want to protect their wealth where it can't be debased by the government or confiscated by banks.
https://www.cnbc.com/2022/11/05/-in-bankrupt-lebanon-locals-mine-bitcoin-and-buy-groceries-with-tether.html
It is incredibly ironic to watch many of the pro-Chesterton's Fence people, including Scott with a time machine, defend cryptocurrency, fiat money, the south sea company, free silver, bretton-woods, the solidus...
Kind of....
Like, on the one hand, yes, I think there's validity to this. If a significant chunk of the world's currency transactions shift to crypto, there's going to be a ton of unexpected side effects and people are underestimating that.
On the other hand...like, have you talked to a crypto bro? Because they are 100% confident they know exactly why that fence is there, quoting a scholarship going back 100 years to a bunch of dead Austrians. I would accuse them of many things but never insufficient research into money and monetary policy, at least not the old/hardcore Bitcoiners.
I think humility is called for but I can't accuse the Bitcoiners of not having done their homework.
I've encountered a wide range of nuance and depth from enthusiastic pro-crypto arguments. There's a decent amount of thoughtful (albeit usually deeply unorthodox) economic and political philosophy arguments of favor of crypto, but there's also a strong strain of what seems to be crypto maximalism as a tribal affinity marker, as well as unnuanced "BTC TO THE MOON!!!" cheerleading based on little besides linear extrapolation of past trends.
The last of these camps has been pretty quiet for the past several months since major cryptocurrencies have slid substantially from their late 2021 peak, leaving the former two camps much more visible by comparison. But they were very, very loud throughout 2020 and much of 2021.
If you compare the arguments of even the top 20% least bad crypto bros to the kinds of arguments Chesterton was opposing when he talked about the fence, they are substantially similar.
I think a good analogy now from student politics types, but also credentialed professors in relevant fields, is pro-puberty blocker trans activists.
Oh yeah, there's a lot of crap Bitcoin arguments, but I don't blame movements for their fanatics. I always think back to that old Steve Yegge article on Ruby vs Python: https://sites.google.com/site/steveyegge2/anti-anti-hype
"The thrust of Eckel's article appears to be that hyper-enthusiasm is diminishing the Ruby camp's message, and it's spoiling a good gentleman's argument. Those darn hyper-enthusiasts are focusing relentlessly on how cool Ruby is and how much they like it, when what's really needed here is a balanced, objective, neutral, moderated, standards-based, point-by-point, academic discussion of Python vs. Ruby, in which we can all make well-informed decisions, and may the best language win, as long as it's Python.
Python folks never really did understand marketing."
Like, I wish everyone liked rational argumentation, but they don't, and dumb cool hype with giant boobs is a necessary step in the widespread adoption of new technology.
The key is they absolutely don't understand why the various fences crypto exists to tear down were built. It is clear from their arguments. They are exactly the kind of reformers Chesterton was talking about, even if the topic differs.
I think "understand" is doing a lot of work in this critique. After a few hour long discussions over Bretton Woods, I think they know what the arguments for fiat currency are, I think they could pass a political Turing test, I just think they don't disagree with them for a variety of...debatable reasons.
I guess my focus just isn't on Bretton Woods specifically as a major issue.
I'm more thinking about domestic intra country regulation.
Dude, I'm not the right guy to defend Bitcoin here, but you can't claim that the Bitcoiners don't understand Chesterton's Fence if you're avoiding their big theoretical argument for Bitcoin.
I'm not sure what "domestic intra country" means but it sounds like you have a highly specific critique of Bitcoin that, to be fair, they might totally be ignoring, but that doesn't mean you get to ignore their core arguments, otherwise you're just kind of arguing past each other.
What is the existing fence? Cryptocurrency is a new idea, non-governmental currency is not. Privately issued currency was common in the U.S. in the 19th century, in Scotland in the 18th.
I can't tell if this makes it better or worse.
If cryptocurrency were truly new, Chesterton's Fence would be irrelevant - the reason it was never tried before is because it was impossible to try it, not because there was some logic to the previous state of affairs.
If it's just a higher-tech version of privately-issued currency, then the Chesterton argument makes more sense, in that presumably there was time to test the privately issued currency and decide they didn't like it and switch to the current system.
"then the Chesterton argument makes more sense, in that presumably there was time to test the privately issued currency and decide they didn't like it and switch to the current system."
It's not a surprise that "they", government, didn't like competing currencies. That's not an argument why the general populace should oppose private money.
Fiat currencies in the modern sense are also pretty new, and were invented in a pretty well-documented age.
There are now widely available apps that can generate novel images of an individual using a small 'prompt' of images provided by a user. An example of this is the Lensa app. This was immediately followed by abuse, e.g. generating celebrity nude images using a prompt that combines celebrity face shots with nude shots of others. It's not hard to anticipate an ever expanding category of embarrassing deep fakes that could be generated using just publicly available face shots.
Is anyone aware of strategies to counter this technology, when your career requires making a small number of profile pictures available on the internet? Adversarial perturbations seem promising (like the barely visible noise that makes image classifiers mistake pandas for worms or whatever) but I haven't found any hits on Google using just the obvious keywords for this kind of technique.
"People can already craft fake nudes ... So I don't see any real impact"
Presumably a lot of the distinction will be when you can use AI to deepfake a celebrity's appearance and voice into a pornographic *film* and not just a still image. Then once you have a 4K clip of that, you make a "fan cut" of your favorite Marvel film crammed full of obscene, degrading sex scenes and post it on torrent sites. There's undoubtedly a market for that kind of thing, and society such as it is now is totally unprepared for something of that sort to proliferate. Think of how feminists talk about what young famous actresses go through *now*: how are they going to sound when everyone can have non-stop access to hyper-degenerate, realistic pornography featuring every famous woman, on command, personalized by AI, 24/7, and you can't possibly stop them without shutting down effectively the entire internet?
I freely confess to not being an expert, but my impression was that the quality of these fakes has hitherto been poor, and they're also restricted to still images.
Also, I wasn't suggesting gimping AI; I was suggesting that society will have to adapt to this new reality but is wholly unprepared to do so. The feminist discourse around a woman's dignity, revenge porn, Weinstein, and so on is entirely ascendant and also entirely incompatible with almost every conceivable future. My claim was not "here's what we should do", my claim was "I predict a conflagration".
The long term solution is to change cultural norms to normalize sex work and make it so that having identifiable nudes on the internet isn't career-ending, because there's no reason it should be; despite OpenAI's attempts at neutering models, this genie isn't going back in the bottle.
In the short term, it seems the best bet is to use simply reuse the same headshot or at least ones from the same angle. Most of the apps I've seen recommend getting faces from differing angles, and different photos. Limiting the number of choices seems like the best bet there.
But really, I feel like this isn't something worth wasting brain cycles worrying about. Unless you have some sort of very specific job, it's unlikely that someone is going to be particularly motivated to single you out for fakes; and if they really are motivated, there isn't much you can do.
If convincing simulated nude photos can be made of anyone you have a few clothed photos of, then the existence of a portfolio of nude photos of you will not be evidence of anything. But there's a transition time where someone generates a simulated set of incriminating photos of an enemy and uses it to convince an unsophisticated audience (or one whose motivated reasoning in the case turns off their skepticism) to fire or shun or refuse to vote for the target.
I struggle to understand evidence of what should a naked picture of anyone be.
For example, a picture of a politician and a woman who isn't his wife naked together.
I think the counter is slowly society will go back to older methods. There was a period of time when photographs were taken very seriously, because they were very hard to doctor. Before that you had to rely on eye witness testimony and cross checking with multiple accounts and pieces of evidence.
As photoshop skill became more common people became less trusting of photos, and the same will happen to video eventually. So rather than needing to "hunt down" all the offending fake videos because they will offend people. People will (hopefully) just update their world model with the information that videos they see are very possibly doctored and not very good evidence of anything absent other pieces of evidence.
Photos have been fairly easy to doctor for most of their existence, compare things like the fairy hoaxers (and similar) or the Stalinist practice of removing people from pictures. I think it was at most a few decades between the development of "casual" highly detailed colour photography and the time photo editing caught up with it.
So I think photos were never really considered good evidence, unless there was some circumstantial evidence that would make them seem tamper proof even today (e.g. they were taken on Polaroid by a person known to have no image manipulation skills).
The ability for anyone seamlessly doctor videos (without heroic effort) is somewhat newer but has also been around for a few years.
I think deepfakes in the sexualized sense are great, the more convincing the better, because they offer plausible deniability for people whose actual nudes come out, and make it easier to turn the conversation around to be about the publisher and not the subject of the photos or "photos". I don't feel as sanguine about the social possibilities of deepfakes in the non-sexual sense because they offer the same plausible deniability to people doing things that are actually bad.
There's something a bit chilling about this idea, that the impact of a controversial event can/should be mitigated by a background level of distrust. I can see how in this case its beneficial for a victim, but this mechanism is also the basis for e.g. state-sponsored misinformation campaigns. In Russia, instead of denying a government misdeed it is more effective to circulate nine more very different accounts of the event until the first sounds just as implausible.
There may well not be a fully general one - the genie is out of the box.
I believe it's quite easy to tweak this sort of AI to add imperceptible watermarks, which is something responsible designers may do, but that just means that that section of the market will go to their irresponsible competitors.
And there will probably be an arms race between AIs that generate deepfakes and AIs that identify them; I have no idea how that will go.
But even then, identifiable deepfakes are still probably usable for a variety of nasty purposes, and they're only going to get better, easier to generate and commoner.
After toying around with ChatGPT and reading various twitter threads/articles about it I've come to realize that I am far easier to fool than I would've guessed. The replies ChatGPT gives me that I am confident are incorrect sound remarkably persuasive on first glance. Combine this with the fact that my knowledge of facts is limited (though above the Clearer-Thinking-test-taking average apparently) and I'm now second guessing many of my favourite substack/blog posts. I know Knoll's Law basically covers this, but I think this deserves more attention. So much of the focus has been on the ability of ChatGPT to be persuasive, rather than on how easily we're fooled.
Oh yeah a well crafted piece of bullshit that has a few of the right "protein receptors" to lock into our existing knowledge in a smooth way, and perhaps an enzyme or two to flatter our preconceptions generally undergoes very little scrutiny. There are a million examples.
> that has a few of the right "protein receptors" to lock into our existing knowledge in a smooth way, and perhaps an enzyme or two to flatter our preconceptions
This comment is an example of its own main point.
>So much of the focus has been on the ability of ChatGPT to be persuasive, rather than on how easily we're fooled.
Great way to phrase it. I think overestimating our own intelligence leads us to underestimate a lot of current AI capabilities.
>much of the focus has been on the ability of ChatGPT to be persuasive, rather than on how easily we're fooled.
Excellent point.
I see. I understand the valid points. I am not going into debating the validity of the formula. ( that I even misspelled in my initial post, as I was legitimately promptly corrected.
My initial intention ( probably not clearly expressed), was to ponder on the universality attached to it by the following message of the substack motto .
" All the rest is commentary ".
Which is exactly what I am doing.
Is it axiomatic or relative (and I am not asking philosophically ( death is absolute or relative because atoms get recycled).
Just from a pure generalized ( or not ?) applicability to that.
I am not debating, nor have the intention. I am genuinely asking and gave a ( flawed ?) example.
Looks like you didn't reply to a comment properly. If you do it by e-mail, it will go into the main thread, not into a reply.
I've always thought of Bayes Theorem as analogous to Carnot Efficiency. It's supposed to give you an idea of what's conceptually being worked towards. You probably can't draw a perfect circle, but understanding the definition of the platonic ideal of a circle (viz. the set of equidistant points around a locus) will allow you to draw an approximate circle with reasonable fidelity. artistic heuristics about drawing a circle are "optional commentary".
Also, Bayes Theorem is an identity. It's is axiomatic in the same sense that trig identities are axiomatic. not sure what you mean by the "death is absolute or relative" thing though.
I'm conducting a 2-minute survey to better understand how people with niche or intellectual interests find friends. Any help is greatly appreciated!
https://docs.google.com/forms/d/e/1FAIpQLSeCh2YASTcam5saz3aUhLF6hMJ10ubbHcZhrAI7nhaj1DWovg/viewform?usp=sf_link
Just filled out the survey!
(For those who care: my defining niche interest is high-level mathematics - math *qua* math).
They kidnap them, of course!
The collective noun for a group of wombats is a wisdom of wombats. This is one of those supposed facts which appears nowhere other than on lists of collective nouns.
My question is: who the heck comes up with these collective nouns, and what the heck is the process behind it? I can see them arising naturally in animals which naturally hang around in groups, but these animals are the ones with the dullest collective nouns like "flock" and "herd". The animals with really exotic collective nouns like "wisdom" or "bloat" tend to be solitary or small-group animals anyway.
Okay, you led me down the rabbit hole. I had a vague notion that collective nouns like that probably came from mediaeval bestiaries, and while the period is about correct, it seems (in English at least) that it's "terms of venery" - that is, specialised terminology that the nobility and well-educated would use, to mark themselves out from the common sort.
Also following fashion. Also probably scholars inventing and distributing these kinds of lists as a word-game. And in modern times, when new animals were discovered, people making up collective nouns to fit in with the traditional ones.
https://en.wikipedia.org/wiki/Collective_noun#Terms_of_venery
"The tradition of using "terms of venery" or "nouns of assembly," collective nouns that are specific to certain kinds of animals, stems from an English hunting tradition of the Late Middle Ages. The fashion of a consciously developed hunting language came to England from France. It was marked by an extensive proliferation of specialist vocabulary, applying different names to the same feature in different animals. The elements can be shown to have already been part of French and English hunting terminology by the beginning of the 14th century. In the course of the 14th century, it became a courtly fashion to extend the vocabulary, and by the 15th century, the tendency had reached exaggerated and even satirical proportions.
The Treatise, written by Walter of Bibbesworth in the mid-1200s, is the earliest source for collective nouns of animals in any European vernacular (and also the earliest source for animal noises). The Venerie of Twiti (early 14th century) distinguished three types of droppings of animals, and three different terms for herds of animals. Gaston Phoebus (14th century) had five terms for droppings of animals, which were extended to seven in the Master of the Game (early 15th century). The focus on collective terms for groups of animals emerged in the later 15th century. Thus, a list of collective nouns in Egerton MS 1995, dated to c. 1452 under the heading of "termis of venery &c.", extends to 70 items, and the list in the Book of Saint Albans (1486) runs to 164 items, many of which, even though introduced by "the compaynys of beestys and fowlys", relate not to venery but to human groups and professions and are clearly humorous, such as "a Doctryne of doctoris", "a Sentence of Juges", "a Fightyng of beggers", "an uncredibilite of Cocoldis", "a Melody of harpers", "a Gagle of women", "a Disworship of Scottis", etc.
The Book of Saint Albans became very popular during the 16th century and was reprinted frequently. Gervase Markham edited and commented on the list in his The Gentleman's Academic, in 1595. The book's popularity had the effect of perpetuating many of these terms as part of the Standard English lexicon even if they were originally meant to be humorous and have long ceased to have any practical application.
Even in their original context of medieval venery, the terms were of the nature of kennings, intended as a mark of erudition of the gentlemen able to use them correctly rather than for practical communication. The popularity of the terms in the modern period has resulted in the addition of numerous lighthearted, humorous or facetious collective nouns."
They've also been very popular with crime novelists, how many variations on "A Murder of Crows" or "An Unkindness of Ravens" have you seen used as a murder mystery title?
I’ve read that the term “pack of wolves” originates from this game, which if true is surely the most essential term produced by it, the sociality of wolves being their defining trait. A word like “herd” just doesn’t seem appropriate for a predator. And there’s nothing fanciful or silly-sounding about “pack”.
A murder of crows
A murmuration of starlings
A gaggle of geese
A den of foxes
A pandemonium of parrots
We have done better than flock and herd.
> A den of foxes
That's invalid; if you say "a den of foxes", you will be referring to the set of foxes that live in a particular den, a family, not a collection of foxes that happen to be in physical proximity to each other.
Quite right.
The correct term is a skulk of foxes.
Den is reserved for thieves.
> The animals with really exotic collective nouns like "wisdom" or "bloat" tend to be solitary or small-group animals anyway.
The exception would be "pride of lions", which has a lot of popular currency and sees a lot of legitimate use no doubt _because_ lions are social and therefore referring to a group of them makes sense to do.
There's a murder of crows-- common bird, fancy collective noun. I have no idea whether the term was ever in common use, but it's handy for mystery writers.
Hey Scott! I was playing around with DALL-e and got it to make a pretty good stained glass image of Darwin and the finch. While I somehow think you don't really want your house to be dominated by rationalist stained glass windows, you might want to have a look. It's here: https://photos.app.goo.gl/2XNbUJxh84hbFgYz5
Wow that looks great.
You should know your name appears in the photos link. If you're uncomfortable with that you should consider posting it to imgur or another free image hosting site.
Thanks. It’s not my real name anyhoo
Took the clearer thinking test. I scored below average across the board. Is there any hope for me left in this brazen world ?
Keep in mind that the sample is other ACX readers.
Yeah, but even average me came out kinda top - Maybe I was early? -If dices are not your thing, it can be frustrating. - I found much more irritating that they abuse that "test" to advertise EA and push their book. Felt like an encounter with Jehovah's witnesses.
> I found much more irritating that they abuse that "test" to advertise EA and push their book. Felt like an encounter with Jehovah's witnesses.
Yes, I complained about that in my disappeared comment. "Would you be interested in our newsletter? How about reading our book?" is not a personality inventory. It's advertising.
> I found much more irritating that they abuse that "test" to advertise EA and push their book. Felt like an encounter with Jehovah's witnesses.
I did not see anything like this either. It's interesting how different everyone's experiences were!
Did you put that you identify as rationalist? I didn't, so maybe they only advertise if you do.
I did, and I didn't notice any 'Read our book!' advertisement.
I didn't notice any ad either. I do have comprehensive ad blocking, and I also chose "reject all labels" or however that was worded.
Well, I marked "libertarian" - as among that looooooong list there was no option "irrationalist" (Tyler Cowen). - They asked a battery of questions that were - obvious to AXC/SSC readers -arguments to join EA. Then they asked: Do you know a philosophy/movement with such ideas as suggested? What is the name? I typed: "EA (as utilitarianism is no movement)". Then there was a bit more about EA, this time explicitly, and then a large book cover popped up, no link to to buy, though; I remember it as "Doing good better" by William MacAskill - it was blue, I recall. :) So, yeah, "EA going YW" was my thought. - Btw: I only had two kind of"IQ-tasks" the spatial ones with folding dice and those verbal: 4 words, find the odd one. (I am ol'n'confused, much dumber than the results suggested; I might mix up some details of this EA-PR, but that was how the experience went for me.)
Oh man, given the current Sam Bankman-Fried imbroglio, maybe cooling it on the William MacAskill Will Change Your Life bit might be a good idea 😂
"It was his fellow Thetans who introduced SBF to EA and then to MacAskill, who was, at that point, still virtually unknown. MacAskill was visiting MIT in search of volunteers willing to sign on to his earn-to-give program. At a café table in Cambridge, Massachusetts, MacAskill laid out his idea as if it were a business plan: a strategic investment with a return measured in human lives. The opportunity was big, MacAskill argued, because, in the developing world, life was still unconscionably cheap. Just do the math: At $2,000 per life, a million dollars could save 500 people, a billion could save half a million, and, by extension, a trillion could theoretically save half a billion humans from a miserable death.
MacAskill couldn’t have hoped for a better recruit. Not only was SBF raised in the Bay Area as a utilitarian, but he’d already been inspired by Peter Singer to take moral action. During his freshman year, SBF went vegan and organized a campaign against factory farming. As a junior, he was wondering what to do with his life. And MacAskill—Singer’s philosophical heir—had the answer: The best way for him to maximize good in the world would be to maximize his wealth.
SBF listened, nodding, as MacAskill made his pitch. The earn-to-give logic was airtight. It was, SBF realized, applied utilitarianism. Knowing what he had to do, SBF simply said, “Yep. That makes sense.” But, right there, between a bright yellow sunshade and the crumb-strewn red-brick floor, SBF’s purpose in life was set: He was going to get filthy rich, for charity’s sake. All the rest was merely execution risk.
His course established, MacAskill gave SBF one last navigational nudge to set him on his way, suggesting that SBF get an internship at Jane Street that summer."
And as we know, the rest is history!
18 hours after you commented (which is when I am commenting), the average cube rotation was around 0.2, the average hidden pattern was around 0.4, synonyms around 0.42, treasure 0.42, and made-up words was around 0.7.
I'm pretty shocked at the low average for cube rotation, and the high average for made-up words. I would have guessed the average performance to be 50% across the board.
Cube rotation was 0.24 - words:odd-one-out was 0.62 (even there I "scored higher than 79.07% of the people who took this test." Which gives me a warm glowin' feelin'. Kurt Vonnegut described once how aliens take over earth by demoralizing humans with "fake facts of high averages" that make most feel like miserable failures. I suspect a sales-trick when I get yummy scores.)
Don't worry, as long as I'm taking any kind of intelligence-measuring test anywhere, you will always score higher 😉
The way things have being going in the Reality Department, I expect Musk to lose the poll, but then declare the entire election process corrupt, and override it.
He'll find people stuffing ballots into mailboxes in Arizona, and plead with the technicians that all he needs is 11, 000 votes. Mobs will rush the Twitter Tower, and Musk will declare himself Emperor of Twitter, Thank You Very Much. That old CEO business was highly overrated, anyway, he'll say.
Musk already lost the poll (it about half a day ago as of the time of writing). It remains to see whether he will abide by it.
https://twitter.com/elonmusk/status/1604985324505030658
Too late.
https://babylonbee.com/news/elon-to-stay-as-twitter-ceo-after-counting-mail-in-votes
You're so out of touch with what you're talking about that you missed the part where the poll had already been long closed by the time you posted this nfoolishness.
Ahem. "Twitter is a private company, they can do what they like, if you don't like it then set up your own version!"
xkcd: https://xkcd.com/1357/
"Free speech just means the government can't arrest you; if you're...banned from an internet community, your free speech rights aren't being violated. It's just that the people listening think you're an asshole, and they're showing you the door".
Take note, Thursday Night Massacred!
So Elon can do what he likes with his own private company, including running polls to troll the users and then declaring himself Emperor. Or did you mean it only applied when done to your *enemies*, it wasn't supposed to happen to *you*?
Nearly everyone that I follow both acknowledges that Elon is 100% allowed to do all of the random bullshit he is doing. What he's being called out on is in his repeated claims that he's actually not doing the same thing he's accused everyone else of doing and saying it's the worst thing in the world for (ie content moderation), except in a less consistent/predictable manner.
I don't think, for example, that you could find any significant number of people seriously claiming that, say, Parler is "violating free speech" if they banned whatever random leftists showed up to pick fights with their users and call everyone a racist sexist douchebag. They might point out that doing so is fundamentally the same action as banning someone for saying something racist or sexist on Twitter, so they're clearly no *better* at what their marketing claims is "free speech". But, you're not going to find many (if any) people suing Parler for getting banned and claiming their 1A rights are being violated, any more than you would for getting banned from editing Wikipedia.
People are definitely annoyed that Elon both has enough money to simply "buy a platform" in the first place, as well as annoyed that he, y'know, actually did so and then did his best to recreate the entire culture of the platform to better suit his ego. Broadly speaking, most people don't think what he did was *illegal*, just, y'know, a bad thing to do. It's like the old school forum days - someone new becomes a mod, starts changing everything that's already been in place, and you're likely going to find the old regulars aren't happy. He's a bad dude, not a criminal. (Probably. He might also be a criminal for some of the workplace changes that he implemented, but almost certainly not for his policy changes, at least not within the American legal system. EU may likely be different, I know very little about their laws.)
Do you think that people suspended or banned under the previous Twitter leadership were "assholes" who were "being shown the door" as per the linked comic? Does that equally apply or not apply to people being banned or suspended now? Why or why not?
I think Musk is beclowning himself on a daily basis, and it's not good for him to do so. I also think he has a lot of support and his base reasoning for buying Twitter was the arrogant "when you are banned from Twitter it's because you're an asshole, learn to read the room" response that was clearly not an unbiased and accurate retelling of the situation. Is Musk doing something uniquely wrong by reversing course and instituting bans on his ideological opponents, or is he going tit-for-tat on previous defectors? Is it worth admitting that the previous rules were in fact a defection on norms (and often directly against the written terms of service) that were inappropriate? Should a person have to spend $44 billion in order to make that point in a way that others take notice?
I'm not a fan of Musk and his antics, but I'm unable to find a good alternative to this situation - other than to shut down Twitter and all social media, which while I'm in favor of that, is both illegal to impose and very unlikely to happen.
I'm a (certain type of) data scientist, and a boss of mine asked me about how to think about ChatGPT. In order to organize my thoughts on it, I wrote a post about it. I'd be really interested to see what folks around here think of it: https://ipsherman.substack.com/p/an-opinion-about-ai-chatgpt-and-more . P.S. I don't spend a lot of time in these Open Threads, so I apologize if self-promotion here is gauche. Please let me know if it is, and if there's a better way to solicit feedback!
> I’ve been hostile to “AI” as a term for a long time. Probably more hostile than I should be, but hostile I have been. I've changed my mind somewhat in the past several months, for reasons we'll get to, but I'll try not to bury the lede. Here’s my current opinion on ChatGPT in particular, the AI iteration that everybody's talking about as I write this:
>> It’s the first “modern AI” that I have personally found useful, but even then, I only find it limitedly useful. I use it (when its servers aren’t overloaded) as a very specific type of search engine, for the types of questions I would otherwise Google, namely esoteric, fact-based questions like “What’s the best way to dedupe a list in Python?” or “What’s the origin of the phrase ‘For all intensive purposes’?”
I don't like when explanatory essays give a bunch of caveats to a topic that hasn't been revealed. because i have to hold all the caveats in my brains while i wait for the lede to drop. If I were you, I'd reorganize this as:
> I have mixed opinions on GPT. It’s the first “modern AI” that I have personally found useful, but even then, I only find it limitedly useful. I use it (when its servers aren’t overloaded) as a very specific type of search engine, for the types of questions I would otherwise Google, namely esoteric, fact-based questions like “What’s the best way to dedupe a list in Python?” or “What’s the origin of the phrase ‘For all intensive purposes’?" But to fully grasp the ramifications on the field of AI and society, there's a lot of context to explain.
> AI - WTF Is It?
> I’ve been hostile to “AI” as a term for a long time. Probably more hostile than I should be, but hostile I have been. I've changed my mind somewhat in the past several months. One reason is that it just doesn’t really have a definition. Or maybe its definition is so broad as to not mean anything.
----
> Sure, machine learning is called machine “learning,” but what it’s doing isn’t “learning” so much as “systematically updating parameters in increasingly complicated equations to minimize loss functions.”
This still looks kinda abstract. Borderline eulering. I know it's not your intent, but your boss might not know any better. I'd change it to
> ... isn't "learning" so much as just doing linear algebra, which is a mathematician's equivalent of using a spreadsheet.
and then maybe take another paragraph to explain that ML takes pixels, characters, etc and automatically loads them into something that resembles a spreadsheet. I think this more concretely relates ML to something your boss will understand, who will therefore conclude "if ML is just a fancy version of MS Excel, it's probably not conscious".
----
I like everything else about the essay.
Great feedback, I really appreciate it! The order in the beginning is a bit of a historical accident of where the idea for the post came from, and your reordering definitely does make it easier to read.
And the eulering point is a good one too. If I'm honest, I never felt that comfortable about how I phrased that part. I just couldn't think of a way to elegantly make the point that the word "learning" in "machine learning" is really different than what most people mean when they think of "learning," and that's what I ended up with. I like your phrasing (especially comparing to a spreadsheet) a lot better.
Glad you liked the essay. Thanks for reading and commenting!
don't worry, i immediately recognized what was happening in both scenarios. i've been there. I.e. at the beginning of your writing, you were trying to explain a fuzzy perspective, but you don't know where to start so you just ramble a bit until you hit your stride. And then you weren't sure how to best summarize what ML actually was doing, because you're so used to being deep in the weeds.
That's exactly what happened in both cases. Thanks again for the feedback. I really appreciate it.
Random thought: I wonder if you could use the same principle of noise cancellation to create active anti-sonar tech - basically, create a perfectly out-of-phase pulse of the same amplitude as the inbound ping, thus cancelling out your echo. This would obviously be very useful for military submarines / USVs, so I imagine either it's already being investigated or there's a technical reason it wouldn't work.
Not an expert, but I doubt it would be possible to perfectly cancel the noise across all frequencies and directions. Real life "noise cancelling" headphones certainly don't.
I know modern radar systems rapidly switch frequencies (even in a single pulse!) to bypass jamming and avoid detection, so I would assume sonars do the same.
> Conventional wisdom is that intelligence-related studies replicate better than other fields, and Clearer Thinking is testing that now by trying to replicate 40 intelligence-related claims. They’re looking for experimental subjects to take their online tests; click here if you want to help.
Heads up, but I think the second example question in the "minimum average distance to red dots" section is incorrect. The test claims that the correct answer is to choose the middle, which has a total distance of 3sqrt2. However, if you choose one of the sides instead, the total distance is only 2+sqrt5, which is slightly lower.
Hmm there must be a number of different tests. I got nothing with red dot distance, nor image folding. I did fairly well at remembering flags, till my brain got a little flag fogged in the end.
More feedback on the intelligence test:
* In addition to the incorrect example question for centroids, one of the paper folding questions is impossible as well (near the middle, I forget which one). There's no way to fold the first image into the second unless you flip it over first.
* Why do we need to click a link to explicitly pause the timer between sections? It should be paused automatically when you're between questions or sections!
* SAT question enforces max of 1600 rather than 2400, regardless of when you took the test. Also not clear how to answer if you took the SAT multiple times
* no option to list a double major in college
* might be nice to have examples of what is urban, suburban, highly urban, etc. for comparison
* No way to go back in the demographics section if you mess up. Only option is to "reset everything", which is rather scary if you just spent an hour doing the previous tests and don't want to lose progress
> SAT question enforces max of 1600 rather than 2400, regardless of when you took the test
I haven't taken the online assessment that Scott linked to, but here's a general trick for comparing 2400-SAT scores with 1600-SAT scores (without using percentiles). Take the scores from the two non-math sections of the 2400-SAT. Average them. Then, add that average to the math section score, for a total out of 1600.
If you want to get non-garbage results, you need to be explicit about that, or else everyone will just guess about what they're supposed to do (in my case, I just used the math and verbal section and ignored the writing section). Or better yet, fix the question so people don't have to follow confusing workarounds in the first place.
It feels to me like it's going to be much easier to write AIs that can give 80% correct solutions 80% of the time to "hard" problems than to write ones that can reliably and consistently solve "easy" problems (definitions of "hard" and "easy" intentionally not gone into).
So what are some applications for which a quite-likely-to-be-correct output that you can't rely on is valuable?
And what are good ways of using plausible-but-unreliable applications plus secondary testing to get reliable applications?
Video game testing/playing is one obvious application.
Maybe. Accurately simulating the way humans play may be harder than it sounds. Basic checking for bugs might be possible.
I await reviews saying that new games obviously weren't tested on humans.
Coming up with 0-day exploits, or scientific hypotheses, for the first question. Doing the actual science, for the second.
Already started with legal contract review , corporates are becoming happy to take a risk on a low value deal (under $1m, not strategically important etc) if red flags (high risk clauses) are not present when reviewed automatically.
Today high volume deals may not even be getting a real legal review , so 80% is better than 0%. Expect consumers to be even more keen to get a basic review due to the costs of law firms.
I have three more subscriptions to Razib Khan's Unsupervised Learning to give away. Reply with your email address, or email me at the address specified in my about page:
https://entitledtoanopinion.wordpress.com/about/
All 3 subscriptions are now sent out.
Re. this:
"As soon as new psychology and behavior papers come out in Nature and Science (the two most prestigious general science journals), ..."
I don't think scientists should treat Nature and Science as the two most-prestigious general science journals anymore. Both have written editorials and/or letters to the effect that science must be politicized, and articles should be judged on the desirability of their expected impact on society, in addition to their scientific merits.
On the one hand, it's true that they've succumbed completely to Lysenkoism; on the other hand, the massive political skew of academics means they retain their old prestige anyway. In fact they probably did it to retain their prestige. And doesn't that just make it that much more urgent to test their articles for replicability?
I think what you call Lysenkoism is almost entirely in the humanities. Once I went through all of the professors at a university who had signed a particular petition asking for the reinstatement of a professor who was fired for having students physically eject a journalist who'd come to report on their protest. (I don't understand the logic of staging a protest and then attacking reporters who try to cover it, but whatever.) 100% of the English department signed it, and a pretty high fraction of all humanities profs (but less than half); IIRC about 3% of profs in the hard sciences did.
Seems to me I saw some other survey with similar results: something like 40% woke in the humanities, and around 3% in the sciences. Maybe. My memory is terrible. Also, it depends on whether sociology and psychology are counted as sciences or humanities.
If it's only happening in the humanities, why are Nature and Science doing this ridiculous shit?
No, its happening in the sciences too, though it's not as far gone as in the humanities. Also 3%/40% is laughably low. 20%/80% might be credible. Source: my lived experience as an academic scientist.
They're still the most prestigious, even if they aren't the most reliable or rigorous. Even apart from politics, the desire for flashy surprising results means that their choose articles that are more likely to be wrong than usual.