> the only theory I have is that it's due to the lack of social cues
I’ve figured it’s some combination of depersonalization and that it’s hard to read text empathetically. Probably those two and what you mentioned are all strongly related.
I’d wager for most people, in person it’s easy to be amicable and listen to people’s words generously unless you’re motivated otherwise. Reading text does not do that same thing for me. I have to make a conscious effort to even remember that one block of text from commenter 1 and another by commenter 2 were even written by different people, much less remember to tie them back to specific identities with their own points of view and such.
If that’s at all common, then yeah, I figure something like what you suggest would help a lot. Even phone calls without video are much, much better at engaging the “yep, that’s another person” mode.
But now that I think about it more after writing that comment, maybe there’s something more subtle. Even on, say, a work Slack with people I’m very familiar with it’s too easy to accidentally be harsh or unfair in communication and I wonder if your social cues theory is onto something.
Depersonalization seems like a big part of it. Road rage is another example - people interacting with each other in a system where they see each other just as blank glassy windows can get *much* more heated.
I've seen studies finding that people in expensive cars are less polite to pedestrians than people in cheap cars, but I wouldn't be surprised if the effect is mostly due to whether your windows are open or closed, and tinted or clear.
I agree with Dalton, depersonalization is a big thing. I read texts from my wife in her voice and (possibly due to a long-ish history with her) I seldom get her tone wrong. I've yet to enter a heated debate with almost anyone except online commenters I don't know - rare exceptions happening in real life with close friends on religion or morals.
Perhaps this could already be lessened if there was a photo next to the name, maybe even more so if you add three values and three causes important to this person or something like that, providing context for this person's existence.
That is probably part of it, but even bigger part is that in the spoken-word world people usually don´t have conversations about hot button topics with complete strangers
I am not an American, so no. Probably every country has its out of bounds topics, but they are slightly different in each case. Social media dynamics are similar everywhere, though
Possibly the case that online discussion is less polite than in real life, but is that a desirable goal? My desire in participating in these discussions is to test my own thinking against others, and hear other people’s ideas. Ftf introduces a lot more social calculations which cloud that truth seeking. Also I find if I type out my arguments rather than express them verbally I am much more precise and less prone to glossing over issues or conflating things and I think discussions online makes spotting that easier in others. Of course it is unpleasant to be insulted by someone online, but if all somebody has to offer is insults it makes them easy to ignore.
Of course it is always better if people are polite, but I mean that being polite is a secondary not primary goal of online discourse. If being polite meant that discussions were meaningless because everyone was too scared of offending then it would not be a good in my view.
On whether online discussions are the cause of polarization, I think back to the pre internet days and my impression is that it was just as bad or even worse. Racism or sexism is called out very quickly in most online forums today, but it was certainly not the case in ftf discussions in say the 1980’s.
Filter bubbles could be a factor as well. I am sure plenty of people never actually talk to others who hold views that they find deeply objectionable in their normal lives but will encounter such people in digital spaces, this space is probably particularly bad for that.
You're rediscovering the reasions for the invention of the emoticon, and later emoji? We had this debate in the Usenet era. But I think the canonical hypothesis (lack of nonverbal cues) is mistaken. People can and do communicate by writing and other nonvisual means cordially, and have done so throughout history.
What I think makes the difference is the expectation of whether you will interact repeatedly with the same person, or whether future interactions are avoidable. If you are certain you will be interacting with the same person over and over again, then I think conversational styles -- face to face or not -- are always far more cautious. On the other hand, if you doubt you'll necessarily have any further interaction, they can easily become inflamed, whether they are in writing (the Internet) or even in person (cf. road rage).
So even Internet fora can be courteous places *if* almost all of the denizens interact over and over again, and there is a strong expectation from the beginning that they will, and that there is no easy way to avoid interaction with others in the same space.
Arguably, the major reason the Internet degrades conversation is that it too easily allows us to interact *just once* with others, and then never again.
Well, let me be careful here. I'm not referring to social reputation in the sense of how many "likes" you have -- *that* kind of thing, designed to solve the spam/flame problem has in many cases only made it worse, cf. the dumpster fire that digg.com became, the distortion of dating sites, et cetera. The only case that comes to mind where it kinda-sorta worked early on and never blew up was Slashdot, which used the concept of meta-moderating to keep the moderation on point.
What I mean is your *actual* long-term reputation in some other individual's head, and the chance -- nay, probability -- that you will need to deal with the consequences of that reputation in the future. It's surprising how moderating just the knowledge that you will be forced to look the other person in the eye and say hello some time in the future can be. People *really* don't want to look forward to a look of contempt.
That is, I'm saying predictably repeated long-term interaction is the missing key. In real life that's guaranteed by the historical difficulty of long-distance communication: typically *anyone* with whom you had a conversation of any length would be colocated for a while, for better or worse. You were at the same office, deployed at the same place, on the same train, living in the same neighborhood, et cetera. But now it's very different. You can meet 10,000 strangers a day, and never see or hear from them again. It's not natural, evolution hasn't provided us with good instincts for that.
You might find exploring Albert Mehrabian's findings (often summarized as the 7:38:55 rule) relevant. I can't add much more ... I'm familiar with them, but haven't gone into depth with them (the basic point seems SO obvious ... and I don't care too much about the exact numeric partitioning).
I, on the other hand, wish more people would talk in person like most of us talk on the internet in text. Be more direct. Don't pull punches. Say what you mean. Sound angry if you're angry.
Anyone got recommended resources for learning about fusion power research?
(An interesting caveat to some news stories: the [energy out / energy in] ratio is called Q. But there's Q_plasma, and Q_total; the former is what's often reported, but the latter is what really matters if you want to run a power plant. [E.g. the plasma outputs more energy than the laser inputs to it, cool, but the laser itself takes power to run.]
Companies talk about Q_plasma>1 as an important milestone, and it is, but often the implication is that that would be enough to actually get energy from it, and that's false; only when Q_total>1 can you actually run a power plant that gives you more energy than it uses. https://www.youtube.com/watch?v=LJ4W1g-6JiY)
I was curious about this same topic after seeing the same video, thanks for posting the question. Aside from the physicist in the video, I've found it hard to find non-hyped-up sources for fusion power. I'd like realistic estimates and timelines.
I've been thinking about how to do a series of blog posts about fusion. One would describe the timelines proposed by the major players and the extent to which I think they're plausible.
When I get this finished / up, I'll comment a link in an Open Thread.
Everything I've seen about the current state of the field (e.g. MAST Upgrade) has been from random people who work in fusion research. I seem to remember some of them being in the comments section here .
(What's the field like, people who work in fusion research? I was thinking about applying for CCFE's mechanical grad scheme. Do engineers get to do fluidy things or is that all crazy MHD that gets left to the physicists?)
The hottest area in fusion engineering right now is probably magnet design with high temperature superconductors. Most people doing fluidy things for fusion are physicists, but the boundary between fusion science and fusion engineering isn't well defined yet. You don't really need a physics background to do fusion science: you need to know E&M and fluids, but not quantum.
Culham is great. They have had (debatably) the best fusion experiment since the 90s. I only know 2 theorists who work there, so I can't give you personal advice on who to work with. But the group as a whole is good. The startup Tokamak Energy Ltd is nearby and might also want engineers.
MHD is weird in that people who don't know it think it's super accurate & complicated, but people who do know it think it's jank. MHD is fluidy: it's fluid equations with an extra force and a few constraints. People use it for preliminary tests: if MHD shows that your plasma is unstable, it definitely is. The crazy analysis mostly involves kinetics. You might have heard of the Boltzmann equation. The Vlasov equation and gyrokinetics are the best equations to use to analyze fusion plasmas.
The past year or two have been very exciting, probably the most out of the ~10 years since I first joined the fusion field. Now's a great time to join.
Yeah, engineers may not be doing plasma-MHD, but actually there is a great need to study how fluids like liquid metals flow in the strong magnetic field environment. This is critical for heat transfer and design of the components surrounding the plasma. It's also under the banner of MHD, just in a different regime.
Fusion has a lot of challenging applications for mechanical engineering in general. Here's a sampling:
1) With high-temperature superconductors, some magnet designs can be limited by the strength of materials. The magnet assemblies need to be incredibly strong yet not take up a large area - one engineer called them "[one of] the most structural members ever".
2) The 'breeding blankets' that surround tokamaks need to withstand extreme fluxes of heat at the surface and neutrons, and operate at high temperatures: they're kind of like a heat exchanger between the neutrons and your primary working fluid. They need to be strong in order to withstand transient magnet field change events which drive currents in the walls; these currents react against the magnetic field and cause large forces. One strategy is to make cuts so that the electric current paths are much longer, but this makes the structure more flimsy. In future designs the blankets may need to be lifted out of the machine vertically to be changed; a flimsy structure makes this more difficult.
3) The machines will need to be handled by specialized (radiation-resistant) remote-handling devices. See what RACE (Remote Applications in Challenging Environments) at Culham Science Centre is doing; they're doing really impressive work.
4) In general the fusion machine environment has high temperatures, heat fluxes, neutron fluxes, and can have large mechanical and electrical loads. There are only certain classes of materials that can be used due to the neutron environment. The whole structure can be compared to building a pressure vessel out of magnets. The tolerances are very tight, sometimes 1 millimeter for a structure 20m high.
Maybe a bit high level but I recently read Arthur Turrell’s “The Star Builders”. Although my reading is he is pretty skeptical of some of the private enterprise efforts in fusion, perhaps seeing them as scams. He seems to favor the big Government led efforts as most likely leading to success, on much longer time tables than perhaps people are hoping. Note he is a former physicist researching this area but now has a role in UK Central Statistics agency and also has a website that teaches coding, interesting guy.
Some of the private enterprise efforts are not that different from scams. Lockheed Martin is big and powerful, so I'll pick on them. The only data they've published is for an experiment 1,000 times too cold for fusion - it wasn't even fully a plasma. And they say their first reactor will be small enough to fit on a truck.
In terms of the triple product (density * temperature * confinement time), the big Government fusion efforts are 10,000 times better than they were in the 1960s. While I love completely new designs and new ideas, they have a long road to catch up.
It is worth distinguishing between private enterprises that build on everything we've learned & have a plausible path to fusion and private enterprises that are little better than scams. Plausible private enterprises include Commonwealth Fusion Systems (building SPARC), Renaissance Fusion, Type One Energy, and maybe Tokamak Energy Ltd. Helion also deserves mention: I think their strategy is unlikely to work (at least in the next few decades), but they are taking the challenges seriously. Plausible private fusion really got started in 2018 after some advances in high temperature superconductors from MIT.
Commonwealth Fusion is currently leading in the race to First Fusion. Their experiment SPARC is being built over the next 4 years and should have Q ~ 10. ITER (the Government effort) should be finished slightly sooner, in 2025, but isn't planning on using tritium until 2035.
That seems very aligned with Dr Turrell’s analysis. Would not have thought Lockheed was a company likely to engage in a clear fraud, so surprised at that one. From my reading my uneducated guess is that fusion is very likely (90%) to be cracked in next 20 years and we have working commercial fusion devices that compete with fossil fuel. That is some black swan ahead of us. Would be interested in your probabilities on this.
One possibility is that Lockheed is credible, but secretive, and do have a plausible path to fusion. Another is that they know that it is implausible, but figured that they'd expose themselves to a high-risk, high-reward chance. A third possibility is that they are too specialized / don't bother distinguishing between plasma physics and combustion physics.
My probabilities vary widely depending on what "cracking fusion" means. If it means Q > 5 in some experiment, I would put it at 95% (and some of the remaining 5% is X-risk). If it means anyone is selling electricity produced by fusion to the grid in 20 years, then 90% is a good estimate. If it means that new fusion capacity is cheaper than or similar to the cost of new fossil fuel capacity, then I'd lower it to maybe 75%. There's just too much we don't know about how the economics will work, or how to scale up reactor manufacturing, or the regulatory environment for me to have much confidence.
In Taleb's zoo, this is a gray rhino rather than a black swan. Something we could see coming, but didn't bother to look at it.
I assume Lockheed is participating because they can possibly get government money flowing to keep it going, and it's not really meant to be much more than a jobs program.
For magnetically confined fusion, an optimized power plant should be able to get Q_engineering > 1 as long as Q_plasma > 5. The leading contenders for First Fusion, SPARC and ITER, both are aiming for Q_plasma = 10. They are not aiming for Q_engineering > 1 because they are designed as experiments, not as power plants. For example, ITER has 4 different ways to heat the plasma, to see which works best at these temperatures & densities, while a power plant would need at most two. Power plants would also need a lot fewer measuring devices looking at the plasma.
Hey, could you recommend a source that really focuses on the power plant side of things? For example, how did they come up with the optimized power plant numbers for magnetically confined fusion, and do these change a lot for different methods like inertial confinement or electrostatic confinement?
By way of example, the electrostatic people frequently make the claim that their plants will be able to use direct energy converters, which have much better efficiency than any of the boil-water-for-turbines options. At least intuitively, this seems like a big advantage (assuming, of course, the claims are true).
The Future of Fusion Energy has a chapter on the power plant side of things. I don't have it with me right now, so I'll respond again tomorrow after I look at their references.
A lot of the power plant is already standard boil-water-for-turbines. I am not an expert on how this works, and I don't think that there is too much innovation from the fusion community here. I have heard the idea of building fusion reactors in decommissioned nuclear (fission) power plants so we can reuse a lot of the plant.
For magnetic fusion, the main innovation in the rest of the power plant is the tritium breeding blanket. A fusion reactor will be surrounded by a layer of lithium, which absorbs the neutrons and produces tritium: Li6 + n => He4 + H3. ITER will have multiple Test Blanket Modules to see what works best.
Less work has been done on the rest of the power plant for inertial confinement fusion than magnetic confinement fusion. The main experiment, NIF, is really a weapons lab that publishes some energy-relevant work on the side.
The startup Helion has a very different electricity generation mechanism. They want to use the magnetic field generated by the expanding plasma (after the fusion has occurred) to directly induce a current. They publish less than ITER, but they have some patents that might be interesting. I think that Helion is unlikely to get First Fusion for other reasons, but their electricity recapture might be interesting.
Can you link some of these electrostatic people so I can see what they're talking about? The most common electrostatic confinement method is the fusor, which can be used as a neutron source, but (almost certainly) won't get net fusion energy. To get net fusion energy, you need to have the nuclei confined for long enough to collide a few hundred times. Fusors only confine the plasma for a few collision times.
On power plant design: you might check out Principles of Fusion Energy by Harms, Schoepf, Miley and Kingdon. It's a bit older but it does a good job of introducing the basics at an advanced undergraduate level.
It came out recently and I have not yet read it, but there is also "Fusion Reactor Design: Plasma Physics, Fuel Cycle System, Operation and Maintenance" by Okazaki.
What do you think of the focus fusion approach (https://lppfusion.com/)? I've kept half an interested eye on it since running into one of their researchers at a conference a number of years ago and having a nice chat, but it is far enough out of my field of expertise I've never really been able to evaluate if they're just crazy or not, and I rarely see the approach brought up in wider discussions of the fusion research landscape.
I should read through their 6 relevant papers [1] before really saying anything definitive, but here is my impression from their website [2]:
I am very skeptical.
I do respect them for being open about the scale of the problem. They need to increase the energy produced by fusion by a factor of 100,000 from their current experiments. Their executive summary [3] also clearly shows that their triple product needs to increase by a factor of ~5,000.
So far, they've gotten high enough temperatures, but too low density. This is especially a problem because proton-boron fusion requires an almost 50 times higher n t (density*confinement time) than deuterium-tritium fusion.
Their strategy is to use a sequence of plasma instabilities to compress the plasma. I don't think that this is a good strategy. You have less control when most of the process is self-generated. I also think that the method will stop working as the density increases. Magnetic field-driven instabilities are likely to compress a plasma, while pressure-driven instabilities almost always expand a plasma. When the density is higher, pressure becomes more important compared to the magnetic field.
This is a pulsed reactor: the cycle of compression - fusion - expansion - electricity generation happens in sequence instead of holding the plasma in a steady state. In other pulsed fusion plans (e.g. NIF or Helion), you would need this to happen about once per second to make a power plant. NIF is currently at about once per day, but they're really a weapons lab, so they don't care as much. Helion has been trying to improve shot frequency, and are currently at about once per 10 minutes. LPP doesn't talk about shot frequency on their website, so I assume that they're not close to once per second.
An Indispensable Truth is a good starting point if you don't have a plasma physics background. Plasma Physics and Fusion Energy or Ideal MHD by Freidberg are good if you are going into fusion or are trying to understand it more deeply.
By the way, you would likely need Q of at least 10 to make fusion economical. And it is probably more likely that you would need a steady state device and an ignited plasma (so Q = Inf) to get a massive reduction in the cost of electricity.
There's a lot of movies with an obvious formula and archetypal characters, but where the film can still be really fun and appealing if the execution and acting are charming. "Sing 2" is one of those films, and so are a lot of romantic comedies.
If we develop the technology to emulate a human mind from scanning brains, and then eventually to just create entirely original minds based off what we know about brain mapping, etc in digital form, do you think we'll try and "resurrect" people who died before the technology based off of whatever we know about them in records? Like will we try to create a plausible version of emulated Mozart based off his writings, his music, depictions of him, and what others wrote about him? I wonder how you'd test them for fidelity.
I ran into a weird anti-fusion power argument the other day. It argued that because fusion produces a ton of neutrons, you could use it and some uranium oxide to produce nuclear bomb fuel. Which . . . aside from the challenges of actually concentrating the fissile material, seems funny to me because if you've got 1)Deuterium/Heavy Water, and 2)Uranium Oxide, and 3)Really want fissile material, you could much more easily just make a heavy-water nuclear fission reactor to get it. You don't even need pressure vessels! That's actually how India got the fuel for its first nuclear bomb - they had a test heavy-water reactor for research purposes.
Speaking of fission, the World Nuclear Association estimates that the "overnight cost" of construction per kilowatt of electricity generated for nuclear plants in South Korea and China to be about $2157/kWe and $2500/kWe respectively, and it was at around $1500/kWe in the 1960s in the US. That's a lot lower than what it is now in the US, but still higher than what the US EIA estimates for "Solar Photovoltaic power plus storage" ($1612/kWe) except in the case of the 1960s US plants, and even then the difference isn't huge.
I bring that up because it makes me skeptical of the argument I've seen recently about why nuclear power hasn't come back heavily in the US, which tends to focus a lot on US regulatory changes. If even countries with far more favorable regulatory and political environments for a mass build-out of nuclear power aren't beating solar PV and storage, then it suggests there's other reasons at play - nuke plants may just be really expensive to build unless you're building a LOT of them.
On fusion and uranium enrichment: fusion plants will still likely require some sort of safeguards/inspection regime, but since they should normally have approximately zero uranium or plutonium around the inspection for that is much easier. It's much easier to find the presence of some radioactive isotopes when there should be none, than it is to track the flows of uranium materials to ensure that a part in a thousand is not being diverted for nefarious uses.
Yeah but on the other hand the fission numbers are real, while the massive PV + storage numbers are speculation. Cf. the massive difference between the sales-brochure predictions and deliverable reality of Ivanpah. You always want to keep your salt shaker handy when listening to investor pitches (including utilities and governments pitching you as an electricity ratepayer).
" Like will we try to create a plausible version of emulated Mozart based off his writings, his music, depictions of him, and what others wrote about him? I wonder how you'd test them for fidelity."
You could do it recursively. Emulate modern people who are known-accurate, then emulate one generation in the past and have them interact with their emulated descendants to test for accuracy. Then emulate two generations ago and test them against your verified one generation ago, repeat until you make it back to Mozart.
It wouldn't really be accurate, the system would surely end up swamped by noise, but it's a neat way to at least maintain consistency.
Delete them and try again. You have some idea of how they weren't accurate, and presumably there are some parameters you can tweak to make version 2 come out different.
But you can't just "delete them", as they would be people in their own right, however a-historical they might be. You wouldn't propose creating 1000 children and then "deleting" the 999 children who are least like a target historical figure after evaluating them as adults. What do you think is the relevant difference here? That the people in questions are simulations and presumably exist on computer hardware?
One of the big reasons death is bad is that people are expensive: it takes a lot of resources to make a person and that investment is lost when they die. Once we have the technology to spawn new digital people on demand that's no longer true. If we value a lot of sims existing then our only constraint is how much computing power we allocate to running simulations, a dead sim can be trivially replaced.
If any deleted sim will be replaced, what is it you value that's being lost when a sim dies?
I would say that the human rights of the sim are being violated if they are deleted without their consent, regardless of the economic value lost. So what would be lost in that case is our framework of human rights and the comfort of knowing that you yourself are unlikely to be involuntarily deleted. Perhaps you’re thinking about “deleted” in a different way than I am? What, specifically are you imagining happening?
That would amount to mass murder and defeat the whole point - which in my view was to protect the original mind from truly dying. If you do it by creating and destroying a 1000 similar-but-substantially-different minds, you didn't exactly decrease the total number of permanently dead people.
If we're troubled by the nonexistence of minds, what privileges minds that happen to have already existed? Those 1000 similar-but-different minds weren't alive at the start of our experiment, so why is it bad for them to be dead at the end of it?
For roughly the same reasons it's bad to kill biological humans, give or take. If you subscribe to some moral theory saying it's actually ok to kill people, I feel like you need to state it explicitly and argue your case, because this is a highly unorthodox view.
I'm doubtful we have enough information for people from the Mozart era, even for the most famous ones, to get anything remotely accurate. But for those who lived in the digital era, with all the photos and videos and messages and a good part of their day-to-day movements recorded, it can perhaps be sorta possible? At worst I hope this method could be helpful to augment another approach of transporting a consciousness into the future aka cryonics. I know some people do extensive life-logging because of this consideration, and considered it myself, though for now decided against it.
Consider the "Grandpa Joe" argument (https://www.youtube.com/watch?v=29_uSlEEPSk). You could interview grandpa Joe for 20 years, record every moment of his miserable life, and still not really understand how he would feel when he gets the "Golden Ticket", because he himself doesn't know, he can't imagine what it would be like to have real hope.
One of the tragedies of these predictive-model methods of resurrecting a person is that they will observe that the person is miserable at the end of their life, and then predict that they will continue being a miserable person even after they are resurrected.
Whereas I think that some of the people who are actually recreated in the future will experience a more profound sense of joy and safety than they have ever felt in their lives, and their reactions (and how they choose to live the rest of their lives) will be extremely hard to predict even given extensive knowledge of their past.
Grandpa Joe is a funny example, because you add another layer to figuring out your proposed person - deception. Grandpa Joe spends 20 years lying in bed, and can suddenly walk when there's an opportunity? Grandpa Joe hasn't gotten out of bed in 20 years, but he somehow buys Charlie a chocolate bar (and his tobacco)? Did he steal the money for that, by the way? If not, where did he get the money from?
Not to make this about a fictional character, but the reality is that Grandpa Joe has a lot of things he really doesn't want people to know about him. So he lies about some parts of his life, and keeps other parts quiet. That's going to make an emulation of him significantly harder to get right.
Ha! Good point with Grandpa Joe. I've personally always seen the story of Willy Wonka as existing on a more allegorical level than I think you're seeing it. In that interpretation the Golden Ticket has magic that "fixes" Grandpa Joe's story by altering his destiny at a narrative level. One consequence of this is that he is diegetically healed and can get up and walk again. He really had felt miserable for the last 20 years, and he really couldn't walk, and he really did experience a miracle of healing.
(though he is still is a bit of a rascal who proves himself unworthy of inheriting the chocolate factory!)
Under this interpretation I feel that it's a better analogy for how someone might feel after successfully being preserved / uploaded.
I'd say you are correct, that his sudden change in circumstances may very well have a sudden change in his behavior and even abilities. So do you model Mozart based on how he would have reacted right before getting his "golden ticket" or with the idea in mind that he may react positively?
Do you model Van Gogh as the guy who cut his ear off and later committed suicide, or do you model him as the guy who would love the 21st Century and have a different outlook on life? Would modeling him one way instead of another change something fundamental about him? Fun-loving-happy-Van-Gogh may (or may not, who knows for sure?) be accurate for the person, but does it defeat the purpose of modern people bringing him back to meet the "real" Van Gogh?
I think that if you're valuing the people from the past as ends unto themselves then you would end up just clearly explaining to them how the world has changed and help them to make of that what they will while respecting their fundamental human rights.
I imagine that you might see some very interesting groups form around shared pasts, while still being modernized in their own ways, and that that might end up being more interesting anyway than making a bunch of fake unrealistically grumpy people! For example one thing a person might do upon waking up and learning that their religion doesn't exist, after the initial shock wears off, is to get together with other people from their time and start it back up!
...it's interesting that you mention Van Gogh in particular, as this scenario has been partly explored in Dr. Who (https://www.youtube.com/watch?v=ubTJI_UphPk). To me this was a pretty powerful scene, and it felt profound and even just that Van Gogh would have a chance to be in the future.
> does it defeat the purpose of modern people bringing him back to meet the "real" Van Gogh?
IMO, If, after seeing the future, Van Gogh wouldn't want to paint again for another 100 years, that's OK. He doesn't owe anyone anything. In fact the future people have an obligation to help him flourish in his own way. If the people of the future can bring him back without an expectation that he "owes" them, simply a desire that he be, the same way they already bring children into the world, then I think that's the right place to be in to contemplate these things. I think that's what being a good steward of these souls looks like. Whatever way the people of the future treat themselves and their past, that's the precedent they're setting for even more powerful, vast, and alien people in the even farther future to deal with them.
You're talking about "predicting" reactions in a way a simple modern-day ML system does, I don't think there's much value in it. What I have in mind is that if we had an immensely better understanding of the internal workings of human's mind, and huge amount of computational resources to throw on the problem,we may be able to some extent "reverse-engineer" the mind based on external obaervations. As an analogy, imagine for example that you're trying to understand how a certain WWII-era plane worked. Not a single copy or even a blueprint survived, but you have a lot of photos and video footage, and reports, and field manuals, and interviews with the people who've flown it, and you also know a great deal of the engineering practices at the time in the company who built it. Will you be able to recreate the plane in every detail? No. Will you be able to build a plane that is functionally identical, have the same layout, same specs, but maybe some arbitrary little details are different? Maybe? With enough information and effort and some luck, I think you can get close. And if you fly this recreated plane in some conditions of which there's no records for the original, it'll likely behave the same way as the original would.
But I recognize that humans are far more complex than airplanes, so even a very detailed life logging probably won't be able to constrain the mind-design space narrowly enough to pinpoint a single mind. This is why I'm talking about using it to augment cryonics, which preserves all or most of the information in the brain, but there may be losses due to imperfect procedure or storage.
I don't think the airplane analogy applies to human minds in this case, because you can mostly figure out the different situations the airplane is expected to be in and what it's expected to do in those situations. But people experience really new situations all the time that don't look like previous situations you've observed them in, and they have a lot of hidden depths like grandpa Joe that seem to me to be beyond the ability of language to actually derive.
To make it concrete, you can ask me as many questions as you like, and look at all of my social media posts, but you still can't ever get at even a relatively simple, legible hidden layer of my mind like the password I use to unlock my computer. Good luck predicting how I will behave when I learn that my country doesn't exist, or that no one believes in my religion anymore, or that the world really is 1000x more just and safe than it was when I was previously alive. My response to these events relies on hidden layers of my mind that are even more difficult to infer from observing me than my password is!
I would change my mind if you could convince me that it's plausible to do some kind of life-logging that would be able to infer my laptop password; do you think something like that is likely in the future?
No, I definitely don't believe this level of depth can be inferred from the information that normal people store on digital devices. Although to be fair your password or at least its hash is definitely stored somewhere, and we can probably assume the future mind-simulating civilization will be able to crack modern hashes, but I totally agree with the broader point. An extreme version of life logging where you really honestly videotape your every waking hour *and* keep a diary with your thoughts can maaaybe come close, but I'm extremely doubtful and the point is moot because nobody would do that. So as I said, I don't believe it's practical as the primary means of mind preservation, but it can potentially provide necessary context or missing pieces for someone trying to repair the damage done by imperfect cryopreservation.
One abstract point where I disagree is that there's qualitative difference between this and the plane example. Both minds and planes are fundamentally physically deterministic, so if you had enough information to pinpoint the *exact* design, this design will behave, and be, fully identical to the original. (I include memories and life history in what I mean by "design" here) The difference is that you need vastly more information to pinpoint the exact design of a mind, much more than even a very digitally active person can plausibly have recorded about them.
I suspect that we perhaps don't disagree about anything then, since I also agree that with "sufficient information" you can of course recreate someone. Though I don't think that I'm attacking a strawman either, since there are actual projects such as Terasem (https://terasemmovementfoundation.com/) which work to make a "Mindfile" and eventually recreate a person through, essentially, a detailed personality quiz.
One thing that I think is key here is that language is almost certainly incapable of pinning down what a person is in enough detail to recreate them, mostly because the subconscious is relatively inaccessible to language. It's not that the details don't EXIST in some physical form, just that language is inadequate at getting at those details. It would be the equivalent of taking a photograph of a ROM chip in our metaphorical plane, but not being able to read the firmware off of the chip.
The only thing I can think of in today's world that can capture enough detail would be to preserve a person's body and physically retain all of that language-inaccessible information encoded in their brain / nervous system.
Just embarrassed myself posting in MR comments thinking it was this blog. But I'll repeat the same point: if you review Nixonland please also read Before The Storm. It sets up Perlstein's hypothesis and yeah, he hates Nixon, but he also hates LBJ and kind of has a soft spot for Goldwater.
Goldwater was just a weirdo. You can pretty much just copy everything about Ron Paul 2008 onto him. Would he have been a good president according to Perlstein? No, absolutely not. But he really believes in his message and is sincere.
LBJ according to Perlstein is a nervous miserable wreck, fraught with doubt about everything and massively insecure. He tries to do good, but mostly he tries to get re-elected, and will use any excuse to get closer to that goal.
Weird. Never heard anyone describe LBJ as a nervous anxious wreck. Ornery vengeful bastard, yeah. Out of his depth in the Presidency, sure. Baffled by the leftovers (both voters and staff) from Camelot, OK.
Is there some kind of SSC/ACX Readers demographic survey that's been done and that someone could link me to? Would be curious to peruse such a thing in general, but am especially curious about what percentage of Scott's readers follow a (sincere, traditional) religion.
re: Engines of Cognition, (a) ordered and looking forward to having a physical copy of some LessWrong essays, and (b) distraught that I missed out on the previous offering of A Map that Reflects the Territory — will this ever be in print again? I can’t find it in stock after a cursory search online.
Hey daltonmw! If enough people request more copies of A Map That Reflects The Territory we may do another run. Book printings are pretty expensive and a bit time consuming so we haven't done it yet, but if enough people request it we may.
Great to hear! Totally understand the circumstance, not to worry. If I could make a suggestion and it’s at all in your power to influence, perhaps adding some kind of a “Notify Me When Available” mailing list here https://www.lesswrong.com/books/2018 would be an easy way to gauge interest in another printing run. And perhaps Scott (I think he hangs out around here too, wink wink) could signal boost the idea.
I'm an editor in the relatively lower-stakes world of sports journalism, but the posts about legible expertise from a while back still haunt me. They stoked my existing hunches that there is no way to reconcile a publication's ceaseless pursuit of growth and attention, and its ostensible goal of producing truthful stories that also give their readers a clearer sense of the world (to say nothing of my hunch that most breaking news reporting is inherently immoral).
What I'm trying to figure out is: Are there any steps that can be taken to drive progress, however incremental, in this area? I hold no illusions about fully disentangling journalism's competing interests, but I remain hopeful that not all is yet lost.
This presupposes people want to be lied to. The only reason there's an inherent conflict is if there's significant market demand for stories that are not truthful or helping inform the reader. If that isn't true then there's no inherent conflict.
That doesn't mean misinformation wouldn't exist. It would simply mean that the cause would instead be something else. For example, that people have trouble figuring out where to place their trust. That's the case with a lot of education. It's not that there's inherently some conflict between student's desire to learn and teacher's desire to teach. But students are, definitionally, not in a position to judge the quality of education in a particular subject. And teachers have every incentive to portray themselves as maximally capable of teaching to attract money and students.
I do think there's market demand for stories with narratives (or just opinions) that inspire strong emotions in readers. They're what many people are understandably drawn to. I don't believe that means they "want" to be lied to, in the sense that there is a conscious desire to seek out misinformation.
I suppose what I'm gesturing at is "falsehood" not just in terms of the words on the page, but also how a story alters one's model of the world. For example, if someone sees a lot of stories about young people dying of Covid, they could get an inflated sense of how dangerous Covid is to that demographic. The stories are accurate and they inspire strong emotions, but they may not improve overall clarity- and journalists can't help but write them.
This seems to fuzzy to test against. Aside from, "I don't like the narrative," how would you rigorously determine whether a story improves overall clarity or not? At scale?
It's true - quantifying would be incredibly challenging, perhaps impossible, and that's a big part of the problem.
I think your other comments are helping me understand your thought process a bit better, and you may be right that there isn't a conflict between truth (as you define it) and success. Instead, maybe it's a conflict of media incentives + our human inability to properly update based on new information vs. overall clarity. Good journalists' jobs aren't about giving you a story and telling you how much it should matter to you, they're about telling a story in as compelling a fashion as possible, within the bounds set by journalistic standards and reality (which gives journalists plenty of wiggle room re: story selection). And we certainly know why it's impossible to expect individuals to give new incoming information proper weight. Don't think that helps with testability, though.
Exactly - journalism constantly tries to maximize reader interest, constrained by the facts. In general, the fewer constraints your maximization problem has, the more you can maximize it.
Yet if they weren't constrained by that requirement AT ALL then you'd expect the media to make things up wholesale. That rarely happens and when it does it's usually punished. Instead what happens is exaggeration or just barely plausible interpretations or laundering sources. So clearly they are constrained to it in some degree.
The question is where that constraint comes from. Since it's clearly advantageous for the news writers to be able to make things up but disadvantageous for the news readers to read something made up I think it's pretty clearly coming from the readers. The question is then... what I said at the start. To what extent is there demand for truthful stories that inform the reader.
You can imagine a world where it's relatively low and they're actually happy with the world we're in. Though if that's true, why do news outlets have such low approval ratings? Or you can imagine a world where it's some form of structural failure. For example, maybe they have no good way of vetting the news or maybe some effective cartel has been captured by interest groups or something.
I think the danger here is a few false dichotomies. "People want interesting stuff" v. "people want the truth" (they want both). "News sets out to deceive" v. "news sets out to inform" (As a general rule the journalists I know desperately want to inform and also desperately want to have enough money to keep informing and also are entirely unaware that there's a difference between facts and the conclusions they personally draw from those facts).
The biggest issue is that we have minds that crave narrative and society is now large enough that any given narrative is probably misleading. If you want to believe that kids are dumber today, a national news engine can find 6 stories about kids doing dumb stuff in an hour. There's tens of millions of kids in the U.S., you just gotta look around. But to get to the bottom of the "are kids dumber today" issue, all that is irrelevant. Instead you'd have to take all recorded instances of kids doing dumb stuff from the present, all recorded instances of kids doing dumb stuff from the past, control for a bunch of stuff and work out per-capita whether any given kid is more likely to eat a Tide Pod (or the 1980 equivalent). Such reporting would no longer be interesting and it wouldn't satisfy journalists or readers.
I think there's a kind of weakness of will in the readers. Over the long run, they want stories that tell them the truth and help them understand the world. Over the short run, they get their attention won over by stories with good narratives and the right sort of poetic justice.
It's the same problem you have with sales of "healthy" food, where people do in fact want the thing with the long-term benefit, but keep making short-term choices that subvert that, and the companies aren't incentivized to figure out the long-term benefit, but just to figure out how to keep appealing to the short-term incentives.
Inherently no, but practically yes. When there's a hot market competition for writing *anything* that elicits emotion, and when most of the things one could possibly say are false, you're inevitably going to end up with writers in a situation where they have to choose between saying one boring true story and ten false-but-exciting stories.
Another way to think about it is that writers are trying to differentiate themselves. But the truth is not differentiated. If you're trying to stand out then the truth doesn't help you; anyone can report the facts of the story and their article will look like all of the others. So where can you look to find interesting things to say that no one else has said? That's when motivated reasoning starts picking up some of the epistemic slack.
If this were true then wouldn't organizations with lower journalistic standards outcompete those with higher? Is that actually the case? I'm not sure it is. It could be but it seems like CNN outcompetes the National Enquirer.
Reality is a constraint. If a story about a demonstration is most moving if there are fifty thousand demonstrators but there were only twenty thousand, a reporter who insists on telling the truth will write a less moving account than one who isn't. If a one-sided bad guy mistreated good guy story is more moving than the complicated interaction that really happened, similarly.
Sure. And yet if this were infinitely true then people would willingly lap up regime propaganda which is designed to be as maximally moving without truth constraints. That is not the empirical case as far as I can tell.
There must be some reason that American news outlets do not go full Pravda. If it's because the consumers will punish them for it then the problem is not that consumers don't want truth. That doesn't mean there isn't a problem. But it does mean the idea there's an effective tension between truth telling and popularity is not necessarily true.
Or that people have limited knowledge (particularly of newly-breaking developments) and don't particularly care to do the work to check the factual accuracy of stuff they read. Or even that people think that (plausibility and/or truthiness) + (affirmation of existing priors) > (factual truth).
The question is, do people read journalism like a textbook? Or like a novel?
I believe I brought up this alternative elsewhere. There's definitely a problem but the question I was responding to is whether insufficient demand for the truth is the problem.
Responding to your general point in this thread: People want things that they think are both true and interesting. As you note, this stops most outlets from simply making up random falsehoods, you can't get far reporting that WWIII is occurring. But there is great tension between truth and popularity within the narrower range of "things that are false but can be convincingly lied about".
Give everyone $100 that they must spend on news media subscriptions of their choice. Increasing the time distance between deciding on the sort of stuff you want read and then actually reading it will cut down on the click bait.
I personally think maybe we worry too much about this issue, the concern is perhaps a transition from traditional media to the new way of consuming ideas via non mediated sources. Eventually we will simply accept traditional media as untrustworthy as the rest and let reputation drive which data and sources we trust. This will be harder for older people, for whom when a prestigious newspaper prints something it means it should be paid special attention rather than a small update in ones priors.
Sometimes I wonder this too. But I'm not convinced. There seems to be a fundamental variation in the ability for untruth to persist across different domains. For example, in physics, a meme like flat earth cannot persist in perpetuity (outside of the fringe). At one point in human history, it was maybe the dominant belief. Now it's extremely fringe, because we can put a camera on top of a rocket and we derive concrete benefit from being correct about the nature of the solar system.
But on various social and political issues it appears there can be no true reconciliation. "Does Donald Trump respect democracy?" For me, Jan 6th was a completely satisfactory answer to that question. But did all the people who claimed otherwise see a reputational hit by Jan 7th? Maybe to some degree, but in general no: those people have as high of a reputation among their audience as ever.
I think COVID gives us a good bound on "how much feedback is necessary for social consensus to converge?" More than a ~1% death rate (or whatever it is).
The conventional wisdom is that this is where competition and The Invisible Hand work their magic. In principle a publication in savage competition with many other publications will have a hard time bullshitting their readers with enticing but false stories, because their competitors will take them down with glee. Everyone wants to print rosy lies, let us say, or is at least driven in that direction by the hunger for growth, but when truth is the rosiest story that *can't* be destroyed by a vicious competitor, that's the market-clearing product.
Of course, there are two obvious caveats to this, even if you believe it in general: (1) you have to *have* competition, so you can't be living in a world where a few huge organizations control most publication in the area, and (2) it probably doesn't work for those stories in which *everyone* wants to be lied to, e.g. stories in which human mortality is a central point.
I'm a traditional horary astrologer who enjoys practicing his art for strangers on the internet. If any of you have a question on just about anything you'd like examined via the toolkit of real astrology, drop me a line at FlexOnMaterialists@protonmail.com. Materialists and even atheists are, despite the email address, quite welcome. I look forward to hearing from you!
I'm going to post my predictions for 2022 soon, and if history is any guide then Zvi and the prediction markets will try to beat me on the global ones. Others will be about my personal life. If you want, I can give you my birth date and any other information you need, and you can try to beat the markets on either category.
A fine idea. There are some techniques which use progressions based on the natal chart to determine personal experiences in a given year of life; I'd need your birth date, birth time (or reasonable approximation), and birth city to check out your nativity, but that's only likely to be relevant for your personal life predictions. For everything else, I'll definitely watch for whatever might be amenable to horary prediction and take my shot along with the rest.
Huge fan of your work since the SSC days btw. I'll try not to gush too much in any personal correspondence.
November 7, 1984, 11:35 AM, in Santa Ana, California. I'd also be interested in you telling me what experiences I might have this year before I make predictions, so that it's unprompted.
EDIT: People on here have often asked how Scott is so good at writing. The answer is simple: it's due to his Mercury (planet of thinking, communication, intellect) in the Tenth House (house of Ambitions. Motivations. Career, achievements. Society and government. Father or authority. Notoriety. Advantage):
Mercury in the 10th House
Mercury in the tenth house tells us about achieving personal goals and ambitions. These people express themselves well both orally and in writing, and they can use this ability to achieve success and recognition in their work. Their job is important to them and their career is carefully and thoroughly planned, including education and preparation necessary to achieve their goals. Mercury brings the ability to present ideas and opinions to the public, and it often generates a talent for making great speeches, publishing or political strategy.
Ah well, that was quick'n'dirty using online chart creation, and the website must be using a different system to yours.
I *am* surprised Scott is a Scorpio, given the general (often overblown) reputation of natives of that sign. Must be a lot of influence from his Rising Sign! 😉
(My natal sign, as you can guess, is Gemini but I have a *ton* of Taurean influence in my chart and it really shows).
FWLIW @Deiseach, I looked up the astrological descriptions of myself and folks who’re important to me, and in many cases the descriptions were frighteningly accurate, with one important exception:
My wife.
“That’s not her at all!” I thought.
But then I looked up our “compatibility”, and: “Oh jeez, that is us!”
Memorably, near the end of the description was:
“Is this an ideal couple? No way; in fact, it's a dumpster fire”
And on that: thank you again for your advice, previously I’ve mostly heard either (stay with your wife, divorce will cost you too much money and the kids, visit whores if you have to” (usually from men) or “She’s a monster! Follow your heart, leave her and be with your true love instead” (usually from women), and while you made some wrong guesses about the history I don’t think that altered the core of what I needed to learn from you, and because of your uniquely unclouded atypical viewpoint it was you I wanted to learn from.
One thing I learned from the disparate advice from others is how much most women really hate housewives, most men regard them as doing a job, most women don’t, the opposite of what I’d expect.
I'm glad the advice wasn't completely rubbish, but I do think that what you need to do first and foremost is sort out what *you* want, and then decide about (a) your marriage (b) any new relationships.
I'm not thrilled by divorce, but sometimes ending a bad marriage is best for all parties concerned. Don't rush into a new, rebound romance, though.
Can I ask what do you think of Indian astrology (use precessed equinoxes so overcoming the objections in articles like this one: https://www.livescience.com/4667-astrological-sign.html , e.g. under traditional Western astrology I'm Gemini but in Indian I'd be Taurus) versus Western astrology (derived from/based on Babylonian, Greek and Egyptian influences)?
Given that mediaeval erections of horoscopes use the same set-up as Indian kundlis, rather than the circular chart layout of modern astrology, did older astrology fall more into line with such practices in your opinion?
I haven't studied Vedic astrology myself; what I've gleaned incidentally makes me think it's an excellent system when approached with determination and skill. I've also heard of a few instances where Western and Vedic astrologers examine the same query and get largely similar results.
Re: square vs circular charts, the data is the same, only the shape is different. Medieval Arabs were using both more or less interchangeably back when giants in the tradition like Mashallah and Sahl ibn Bishr were actively practicing.
The precession of the equinoxes is often used to attack modern astrology, but was actually known to astrologers > 2000 years ago. For example, the Greek Hipparchus circa 129 bce, and 2nd century Ptolemy 'the Alexandrian' wrote about it in his Tetrabiblos. Maybe the Chaldeans before then?
A person who believes in the sole existence of matter and energy. Your comment made me realize there's at least a hypothetical category of "agnostic"(?) materialists--someone who doesn't know whether or not the physical world of matter and energy is all that exists, though I expect this is fairly rare.
Likely the same reason that agnosticism is rare compared to theism/atheism: it's exceptionally difficult to wait for evidence or keep your thoughts cautiously neutral in the face of such a difficult and fraught proposition.
Don't we have to be a little careful here? On almost all factual data, almost of which is by definition unmeasured, I think people are perfectly comfortable with agnosticism. Who will win the Superbowl 150 years from now? Did Abraham Lincoln have a comfortable crap the morning he was assassinated? Did your or my great-grandparents prefer to do it doggy or missionary? Et cetera. About these and an infinity of other factual questions, most people are quite happy to shrug their shoulders and say Who Knows?
Not on profoundly important existential questions, no, but arguably the problem here is not that they are disturbing per se but that you have to *move forward* if you are not a stylite philosopher who can spend his days ruminating over the nature of the infinite. You have to at least act as if you have satisfactory answers to questions of Is It All Worth It? and What Gives Life Meaning? so you might as well just pick a functional answer on grounds of taste or convenience and get on with life.
That isn't especially cowardly, it's just practical.
Hmm... would I still be a materialist if I believed in the number 3? Like, I don't think you could go out and punch a 3, but I still think math is real.
What about linguistic prescriptivism? Or objective moral truth?
Apart from that, your definition seems like a good one. My only real caveat is what if some new thing is discovered that isn't matter or energy, but is still described by simple mathematical constructs with no privileged reference frame?
I don't know if there is such a third thing or not. But I would hardly call myself agnostic about my materialism, other than the fact that 0 and 1 aren't real probabilities.
IMO the materialist would say abstractions like 3 only exist as signals in the brain; you could tie yourself in knots trying to reason out whether or not 3 existed before some mind counted that high, but ultimately the expectation is that 3 does not have meaning outside of a thinker thinking. The non-materialist would say 3 has an objective existence which transcends the cosmos of matter and energy.
On a related note, I know several groups who buy (or would buy) successful predictive power. These range from actuarial to stock picking to whatever. If you can show any consistent predictive power in any of the events they're interested in then they'll be happy to pay you large sums of money. If you're not money motivated then they'd be perfectly happy to donate it to charity or something. And it would absolutely convert a lot of people who think astrology is BS.
And on the flip side, if you refuse to engage your "talents" with people who have strong incentives to rigorously test your predictions, that strengthens the priors of people who think astrology is BS.
Given that I assume chemists are real and astrologers are charlatans (reasonable or not, I typically assume that):
If a chemist is getting paid a lot, I infer that he's very good at chemistry. If an astrologer is getting paid a lot, I infer that he's a particularly craven and insidious liar.
These aren't necessarily reasonable inference in all cases. The point is merely that getting paid won't always overcome the bias against astrologers.
I recently watched this video (https://www.youtube.com/watch?v=rvLFEh7V18A), which explains the recent retro video game bubble, where the top auction price of old, high quality video games jumped from $30K to $1.6MM over the past 4 years. According to the video, a group of connected people (including a specific auction house) have been selling retro games between one another at increasingly higher and higher prices. These circular sales give the impression to the general public that the value of retro video games is rising very fast, which creates extra hype, which causes more people to flood the space and buy these retro games for higher prices, with the hope they'll be able to sell the games for an even higher price in the coming months/years. Since retro video games don't have much intrinsic value (games can be played for free on an emulator or bought cheaply on a virtual console), the price they sell for at auction is mostly a function of how much hype exists around them, and how likely buyers think it is they will be able to flip the game for a higher price in the future.
Do any of you think this practice could be even more widespread?
The video says something like this may have happened in the comic book collecting space. Additional places where this practice seem possible to me include sales of fancy art, sales of NFTs, and potentially the prices of some crypto coins. Sales in all of these places (especially in the crypto space) can be very difficult to trace, making circular sales just to bid up the price very doable. In addition, valuations in all of these places (especially in the crypto space) are incredibly dependent on hype and the belief that prices will continue to go up. While there are some safeguards in place (e.g. auction houses have some controls on who buys and sells products there, and some crypto exchanges have identity verification), it doesn't seem like they would be enough. In addition, such circular bidding does not need to go on forever - it's just needed to get some initial hype - afterwards you can let the market do the rest.
all of the classic methods of fraud in that video have been heavily used in all notable NFT markets, alongside many others that crypto makes easier, yes
This is a variant of wash trading, a very common, and where regulated illegal, market manipulation tactic. I expect it to be all over the place in unregulated markets. As you say, it has two advantages for the manipulator: generating fake trading volume and a fake price. Products which young people trade are more sensitive to start hyping from those initial impulses because there are not many data points in the past 5 years of hypes crashing hard enough to painfully punish early investors. It's a problem.
You don't even need to collude directly if the initial majority manages to cooperate on the prisoner's dillemma by playing playing bid/bid.
Sneakers, comic books, trading cards, crypto currencies, NFTs, in-game items. In a more abstract variant also the hype-stocks like GameStop, AMC. Also Tesla, Amazon, etc.
It's wrong but hard to bet against. Most trades I can come up with have very high variance relative to the return they should eventually provide. "The market can remain irrational longer than you can remain solvent." The long long-run should see these prices eventually correct, but it might easily be a matter of several years if not more.
Question for anyone that hires experienced software developers. I am trying to demonstrate soft skills in front end web development. This is not always easy to do in a short interview. Would you be convinced by a series of blog posts on a candidate's personal website?
For what country? I can only answer for UK hiring.
What are "soft skills in front end web development"? Do you mean having artistic sense? Incidentally if you could write a blog post on how to have good sense that would be nice, there's not much in that space that I'm aware of: even the lowest level stuff like "every layout" gives you a series of components to build with and tells you why they work, instead of telling you how to figure out whether something looks good (without showing it to your friends) or design something that looks good.
Blog posts affect my decision if they're good, in the same way that GitHub repositories might affect my decision if they're good. A generic React tutorial is neither good nor bad, a bad post is negative, a good post is positive.
I would expect an interview to be more valuable for soft skills than blog posts would. No matter how much you've written about communication, the interview will tell me how good you are at communicating.
I also expect most people interviewing to not read your blog posts, especially if they're long.
1. Asking better questions from other devs in the same organization.
2. An easy way to be a better team member (taking good notes).
3. Common ways that designers break web standards, and how to have a conversation about it.
I don't know if most interviewers are even looking for good communication skills above a bare minimum. I don't think my technical skills and experience make me stand out very much. So part of my strategy would be to try and sell interviewers on the importance of these other skills.
Your strategy expressed here requires three things:
1. Write good blog posts about soft skills
2. Convince interviewers that good soft skills can make up for technical deficiencies
3. Convince interviewers that you personally have good soft skills, using the posts written in 1.
I think this is going to be quite difficult. When I think about the people I knew who I would have hired on the strength of their communication, they did a lot of things -- for example, they were easy to talk to and communicated information well. If this isn't coming across in the interview, it's possible you don't actually have standout soft skills.
For the given topics, I see the first two as neutral, and the third as a potential negative. Which do you care about more: delivering software that gives value to users (and the business), or meeting web standards? Are you going to be overly picky? There are definitely cases where accessibility (to give an example) is worth the time (and money), but also cases where it isn't. You could talk about this in the post, but any unusual opinions can sink your chances.
When I think soft skills, I think more broadly: Dale Carnegie's "How to Win Friends and Influence People", Robert Cialdini's "Influence", Communication (and feedback) as one of the four pillars of Kent Beck's Extreme Programming. What do you think about pair programming? Code review? Mentoring? Office politics? How much / how little? What's your experience in them?
If you think your technical skills are holding you back, I'd also expect you to be working on improving them. Trying to sidle around to focus on soft skills is not going to be terribly convincing, I'd expect.
Good advice all around. I disagree on the point about web standards and it is part of why I want to write a post. Web standards exist for good reasons and breaking them is often bad for everyone involved. It makes your product less useable for customers, it makes bugs more likely and increases the cost of maintenance, and it often requires writing custom software. So my argument would be it's worse for customers, and more expensive.
And accessibility is also something where I disagree. Usually meeting accessibility standards means your code is legible to screen readers, which has carry over benefits because it makes it more readable for machines in general, which translates to better SEO, easier automated testing, and easier web analytics.
I look forward to seeing your argument in full, preferably with some examples. That's definitely worth a post.
I agree it often makes products less usable. I don't think it necessarily makes bugs more or less likely, or increases or decreases the cost of maintenance. I don't understand the custom software point, though I'd agree if you're writing custom software to do something that can be done natively that's often a bad idea.
I think it's important to consider the audience for the tool you're building. Look at the browsers they use, look at whether they're using screen readers or not (as in your example). If it's an internal tool and nobody involved uses a screen reader, spending two weeks making it screen reader friendly is time you could have spent solving problems for other teams. If somebody arrives who is colour-blind, or has vision problems, that's the time to start focusing on improving that area -- but not before. YAGNI.
In short: improving things takes development time. This is the tradeoff: you can't code everything. You have to prioritise. And if you choose to prioritise standards, you still have to maintain that code you wrote. So it should be delivering benefits worthy of the time you spent writing it and the time you'll spend maintaining it in the future. And maybe it will, but maybe it won't.
If you want better SEO, focus on SEO. If you want better automated testing, focus on automated testing. If you want better web analytics, focus on web analytics. Improving accessibility and saying it offers all this extra stuff -- well great, but are any of them priorities? If they are, they should be their own focus. If not, it's half price on something you didn't want in the first place.
I do look forward to your post though :). My perspective is software consulting, which has a limited budget (client pays for time) and so work has to pay for itself internally to a much greater extent than outside. If you have unlimited time and budget, all this stuff is great to work on. If you don't, tradeoffs need to be made.
I agree with the points about tradeoffs. I have more often worked at companies that are building a web platform that they will have to support indefinitely.
One example of breaking web standards is how modals are sometimes used. I've had designers that wanted to use full page modals, which is basically creating a page where standard web navigation is broken. it is a surprisingly common anti-pattern (I think the reddit redesign had full page modals for viewing comments on a post, not sure if they still have it.)
The mock-ups for these full page modals often look pretty and clean, but they are crap for user experience. And once product owners realize that web navigation is broken they usually want it back. Which leads to the front end dev having to add in a bunch of standard web navigation functionality.
I've never met a product owner that has been happy with the full amount of time and effort to implement full page modals, but if no one knows where things are headed then each individual step along the path of failure seems worth it.
Interviewers likely have a busy schedule. I expect most would click around and glance at your blog posts for a few minutes, then make a snap decision on whether they like your style or not, rather than reading the full thing. Doing anything out of the ordinary is normally a net positive, but given how short the time to impress is, treat it like your CV: perfectly styled, thematically curated and organized to bring across the points (about yourself) that you want.
I'm a FE dev for a US startup, but I don't do interviews. I would say that our interview process already evaluates candidates' soft skills, and our interviewers wouldn't have time to read blog posts.
For what it's worth, I've reduced a candidate's ranking based on Internet writing -- yes, I google everybody, naturally -- but never elevated it. I wouldn't say it's worth the risk. On the down side, if there's *anything* there to which exception can be taken, you get questions raised about your judgment in signing your name to it publically. There isn't much of an upside, as far as I can see, because your ability to craft a piece of human or humane sounding writing given plenty of time, solitude, and unknown assistance is not especially relevant to the question of how well you would interact with colleagues and management in the moment, day to day.
Maybe it's old-fashioned in this day of building your own brand, but I assess "people" skills of candidates by looking them in the eye, talking face-to-face about this and that, and seeing how they react to even the minor twists and turns of an interview process. Are they gracious, polite, considerate, neither easily offended nor a doormat? Big plus. Are they brittle, self-centered, needy, unaware, or see the job as some aspect of a personal crusade? Next!
I am a U.S. based software engineer who has done many interviews mostly for backend candidates. I would say there would be a 50% chance I would see your blog and spend more the 5 minutes on it. If anything I may ask you one question about your blog.
For me one thing that makes a candidate stand out for some of the soft skills you mentioned is if they are able to teach me something in the interview. It definitely hits point 3, and potentially shows you can ask the right questions to learn a topic. In my experience this doesn't happen often, but a blog can potentially increase your probability.
Short answer: No, I wouldn't be convinced of much by blog posts.
Long answer:
If the folks doing the hiring care about soft skills (and they should), then their interview process should be designed to get those signals regardless of whether you have any public writing available.
Most software interview processes at most companies should not depend on your doing "work" outside of work — in particular, I would not support any interview process that required you to have a presence on GitHub. There are some good exceptions, e.g. open source focused roles. Likewise, for a developer relations role, perhaps it is reasonable to expect candidates to have some public writing.
A few ways a software interview process can probe soft skills:
* how does the candidate respond to feedback in a coding interview?
* have a written (code, text, or both) take-home portion of the interview; can also add a feedback cycle or two
* questions about how the candidate has handled interpersonal scenarios, e.g. gathering requirements for an ambiguous project or dealing with a difficult teammate
* etc.
All interview loops I've had a hand in designing have had some of this. Typically at all stages from intro call through final round. And the feedback from candidates has been overwhelmingly positive when these things are included.
Recommendation: don't go to Walgreens for your COVID booster. I did this last week, showed up right on time for my appointment, and they made me wait an hour and a half on an indoor non-socially-distanced line full of coughing people asking about COVID tests (miraculously, I don't seem to have gotten COVID). I'm told that non-Walgreens pharmacies are less bad than this, though I can't personally confirm.
I went to a Walgreens for my booster and didn't have to wait in a long line or be exposed to lots of sick people. But they *also* refused to update my vaccine card with proof of the booster, so I'll have to order a new one from the CDC or whoever does this, assuming I'm even in the database at all now. Foolish inconveniences!
The CDC refers people back to where they got vaccinated (and/or the state health department) for the similar issue of replacement vaccine cards.
If the shot was properly recorded, you may be able to get a digital vaccine record you can store on your smartphone as an alternative from your state (depending on the state) or your primary care health provider (if you have one and it does digital vax cards).
(My state didn't have any record of my second shot, and when I was finally able to get through to the health department half a year later, it referred me back to the place I was vaccinated. To its credit, the provider was actually very quick to correct it once I contacted them through their company website. But if Walgreens is having issues, that may not work as well for you.)
Huh. I have a COVID booster appointment at Walgreens tomorrow, since they were the only place I could find with open slots. They also had a great signup system, where I could search appointments at all area pharmacies at once. I'll let you know how it goes!
Update: Pretty much everything was great. There was only about ten minutes' wait. They wrote the booster dose on the back of my old vaccine card, but there really wasn't room on the front (blame the people from last spring.) Two people did come up asking about at-home COVID tests, but neither of them was coughing, and they left within a minute.
And, I don't have any side effects aside from a sore arm and feeling slightly more tired than normal.
I went to Walgreens for my COVID booster because they were the only pharmacy in town with Moderna boosters. When the pharmacist heard I had already gotten two shots of J&J (I was in the two-shot phase 1 trial before they even planned the one-shot regime), he refused to give me a booster, claiming I wasn't eligible, even though it had been over a year since my second shot. (He told me I could get a doctor's prescription for one, but that seemed like too much of a hassle). Two weeks later, I end up catching COVID. I called the pharmacist to rant at him and he had zero regrets about following procedure. Even more reason to avoid Walgreens.
We had an awesome experience at Walmart 2 weeks ago. Random walk in with no appointment flu shots, could have done the same for Moderna booster. They were taking Phizer appointments one week out, which we declined. 30m after we left they called and asked if we could return to fill two cancellations.
For what it's worth, I'm pretty sure the shot I got from Walgreens was a dud, either due to cold chain issues or because the pharmacy tech didn't stab it in far enough. No reaction. I am now an immunocompromised person in a risky work situation. Next dose is coming from a hospital if I can manage it.
FWIW I would not assume your booter was ineffective just because it did not provoke a reaction: I had no reaction at all to a booster of Pfizer after getting AZ originally (having had a very acute fever after the first dose of AZ but almost no reaction to the second).
I had zero reaction to a first dose. It was slightly creepy. Then I had a mild but definitively existing reaction to second one. From some conversations I understand that many people had no reaction, so that is normal
For what it's worth, I had all three of my shots from my local Walgreens, and the procedure was smooth and uneventful on all three occasions. Same for my mother.
About 9000 Walgreens locations in the US. You could have found a clinker. My booster at another Walgreens went smoothly. I scheduled an earlier AM appointment and was in and out in 15 minutes.
I’ve run into this situation at the same location though:
They *made* you stand in line? As in, you were manacled to a post or something? If not, I wouldn't assign Walgreens more than 50% of the liability if I were on the jury.
I realize I’m being a bit cranky here, but — this comment seems to me to be an instance of a low-value kind of gotcha: 1) Break off a piece of somebody’s remark. Out of context that piece sounds dumb. (2) Point out that it sounds dumb.
There’s lots and lots of that kind of thing online, and more on ACX than there used to be. Jeez.
You, I and everybody else know that Scott was not claiming that he was forced to stay in the line in the way that somebody manacled to a post would have been forced. He meant that in order to get his scheduled Walgreens shot without waiting in line he would have had to protest the arrangement in some way that was conspicuous and unpleasant and not very likely to result in his getting his shot.
If you think he should have chosen the alternative of making some sort of public protest, go ahead and make the case for that. But the gotcha you posted is just silly.
I disagree. I'm pointing out that Scott's actions, read as revealed preference, demonstrate that *his* concern for the possibility of catching COVID from the line was low. If he thought it was a serious risk, worth at least the very mild trouble of coming back another day, or going to another place, he would not have stayed.
Therefore in his own estimation, the risk was low. Given that, why should Walgreens assign a higher importance to the risk than Scott himself? Which suggests the issue is entirely cosmetic, like a ritual "thank you" to the cashier or something. Cosmetic niceties are...well, nice, but they are not in the same category as actual carelessness with public health, which is what the implication of the comment was, I thought.
In short, I had a serious point to make. I regret it was not more obvious.
Did you go to the Berkeley Walgreens that looks like it’s about to shut down: broken front gate, many empty shelves, multilingual “no shoplifting” signs everywhere? My girlfriend did; her experience was very similar, but with fewer people asking about tests, as it was pre-omicron.
It's been a very tough few years. As the pandemic continues to evolve, we're doing everything we can to keep everyone protected.
To date, our teams have provided over 55 million COVID-19 vaccines, more than 23 million COVID-19 tests and have filled over a billion prescriptions filled annually. We're incredibly proud of our team members and their unwavering dedication to your health and safety.
Now more than ever, we owe it to you to recognize the challenges we as a company have faced during the pandemic, including longer lines, out-of-stock items and delayed appointments.
We heard your feedback and we're working tirelessly to be better. We're implementing speedier appointment scheduling and hiring thousands of new team members to help make visits faster. You can also save time and trips by using our digital tools online and on our app.
We care deeply about every one of you. Your well-being is our top priority, and we're committed to delivering the exceptional wellness experience you deserve.
We're in this together, and because of your feedback and understanding, we're confident we'll all come out stronger.
We wish you a very happy and healthy 2022.
John Standley, Walgreens President Tracey D. Brown, President of Retail Products and Chief Customer Officer
Oh, and I like how "longer lines, out-of-stock items and delayed appointments" were problems the company faced rather than problems its customers faced.
Like many, I've been "George-Pilled" by the recent series of essays. I'd like to get the other aide of the story. Can anyone link to some good critiques of Georgism?
It seems like the most intuitive critique is that homeowners are a pretty strong political block to most places (read: the US), and would object to their rent pricing model changing significantly, (perhaps wisely) and would not believe that a UBI dividend would be how government would actually choose to spend that tax revenue. I'm not sure I disagree with them, even despite that I found all the Georgist arguments presented quite convincing.
More generally, I suspect it's the strongest criticism, and why so few attractive policies are implemented.
If a policy is attractive theoretically but difficult to implement, either because it is unpopular with stakeholders (homeowners, in this case), or else a radical departure from norms, even if those norms are bad and the stakeholders suck, the policy seems unlikely.
It may or may not meet the "good" threshold, but it is a critique. Personally I found their specific arguments unconvincing, but I can definitely conceive of others in this (very) general category that fare better.
For instance, the person who buys a mountaintop in the middle of nowhere to open a ski resort gets hit progressively harder and harder as the resort town builds up around it. Compounding the problem (but in another category entirely) is that this mountaintop property is qualitatively different from the surrounding town; suitable for different uses from the rest of the area, making neighborhood comparison-based valuation impossible.
The comment sections of those articles seemed to be filled with a lot of criticism, but it was a fair amount of work to distinguish thoughtful economic critique from people who have a lot of equity in their homes.
It's my impression that most economists think it'd be a great idea but almost completely impossible to implement because of political considerations. The central problem is that like 60-80% of this country believes, very strongly, that owning their own home is this incredible opportunity for building wealth. Even people that don't yet own a home tend to be saving up for one. Since, for most of these people, these investments would represent the majority of their net worth, they are extremely averse to suggestions of a different arrangement. Many of these people created a retirement plan decades ago predicated on paying off their mortgage.
In general, our addiction to land-as-an-investment is very strong. I think breaking that addiction is just incredibly hard, politically. Those are the problems that Georgists have to contend with.
"The central problem is that like 60-80% of this country believes, very strongly, that owning their own home is this incredible opportunity for building wealth. "
The more charitable conclusion is that your method of calculating the value of a home fails to capture important aspects that actual people value.
In order for you to state with disapproval "The central problem is that like 60-80% of this country believes, very strongly, that owning their own home is this incredible opportunity for building wealth." you implicitly claim that the value of a home is low enough that you can disapprove.
You may fail to state a method, but you implicitly have one.
I wouldn't say I'm passing any judgement on any home-valuing methods. I think you are misunderstanding my comment in a defensive manner.
If I have disapproval, it's for the suboptimal equilibrium we find ourselves in with our system of taxation. I can disapprove of the system while respecting that every actor within the system is behaving rationally. I didn't say that people are wrong to consider home ownership to be a good investment. Under the current paradigm, it is a good investment.
But I do think that in the long run this is bad for society, and it harms us all economically, disproportionately the future generations. One way to look at it is that housing cannot simultaneously be a good investment and be affordable. Think about it: anything that's a good investment will be bid up to the maximum price the market can bear.
It's not like a Georgist LVT would make wealth building or home ownership any harder. The stock and bond markets would be fine (they would actually be stronger) and homes would cost the values of the houses built upon it. You could still make improvements to a home to increase its resale value. Teardowns would become much less economical and frequent.
You can think this is a bad idea, but saying that "people value homes more than you think they do" does not demonstrate a serious critique.
"I didn't say that people are wrong to consider home ownership to be a good investment."
You implied it. Stating "the problem is that people believe X" implies that you think X is false.
Furthermore, you're confusing "is a good investment" and "is a good value". People don't believe that homes are a good investment, in the sense of providing more money when they are sold than if they had bought something else. People rather believe that homes provide good value for the money put into them, which is different.
I wrote a lot of critiques in those threads, from having been very supportive of the idea when initially posted in the book review contest.
I think Georgism works pretty well in dense cities where most people rent. The more the population leans towards owning their own home, and the more rural the scenario, the less Georgism makes sense. Because so much of the [U.S.] country has made decisions based on home ownership, the transition will be extremely difficult - a fact most Georgists acknowledge. That's not people being needlessly selfish or just holding onto outdated ideas, it's a real problem where they've sunk a lot of their financial wealth into property ownership. A swift move to Georgism would destroy the wealth of most people in the country. A UBI would help, but only by making everyone dependent on government. Taking a lot of people who feel independently comfortable in their livelihood, stripping them of most of their accumulated wealth, and then making them dependent on a government stipend seems like a loser of a deal.
Trying it out in some cities that have very high rates of renting would be a good place to look. If that's not feasible, then there's no way it could be feasible across the country.
Georgism privileges skilled labor and productive enterprise over getting there first and paying your dues.
We really don’t like the guy who gets paid a lot of money because he can write code or trade derivatives. We think it’s wrong that he gets paid so much, and proper that he gets cut down to size by heavy taxation.
Meanwhile we really do like and respect and want to support the family that put down roots and bought a house in a community that turned out to be successful.
Georgism would transfer tax burden from the former to the latter, and for that reason would bring public policy further out of alignment with the public’s values.
I don't have a link, but I remember reading an interesting counter argument in the comments. Basically it was that if I build and own an entire city on my own, I won't get taxed much for the land, because all the extra value is an internality, but if a million people build and own that city, they'll all get taxed on the externalities they get from each other, which seems unfair and like it would distort markets toward concentration of land ownership.
10 000 landowners are surely enough for a city to have almost the same landprices as 100 000.
It seems like only very active planning can ensure the concentration of ownership that is necessary for the land in a city to become significantly lower taxed.
I don't think that much planning is needed. It can start with just a gated community or a shopping complex or whatever, and then the incentives should encourage any land owners between whom there are large positive externalities to merge their holdings in order not to get taxed for those externalities, until the whole city has the same owner.
Could anyone with life sciences/nootropics background comment on nicotinamide adenine dinucleotide (NAD+) as a supplement? This video has some impressive claims with papers backed up by reputable journals: https://www.youtube.com/watch?v=cY25i_bkUys
From my cursory googling this seems to be both available online and safe, yet I haven't seen much discussion of it in the few places that I trust on this kind of material.
Don't most people take this general category of thing as nicotinamide riboside or mononucleotide for availability reasons? Or am I thinking of something else?
My half-assed intuitive take after reading way too much literature is - fasting should give you most/all benefits of NAD+/NMN supplementation, plus many others, so I haven't considered it further as it seems redundant.
For some reason Google books/play has digital copies of hard to find books at less than half of the price of what you can find elsewhere. I don't know if it's legal or not but I was amazed to get a digital copy of a fairly obscure textbook called "The Rediscovery of Classical Economics" for a fraction of the price I could find it elsewhere. It kind of looks scanned in but it's just fine. It's not a downloadable PDF but you can highlight and save your notes to Google Drive. If you do lots of research this should be a source you check.
I don't think this particular diagram is particularly enlightening, but will attempt to give my own classification system/understanding of the landscape.
A useful question about a particular advocate of libertarianism is if they are using a deontological/rights based foundation (eg non aggression principle) or a consequentialist foundation. Note that two individuals could arrive at the same policy bundle through different foundations, so sub-ideologies cannot cleanly be classified into these categories.
I bring this up as the left right groupings are different between these views.
In the rights based view, i see the following 2 axis, positive rights yes/no (the right to something (eg healthcare or food)) and property rights yes/no.
No to property rights places one in the libertarian socialism grouping.
Yes to property no to positive places one in the right libertarian/anarcho capitalist grouping
Yes to property and yes to positive places one in the social libertarian grouping
Alternatively if you are in a consequentialist foundation world, you tend to get more uncertain positions, eg i am 80% sure full anarcho capitalism would work (no state), but if i am wrong i would be happy with a minarchist state (smallest possible state). Which means that individuals tend to be less clearly members of a particular ideology.
But the beliefs on the effectiveness of markets (and thus property rights) is sufficiently bimodel, that you still get a clear socialist libertarian cluster and a clear capitalist libertarian cluster.
I agree, this is an unuseful Venn diagram designed to shoehorn libertarianism into the left-right dichotomy, while deontological/consequentialism is the primary axis.
I... wouldn't say neoliberalism is a subset of libertarianism. They must be using either a *much* narrower definition of the former or a much broader definition of the latter than I would.
(or they're just confused about how Venn diagrams are supposed to work)
I would definitely not put "neoliberalism" at the position they have it. And "Democratic Socialism", even with the qualifier "(in some forms)" seems like an odd thing to include at all!
Is there an effective altruist case to be made for investing in great works of art? Or would a utilitarian framework regard the notion of great art as too mushy? I think art that is goal oriented or dogmatic risks being bad, but surely there is a middle ground between art for its own sake and agitprop?
If the case for beauty is not enough, I don't know what case can be made. Art as Investment Vehicle brings me out in hives, and that's plenty utilitarian.
We can then go on to argue what is and is not Great Art, but there is a strong case to be made for beauty, and some will see beauty in works by modern artists. People can find pleasure/enjoyment/satisfaction/aesthetic stimulation in a work like this one,
Utilitarianism is a means to an end, and the end is maximal happiness and well-being. A beautiful work of art directly contributes to that end. Arguments about utility optimization can only indirectly increase utility, if they convince someone to act in a way that results in more happiness and well-being, and these have to be *really* good arguments if they need to stack up to Plato. (I can't personally comment on the quality of Plato but his fame says enough.)
So, to answer your original question, yes, there is a good case. If you find someone who is a modern Vincent van Gogh, you could justify investing in them over e.g. fighting poverty if otherwise, all future generations are deprived of enjoying their work. The hard part is knowing a modern Vincent van Gogh when you see them.
My thinking is like Laurence's. "Art/beauty" is something like "that which, when people experience it, they think it's valuable, even if nobody understands why it's valuable" Sort of like an experience of value that's direct or unmediated, or whatever mediation there is, is opaque.
So, an argument for beauty from utilitarianism is a tautology. Utilitarianism is probably smuggling beauty in through definitions, no need for an argument.
Have you read Kant's Critique of Judgment? I believe that's where I got this idea from.
I think that I possibly don't "open myself up to" enjoying art when I don't know what it's intending to "say"/convey! But when I look at "Black on Maroon," (and especially the "Rust and Blue" one... I see; they are by the same guy. Hence the resemblance--I am an ignorant oaf about this.) I DO feel myself relaxing a bit. (The "Rust and Blue" colors are delicious.)
Asked another way, could Beethoven, Rothko, David Foster Wallace, or insert your favorite artist here justifiably win an EA grant or an ACX grant? What about philosophers whose work is not rationalistic, but whose work might contribute to beatifying the world through style or just through opening horizons of new thought?
There is definitely a good argument for the utilitarian value of art, but buying high art as an investment isn't a very good way to incentivise the production of more art. You'd be better off donating to art education charities, schools, etc. Or commissioning art from people online.
As an investment vehicle in itself its almost certainly inferior to index funds
Replace the word "investing" with "buying" and you have a better case, assuming you mean the former to be an instrument that will appreciate in financial value. The art market is the ultimate vehicle for insider trading and only very few wealthy insiders working with a handful of high end galleries will ever see an appreciation in value. The other 99.99% of art buyers will see their purchase price drop to near zero after walking out the front door.
On the other hand, if you want to purchase a quality work from a starving artist, I think of art as a liberalizing (in the positive sense) force and not only will you support the art industry but hopefully you and your guests will have a work beauty on the wall to both admire and contemplate.
I would say it's better to invest in artists than to buy art. Not invest in them for financial gain but for the sake of the art itself, if it is worth being produced. From an EA perspective, if you're going to spend any money on art, you should fund the starving artist who has the greatest potential to produce something that will go down in history as a great work, and appreciated by countless people for centuries to come.
Yes, this is impossible to predict, but shouldn't we try anyway?
Here's what Peter Singer had to say about art museums in The Most Good You Can Do:
"Art surely can be an aid to learning, but building new museums is not likely to be the most cost-effective way to do that. We have other opportunities for studying art from which we will learn as much or more than we can by joining the crowds peering at very expensive paintings from behind a rope or through bulletproof glass. If the goal were really to educate the public about art, museums would do better to spend a few thousand dollars on the highest-quality reproductions and allow the public to get as close to them as they like.
To forestall misunderstandings: there is value in creating and enjoying art. To many people, drawing, painting, sculpting, singing, and playing a musical instrument are vital forms of self-expression, and their lives would be poorer without them. People produce art in all cultures and in all kinds of situations, even when they cannot satisfy their basic physical needs. Other people enjoy seeing art. In a world in which everyone had enough to eat, basic health care, adequate sanitation, and a place at school for each of their children, there would be no problem about donating to museums and other institutions that offer an opportunity to see original works of art to all who wish to see them, and (more important, in my view) the opportunity to create art to those who lack opportunities to express themselves in this way. Sadly, we don’t live in that world, at least not yet."
"In a world in which everyone had enough to eat, basic health care, adequate sanitation, and a place at school for each of their children, there would be no problem about donating to museums"
This very much does not follow. What he is implying is that since food is more important than art, we shouldn't fund art. But by his own standards, just because someone would die without food and wouldn't die without art doesn't make food more important than art; you need to do a utility calculation and in the calculation, lives don't have infinite value.
A utilitarian cannot a priori restrict other people's utility functions. There's really no coherent notion of a well-formed utility function being "too mushy." John Stuart Mill tried to give arguments about what utility "ought" to be, but it's pretty clear that those arguments were parasitic on non-utilitarian ethical ideas.
Whether the above tells for or against utilitarianism is a separate question.
Undying glory (found only in great works of art) more accurately tracks human well-being (esp over the long term) than economic modeling, because economic modeling incompletely measures the human condition.
Consider the value of a stay-at-home parent. This is not very easy to track economically, and harder to recompense, but (esp given the right parent) it can pay off on the scale of billions. Easy to mismeasure economically, but practically every Great Book enshrines "honor thy father and mother...that thy days may be prolonged, and that it may go well with thee. (etc)"
A society that understands this (and things like it) well cet par be better off than one that doesn't.
Re: objective framework for art. I think there is an inside view and an outside view of what makes art objectively valuable.
I think the inside view can be summed up in the maxim "nothing too much." Not a stroke out of place. Every arbitrary choice tends to the same end to form unity in variety, a perception of parsimony. Because the choices an artist can make depend on the convention of the form, whether a piece of art conforms to the "nothing too much" standard is something generally only the learned in the field can judge. Sometimes this makes laymen mad.
I think the outside view of beauty can be summed up by the phrase "undying glory." What's the news that stays news? What are the books that stay in print? What are the lindyest of the lindy?
The argument here is that (a) a work's popularity could just be accidental -- a fad -- and so inherently worth nothing, but that (b) the longer a work stays popular (especially through many societal "selections"), the more likely it is that it's NOT a fad -- i.e. that there's something inherently valuable that's causing the fad phenomenon. You generally also find that experts agree on what these works are.
For example: what made Aristotle the philosopher of choice for the Greeks, Romans, Golden Age Muslims, Scholastic Catholics, German Idealists (depending on who you read), Existentialists, and now modern Virtue Ethics? These societies are pretty different, and I'm only willing to buy the "historical contingency" argument for so long before I think that there's something to explain the 2500-year staying power (esp. when we think that everyone knows the name "Aristotle" and nobody knows the name "Longinus").
"For example: what made Aristotle the philosopher of choice"
I'd go with historical contingency.
First of all, everyone in your list is related. The Romans copied lots of things from the Greeks, for instance, and by the time we got to virtue ethics, Aristotle was very well known in the culture.
Second, there's a difference between Aristotle's ideas being inherently valuable, and Aristotle being inherently valuable; it may be that his ideas are useful, but not hard to come up with, and if he hadn't existed, someone else who came up with them would have lucked into popularity.
Third, ideas can spread for memetic reasons; the Church adopted Aristotle's ideas, and the Church is a powerful force for spreading ideas regardless of their merit.
1. How is the Islamic Golden Age related to the Romans except through Aristotle? Likewise, why do the Catholics (who were both anti-islam and anti-pagan) pick him up in the first place?
2. This is somewhat of an aside, but Aristotle easily fits both categories. The ideas are important and they're hard to come up with, both in content and in form (i.e. Aristotle's unique method of inquiry/thought). cf. e.g. the invention of Symbolic Logic, which wasn't improved until either Leibniz or Frege (and depending on who you ask not even then).
3. (a) There are no successful memes without merit. Even if the merit is just "fitting well" in a particular environment. However, like I was saying, with Aristotle, the "contingent-upon-environment" argument becomes less convincing as the years pass. (b) the Church doesn't just adopt ideas for no reason either.
"How is the Islamic Golden Age related to the Romans except through Aristotle?"
Islam ruled vast swaths of land, much of which had been previously occupied by Romans, and adopted some ideas from the conquered areas.
"Likewise, why do the Catholics (who were both anti-islam and anti-pagan) pick him up in the first place?"
Catholics weren't "anti-pagan" in that sense. Notice, for instance, how reference to Greek and Roman gods stayed a fixture of literature in Catholic Europe.
"There are no successful memes without merit. Even if the merit is just "fitting well" in a particular environment."
The argument about Aristotle implies nontrivial merit, not just any old merit.
1. Why these ideas? Why does this idea get selected over and over again and not others? (e.g. Why was Aristotle adopted but not Greek togas?)
2. Catholics pre-Aquinas were pretty opposed to Greek Rationalism in general and Aristotle in particular. It's historically really surprising that they integrated him. Many Aristotelian texts didn't even survive except through the Muslims.
3. Do you think adoption by the Catholic Church is trivial? The church has strict selection criterion esp for stuff they integrate so deeply into doctrine. Even if Aristotle happens to "fit well" with the environment of the Catholic Church, the Catholic Church is also a nontrivially meritorious meme in its own right.
If Greek togas were adopted and not Aristotle, you could say the same thing. Given that they liked the ancient Greeks, some subset of things would survive; Aristotle happened to be one of them.
"Do you think adoption by the Catholic Church is trivial?"
It's trivial in the sense that it has little bearing on the question you're bringing it up for.
FWIW, Aristotle is remarkable not only for his longevity and range, but also for the extent to which his ideas penetrate and then permeate the societies in which he's adopted. E.g. Aquinas doesn't even use his name, instead just calling him "The Philosopher."
Still within the Octave/Twelve Days of Christmas if we're going the traditional route, which only ends on January 6th (Twelfth Night) with the Feast of the Epiphany!
People often say processed carbs are bad. I've never quite understood this. Why would it matter if something is 'processed', and what that does even mean?
Here is my best guess of what is going on, but this could be totally wrong (this comment is very much meant as a question). The guess is that 'processed' is a misnomer and it's really about something like 'modified'. It's not actually the case that people in the ancestral environment never had too much to eat, so are bodies are capable of regulating food intake to some extent, but they're used to food with certain compositions. Therefore, if you take grains and make them into flour and make flour into bread, this is still okay even though it's quite 'processed'. But if you only take part of the grains to make the flour, well then you modify the natural composition, and as a consequence, human bodies are worse at regulating it and so you may put on weight.
This would also suggest that modified carbs are really only a problem if you're overweight and otherwise not particularly unhealthy?
My understanding is that it's more about glycemic index ("fast-acting" vs "slow-acting" carbs) and the fact that most types of processing generally increase the glycemic index; and also about the mixture of carbs vs fiber, where most types of processing keep the carbs but remove the fiber.
FWIW, fiber *is* carbs, just not (very?) digestible. That's why it's normally included under the carbs category. But when the category is broken down it's usually into fiber and sugar, but those don't add up to the grams of the whole carbs category, because they don't usually list starches, etc.
And *that* isn't the whole story. Even things with identical amounts of fiber, starch, and sugar can be different depending on their accessibility. Cooked carrots are high-glycemic and raw carrots very low glycemic. So much so that a lot of the sugar is likely to make it's way through the digestive system without being absorbed (depending partially on how well you chew the carrorts).
So processing is important even in cases where it doesn't change the relative percentages of nutrients. (Though, yes, it usually does.)
Carbs aren't bad, that's just one of many dietary misconceptions. A more accurate approximation might be "carbs without fiber are bad", because they tend to be easy to overeat and have higher glycemic index. Fruits, vegetables, and whole grains are all healthy sources of carbs, and are perfectly appropriately even when on a fat-loss diet.
"Processed food is bad" is not an unreasonable generalization, because processing tends to remove fiber (making it easier to overeat) and some of the more nutritious parts of the food. Going from wheat to white flour is an example where a lot of nutrition is lost; in fact, most flours you buy will be enriched to add back some of the lost vitamins.
All that said, "bad" foods in moderation are usually fine.
> "Processed food is bad" is not an unreasonable generalization, because processing tends to remove fiber (making it easier to overeat) and some of the more nutritious parts of the food.
Ok, but doesn't this mean "it's really about modified carbs, not processed carbs" is accurate?
I'd push back on the sweeping "all fruits are good" generalization here. Many if not most of the fruits you find in your grocery store - including (gasp!) the organic ones - are bred to contain insane amounts of sugars, and either as a result of selective breeding or something else, but they don't contain that much fiber either. Plus, that sugar is mostly fructose which some have argued is actually worse than glucose (though this last part is by no means a consensus).
Charitable interpretation - Processed means concentrated beyond what you could usually find in nature outside of a beehive. More calories per volume or weight. Less fiber and micronutrients. More calories to reach the same level of satiety. More commnonly mixed with fats (fats + carbs is the least satiating combo of any 2 macronutrients).
Unchartiable interpretation - Its a purity fixation.
Arguments from genetic evoltuion in nutrition are usually horseshit. We evolved to eat anything we could get our hands on.
Processed carbs are *typically* bad for WEIRD folks because we *typically* don't need to rapid injection of easily-available energy when we do our *typical* activities like sit at a desk for 10 hours followed by recreationally laying on the couch for 6 hours. But my dad was on a road crew and could put away 4k calories a day and getting processed carbs in was no problem for him staying lean. Lumberjack camps served like 7k calories a day back when trees were felled with muscle power, and they ate a ton of simple carbs.
Its "best" imho to think of food as fuel, and matching your fuel to your needs.
I can only maintain a healthy weight on a "low carb" diet, and here's my understanding of why for n=1:
1) Americans eat just an absurd amount of carbs. So when I say "low carb" that actually just means "my entire diet isn't made up of flower and sugar with hardly a protein to be seen."
2) I do better with "non-processed" carbs. You're absolutely right that this is a nonsense phrase filled with nonsense. But the best way to describe what I'm looking for is that usually a piece of plant nutrition is made up of a shell with a fair amount of fiber/other good-for-you things and a gamete which is where all the calories are. Sugar, white flour, and corn syrup are "processed" to remove the nutrient rich shell. As a result you don't get full when you eat it and you tend to overeat.
The term is bad though because whether a carb is "processed" or not isn't the best way to judge this. A baked potato isn't "processed" by any reasonable use of the word, but it's still a calorie bomb with very little satiation or nutrition. Instead, best to look for foods where the outer layer is doing reasonable work at filling you up - carbs with big fiber content, potassium, omega-3 rich oils and small gametes such as grain, dark greens, non-white rice.
Probably also worth mentioning that if you are *active* and already at a healthy weight those low-satiation carbs are not quite as big a deal. Your body just burns them in your triathlons or whatever.
Weren't there some people claiming to lose wieght by eating nothing but potatoes for a year? I don't know if it's a good idea, but if they can live on just potatoes surely there is sufficient nutrition there?
Spudman! Lived on Potatoes for a year, lost a ton of weight, and did not die. To be fair, potatoes are definitely healthier than, say, white bread. Spudman ate a variety of potatoes, including sweet potatoes and smaller potatoes with thicker skins. He also supplemented with vitamins since he was not getting enough protein.
He does not recommend eating only potatoes and only did it to make the point that potatoes are not super bad for you and are a reasonable part of a balanced diet. I agree. But the fact that it worked doesn't really prove anything other than that he was in a calorie deficit. People have done the same thing with pizza.
I probably overstated my case here - I, like most Americans, am a simple carbs fiend, and so cutting simple carbs does not really put me at a carbs deficit. Carbs are still the foundation of a balanced diet, you just don't want to eat so much simple carb that you put on a ton of excess weight. The biggest issue with American diet is the prevalence of soda, for instance.
The tautological way to put it is that if you're not eating too many simple carbs you don't need to cut simple carbs. But Americans really really overestimate how many simple carbs are healthy.
I think there are some quite serious indication that processed food is bad for you. Most importantly it increases your weight, but perhaps also inflammation beyond the weight, and it has some other bad consequences. This is pretty well established. For example, lab animals gain weight from so-called "cafeteria diets", but not from diets which have the same ratio of carbs, fat, protein, sugar, and so on.
But we have plainly no idea why. There are dozens of popular theories: glycemic index, types of oils, fibers, sugar content,... And as far as I can tell, none of these theories withstands even a modest level of scrutiny. We can most certainly rule out explanations along the lines "it contains too much sugar" or "it contains the wrong types of fat" or "the fats/carbo/protein ratio is bad".
So our best conclusion is "probably processed food is bad, and all major candidate explanations are wrong".
For a book-long discussion (focusing on the weight part), see here:
Their favorite explanation is that lithium is the culprit, and that it enters food because machines are often greased with lithium-containing substances.
it seems like the answer for questions on nutrition is always "we have no idea". I've read about half of the article you linked, and it pattern matches to solid work, so I cautiously trust your take.
But even so, shouldn't we at least be able to define what 'processed' means? Is it the amount of steps taken to do stuff with the grains? The amount of chemistry involved? The amount of machines? The current concept is so vague that I can't tell whether white bread more processed than whole grain bread.
In one of the later pages in the Slime Mold Time Mold series, the writer conjectures that "processed food" is more likely to have a higher concentration of contaminants.
I share your frustration, but I really think we don't know. This is something that I have heard from many sides, not just from the linked article.
The standard recommendation is "have as few steps as possible done by industrial machines and as many as possible at home in your kitchen". This is probably a pretty bad proxy for the unknown thing that actually matters. But we don't have a much better proxy than that.
"Processing" is a vague term that basically means doing anything to food (e.g. milling, chopping, cooking, adding ingredients, removing elements, etc.). It's too broad of a term to always be bad. Some processes, such as fortifying common foods with vitamins, are done to improve health.
There are some reasons why "heavily processed food is bad" might be a good rule of thumb.
First, processing often increases the convenience, which may cause us to overeat. Compare shelled pistachios to unshelled pistachios, or oranges to orange juice. You're much more likely to eat a larger quantity of the processed version.
Second, processing improves taste, which may cause us to overeat. There's a reason we cook meals instead of eating a bunch of raw ingredients. If you're like me, those freshly baked chocolate chip cookies are more tempting than the bowl of apples and bananas.
Third, salt, sugar, and other easily digestible nutrients tend to be tastier, and processed foods may have too much of these.
Fourth, it's too easy to buy large quantities of processed food. Home baked cookies might not be any healthier than store bought ones, but just the effort of having to bake cookies from scratch each time will probably cause you to make them less often and eat fewer cookies overall than if you can just throw boxes of them in your shopping cart.
I think a lot of the harm of processing is simply from making food too convenient, tasty, and easily digestible.
I'd like to add a Fifth to this list which is processing breaks down the cell structure of the food which changes how the food is processed by the GI track in a way which is presumably detrimental to nutrient bioavailability as well as the glycemic response.
Do you have a link to the animal studies you mention? My current model is that "processed food bad" sentiment predominantly comes from observational studies indicating that people who eat a lot of processed food tend to have poor health, but it's a correlation that has little to do with processing itself making the food bad somehow (though of course it's hard to deny that a lot of processed food is just obviously crap). Decent animal studies would indicate against this.
As far as I understand, it's actually not so easy to find a diet that makes lab animals obese, and this was a serious challenge before researchers found out that you can simply feed them human food. High-fat diets work less well for making animals obese (there is a direct comparison study linked above), and high-sugar diets seem to be used even less. I think they just don't work at all, though I am not completely sure about that point.
Citation that I stole from the link above:
""Palatable human food is the most effective way to cause a normal rat to spontaneously overeat and become obese,” says neuroscientist Stephan Guyenet in The Hungry Brain, “and its fattening effect cannot be attributed solely to its fat or sugar content.”"
If you want to dig into it, then this book might the place to start.
Ah I see, so in that study rats were fed ad libitum, that changes everything. Rats on human food gained more weight, but simply because they ate more calories, you can see it from the charts. And of course they did, the human food they used in the study (cookies, cereal, processed meat etc) is hyper palatable to humans and rats alike, but unlike (some) humans rats don't have the common sense to not eat an entire meal of snacks. And they had comparison groups on standard chow, high-fat diet and low-fat diet (with fat calories replaced with sugar so it was effectively high-sugar diet). And all groups were fed ad libitum. Rats on all 3 diets gained a lot of weight compared to the controls on standard chow, with high-fat and high-sugar faring about the same. On the human food they gained the most, but again, they ate more calories, the diets weren't matched by calorie or nutrients content at all. So nothing surprising here except maybe for high-fat and high-sugar giving the same result, I'd expect high-sugar to induce more weight gain and insulin insensitivity.
On the broader point, I have quibbles with all their "mysteries" except for 5. Most importantly, they basically ignore the effect of exercise - they handwave higher exercise in 1, and then completely ignore it in 4 where it could easily explain all the difference. In 8, they directly contradict their sources. The meta-analysis they link says that people on low-carb diet lost 8.73 kg in 6 month and 7.25 kg in 1 year, similar but slightly less on low-fat diet. For the population with median weight of 91kg that's quite a fucking lot. Now the article somehow concludes from this that "diets don't work" and "even if someone does lose 20 lbs, in general they will gain most of it back within a year". No, they're not gaining most of it back, they are gaining ~3lb of it back! And even for this, a part of the reason is the study design - the median duration of a diet was 24 weeks, so by the end of the year many of those people haven't been on a diet for a few months!
Generally, it all sounds like a lot of motivated reasoning twisted to prove that *clicks through to the last chapter* oh right, eat whatever you want, don't exercise, industry and the US government are to blame. Ok this was a bit uncharitable, but seriously, lying about your sources is bad so they deserve it.
They address your points in later parts of the article series (e.g., exercising and 4). Also, I have recalled that Scott has written a book review of The Hungry Brain,
For the question whether "diet's don't work", I think that's a question of what you expect from them. By Amptoons' standards, that they should bring fat people into normal weight regimes, it is pretty clear that diets don't do that, https://amptoons.com/blog/?p=22049. And the 7-10kg of weight loss is not for a population of 91kg, most of these studies were restricted to obese participants. Correct the studies for weight and dropout, and I can see the point of not finding the results impressive.
I do agree that the statement "even if someone does lose 20 lbs, in general they will gain most of it back within a year" is not supported by their link, and they should take the blame for this. But I am not sure that this makes their general point invalid. My first google hit was this, which easily supports the magnitude of such weight regain, though in a time frame of 2+ years instead of one year.
The study says right in it's abstract: "Conclusions and Relevance Significant weight loss was observed with any low-carbohydrate or low-fat diet <...> This supports the practice of recommending any diet that a patient will adhere to in order to lose weight." And they somehow infer from this "Diets Don't Work". At this point I just wash my hands, I'm not wasting time on first finding out what they have to say about hunter gatherers and exercise (it should be right there when the topic first mentioned, otherwise all their comparisons are irrelevant), only to later find out their "evidence" for this is some study which says something diametrically opposite.
>And the 7-10kg of weight loss is not for a population of 91kg
Yes it is, read the paper, specifically the "Results" section. It is the median weight of the population they investigate. Yes median is a shitty metrics in this case, but that's all they report.
And the paper you linked says in almost every paragraph that a regimen of diet and exercise *works fine*, the problem is that it's difficult for people to *maintain* this regimen, for various reasons. Main of which is the same as with the poor rats: hyper palatable food easily available in unlimited quantities. So absolutely no mystery that would require pollutants for explanation - people it less and loose weight, then they stop eating less and regain weight. The entire "Clinical recommendations.." section is focused on ways to *keep people eating less [crap]*.
Besides, I always wonder when people make arguments like this - don't you, like, personally know someone who has lost weight via diet and/or exercise? Because I certainly do know a few such people, all of them basically say they can control their weight insofar as they can control their diet (the latter being challenging for some, for sure).
>The Hungry Brain
Direct quote from the review: "Researchers have been keeping records of how much people eat for a long time, and increased food intake since 1980 perfectly explains increased obesity since 1980". I'm not sure what "few things" you were referring to, but at least this book/review doesn't seem to support the idea that calories and nutrient composition are irrelevant, which is the main point I'm arguing against here.
"Processed" is a proxy for particular, characteristic modifications. Carbohydrates range on a spectrum from simple, (roughly, sugars), to complex, which the body breaks down into sugar for use. Commercial processing characteristically breaks down carbohydrates to simpler form and, importantly, concentrates them. The production and heavy use of high-fructose corn syrup is one example.
While admittedly a proxy, I would say that "processed" carbs actually does have fairly high correlation with a significant problem in the modal American diet.
"Processed" in the context of food being bad for you mostly means the gross composition of the food has been altered. So for example when ordinary wheat or rice is "processed" after harvesting, the seed coat and germ are removed (removing almost all of the fiber and protein), and only the stach kernel is left. When orange or apple juice is made from apples, the juice, which is high in sugar, is removed from the pulp, which contains all the fiber.
There are also some chemical modifications. The most famous of these is the production of high-fructose corn syrup from corn starch, in which the starch is broken down into glucose monomers, and then about half of those are converted to fructose, an isomer of glucose, and the replacement of sucrose (which is a dimer of glucose and fructose) with HFCS, which is cheaper to produce (in the US at least). This alters the chemical composition of the sweetener slightly.
The argument is that that you evolved, 1 million years ago, to eat foods with a certain gross mixture of nutrients, and this is optimal for health. For example, in a primitive state, where only whole raw foods are readily available, it's very difficult to consume large amounts of sugar, because it simply isn't found in high enough concentration. For example, an apple contains about 20g of sugar, which is about 1.25 tablespoons. You can easily eat 5-10 tablespoons of sugar in one sitting, either by itself or in some baked food -- but it would be rather more work to wolf down 4-8 whole apples at once. Additionally, it's thought that a certain mixture of gross nutrients -- so much protein to so much carbohydrate to so much fat -- results in less imbalance to circulating metabolites, like blood sugar, and fewer demands on regulatory pathways like that controlled by insulin. And, finally, there is the more recent hypothesis that some significant effect on our health comes from the health of the "micriobiome" of bacteria dwelling in our guts, and if you don't "feed" those critters the right mix of nutrients, you get unwanted changes in the demographics -- certain desirable bacteria die off, and certain undesirable bacteria populations grow.
All of these are necessarily speculative, with most evidence for or against coming from epidemiological studies if not mere anecdote ("I ate an apple a day every day and my asthma was cured!") so...pretty far from empirical soundness. Of all of these, my favorite is the microbiome hypothesis, on account of it explains how an excess of *an actual nutrient* can turn out to have chemical toxicity -- it's not the nature of the nutrient itself, but rather that it "feeds" some highly undesirable microbes which produce the toxic consequences, sort of like a "red tide".
Edit: I will add that there is also an argument, based again on the microbiome, that we are used to having certain species of gut bacteria regularly replenished from our food, e.g. certain species of soil-dwelling bacteria that tend to cling to roots when you eat them. And that the processing of plant foods we do going from field to table is *designed* to kill off as many bacteria as possible, make the plants foods if they are whole squeaky clean, and if they are turned into ingrediens nearly sterile, and that this may deprive us of some unknown necessary "micronutrients" in the sense of certain valuable bacteria species we evolved to eat regularly.
>This would also suggest that modified carbs are really only a problem if you're overweight and otherwise not particularly unhealthy?
Not really, you definitely can have metabolic issues without [yet] being obese. Insulin insensitivity is the most common of them, and roughly speaking it's driven first and foremost by spikes of glucose in your blood that come after gorging on lot of easily digestible carbs without much to dilute them or fiber to slow down their uptake. Even if these metabolic issues never lead to obesity, they can lead to other problems: increased inflammation, broken lipid profile (leads to CVD), and maybe non-trivially increased chance of some cancers (but don't quote me on that). The standard fasting glucose test your doctor gives you will catch it only when the problem is well underway.
Being young and physically active definitely helps a ton, but the former only works for a while and of the latter they say "you can't outrun your bad diet".
Assuming it's true (any one with expertise care to comment): "science" has spent the past few decades not really understanding how viruses spread. It seems crazy to me. But somehow believable.
IIUC, it's true. There were a lot of people saying from early cases that the spread was probably due to aerosols. I decided that was true based on an early chinese restaurant case. But this wasn't the officially adopted explanation until a lot more proof was available.
AFAIK, there are still arguments about the relative importance of different modes of transmission. People around here are still cleaning surfaces. Perhaps that helps. I'm still washing my hands frequently and using alcohol wipes in between. I suspect this is excessive, but don't feel like testing. I've seen studies that COVID was detectable on paper 3 days after exposure, but I don't know whether what was detected was a live virus or just fragments.
Reads like sensationalized half-assery to me. It conflates a serious and difficult point, which is deciding the relative contribution of aerosols and fomites to infectious spread, which *everbody* has known for centuries varies by disease and conditions, and turns it into a cartoon plucky peasant heroine v. Arrogant Establishment narrative that is probably Chapter 11 in How To Write Hollywood Screenplays. Really, all we need is a tense Hans Zimmer soundtrack and a comely actress to play the lead (while Willem Dafoe can be dragged out of retired to play the WHO baddies) and we're looking at box-office gold.
It's certainly true that the presumption early on was that COVID, like most coronaviruses (e.g. those that cause the common cold) was more transmissable by fomites than aerosols. That was the basis for the early advice, which, alas, turned out to be wrong. But no one could have known that at the time. And the fact that it was hard to turn the gigantic ship o' public health around once it was off and running...well, duh. The presumption that one bright-eyed intelligent worker with a handful of data and a good argument should make such a ponderous org do a 180 inside it's own length only happens in...well, Hollywood. Reality is way more fraught.
But a story that reflected that reality wouldn't be nearly as exciting to read.
Wait, where did you hear that the other coronaviruses spread significantly on surfaces? The viruses that survive long times and spread significantly on surfaces are capsid viruses. Enveloped viruses don’t, in my understanding from virology class years ago and from searching in the literature. I was kind of appalled early on that virologists who should have known better didn’t speak out about the unlikeliness of significant COVID (an enveloped virus) surface spread for months while millions of people were going to great lengths to sanitize their groceries and everything else. (I certainly didn’t bother based on my understanding of capsid vs enveloped viruses.) But maybe I was mistaken and there is literature suggesting that surfaces are a significant source of spread for the other coronaviruses?
You're right, I was sloppy -- I should have emphasized the direct transmission (e.g. touching of hands) rather than fomites. The conventional wisdom for years has been that most rhinoviruses and (ordinary) coronaviruses spread through direct transmission at least as well, maybe better, than through aerosols, hence all the advice about washing your hands, the hand sanitizer craze, et cetera. You can find that advice everywhere. Transmission via fomites varies all over the place, so far as I can tell -- and this is not my area of expertise. I believe influenza has been shown to routinely remain viable for 48-72 hours on stainless steel surfaces and such, which is alarming for an enveloped virus :(
All of which reinforces my argument that, like much else about virology and immunology, It's Complicated, and trying to turn the struggles of the community to ferret out exactly what is what into a Manichaen MCU storyline is at best not helpful.
Ok, thanks. For flu, fomite spread can happen, but I remember reading an estimate that fomites may account for low to mid single digit percent of transmission.
People love to tell the story of Semmelweis - all the doctors of the time thought diseases were caused by bad air, and here's this upstart telling them to wash their hands, and history eventually proved him right. Given the centrality of this story to much of the picture of medicine and public health, it would be not at all surprising if the entire medical community had overweighted "wash your hands, don't worry about the air so much" for over a century.
Regarding the "It should take 0.5-2 hours to write an application (setting aside time actually planning the project)" part, I haven't looked into this particular process, but for the grant applications I have written (various academic options) it generally takes 40-100 hours of work in addition to actually planning the project.
I assume that your process is simpler, but I'm also quite convinced that if you would do a post-application survey of "how much time did you spend on the application" then almost noone would answer 0.5-2 hours, and the majority would be 10+ hours even if you think that it should take less than 2; people applying for grants generally do spend excessive time on various details that may seem trivial to application evaluators.
- By "It should take that long", I mean "It taking that long is fine and is what EA Funds wants". I didn't mean "This is my prediction of how long people will actually spend on it". I should've been clearer about that.
- For context, the EAIF application page says "We aim to keep our application process very simple and fast, and recommend that applicants take about 1–2 hours to write their applications. This does not include the time spent developing the plan and strategy for the project – we recommend thinking about those carefully prior to applying." https://av20jp3z.paperform.co/?fund=EA%20Infrastructure%20Fund
- I'd indeed guess the mean and (less confidently) median length of time spent on applications, setting aside planning, is >2hrs. I think this is somewhat unfortunate.
- I'd guess that the median is <10, maybe around 3hrs or so?
- I'd guess the mean is notably higher, maybe like 6 hours or something? (I feel even less confident here than for the median.)
- I'd guess that how long people spend probably varies wildly, driven largely by how large the grant is, how crucial it is for the person (like does this determine the next 6-12 months of their career/organization?), and how integrated into the "core EA community" they are (the more they are, the more they'll believe that a quickly written application is fine and the more they'll feel confident relying on our prior knowledge of them).
- I'd guess that at least 10% of applicants indeed spent <2hrs. I've seen some applications that were clearly written very quickly (and if I recall correctly most of those were in fact approved, which is at least weak evidence that quickly writing an application does little to harm one's chances).
- These guesses are informed both by applications I've read, interactions I've had with applicants, and the ~3 times I've been involved in writing an EA Funds application myself (before I joined the EA Funds team). I haven't looked for or gathered systematic data, though.
- I also wrote EU grant applications for a company before, and that was horrifically time consuming. I'm quite confident people tend to find the EA Funds application *far* less time-consuming than grant processes for e.g. the EU or academia.
Asking people in a post-application survey how long they spent sounds interesting and potentially useful. I'll look into whether that's ever been done and, if not, whether it should be.
> I'm quite confident people tend to find the EA Funds application *far* less time-consuming than grant processes for e.g. the EU
This bar is awfully low. EU funding requires so much paperwork that in numerous cases organizations decided to abort applying as expected applying cost was similar to expected funding received
> I'd guess that the median is <10, maybe around 3hrs or so?
I expect this to be untrue, with median for funded grants at least full day of work. Maybe even multiple days. (based on own experience of applying to grants that claimed to be easy to apply)
> I'd guess that at least 10% of applicants indeed spent <2hrs. I've seen some applications that were clearly written very quickly (and if I recall correctly most of those were in fact approved,
Though maybe I actually should send first draft after first hour of work?
My guess is that someone with an appropriate level of knowledge of the issue they want to work on, (assuming a Bachelor's Degree), you should be able to vomit out a short proposal in less than an hour. Maybe reread once or twice and submit?
If you are working with a collaborative endeavor which isn't funds-constrained, you only need to pass the threshold for "worthwhile funding" rather than a tournament-style competition where you have to be both "worthwhile funding" as well as "better proposal than most of the other applicants".
ya, when Scott said his application would take "15 minutes" or whatever ... I knew there was no way I could do that. but i also realized that's me, and wasn't sure how much i could generalize there
Does anyone know if Nick Bostrum has ever addressed computation loss in chains of simulated worlds? ie That there might be a limit to how many dreams in a dream there can be? Never understood from what I’ve read of him why simulation B from parent universe A couldn’t turn around and then simulate parent universe A in such convincing fashion you couldn’t tell who gave birth to who if you can just carry on that kind of nesting forever.
Have not read Bostrum, but I wonder if software emulations of older computers might offer some insight?
For example, these days it is routine for newer platforms to emulate older platforms, because they are much more powerful. It’s fairly easy for a current Wintel PC or Mac to emulate an Apple II.
I guess it is theoretically possible to emulate a Wintel PC on an Apple II, but it would take a tremendous amount of programming effort, and the resulting implementation would run incredibly slowly.
What you’re saying makes sense to me and is more or less how I think of it but it seems like that exponentially caps your number of universes. You could bud many from the true parent, some from the children, fewer and fewer from the grandchildren. However, I’m sure Nick Bostrum/someone else who ascribes to that theory must have thought of that and have a counter argument. I’m just not aware of what it is and can’t seem to find it. Or maybe the argument is more limited in scope than I understood (this is true only when you have near lossless reproduction, etc).
There are two obstacles - essentially information and computing time.
Simulating an universe requires storing at least as much information as anything it can simulate - so a finite universe can only simulate equal or smaller universes (although for many needs properly simulating a tiny fraction of that universe, like a single planet, is sufficient and everything else can be faked relatively cheaply, but we'll stick to "proper simulation of an universe" for the sake of argument). I'm not sure if infiniteness solves this problem, since at least we seem to be limited to a finite observable universe.
A similar restriction applies to computing time - simulating an universe takes at least as much computation as any universes it might simulate - however, that can be solved with time dilation, having the simulation "run in slow motion"; perhaps you're spending a year of limited computation power to simulate a microsecond of more complex computation. If you can solve the storage problem (perhaps "cheating" by using an infinite universe) then you *can* have arbitrary nesting, it's just that the "simulated-A-within-simulated-B" would be much slower than the actual parent universe A but have no ability to measure or detect the slowdown.
Also, there is the notion of arbitrary causal intervention. The parent universe can have the ability to arbitrarily alter or destroy or clone the simulated universe. That does provide an ability to distinguish who gave birth to whom - although before *trying* to distinguish them I'd like to refer you to the short story "I don't know, Timmy, being God is a big responsibility " at https://qntm.org/responsibility as an interesting exploration of implications of such nested simulations.
Fantastic story! I think I may need to restate the part of the hypothesis I had in my head to check my thinking and it may impact the story as well. My understanding of Bostrum is that we are more likely than not to be living in a simulation (just like in the story) given a nesting of simulations springing from some kind of ur parent universe that has some semi magical property of being the prime reality. However my thinking has always been that can only be true if the ur universe and its simulations has some kind of infinite nesting power. (This may be related to mathematical vs computational universe arguments but I’ve only been aware of those for about a year) If it’s not infinite it seems like that dashes the argument quite a lot as even near infinite nesting power would see an exponential decline in how far down you can really go before you run out of compute power. (Let me know if I’m thinking about that wrong, my gut is saying exponential but it might be just a straight down trend) That may be a big number still but it makes the whole thing a bit more shakey for me. However, I’m sure that particular problem has been thought through by someone in more detail. The other piece I have not seen is that it seems if you do have an infinite ur universe reproducing infinitely fertile children, that you run into circular child/parent problem that impacts your ur universe as well. If everyone gets the same juice as everyone else so to speak, I no longer understand what could make the ur universe special. I don’t even know if you could tell from a Gods eye point of view if you tried to turn the simulation off because if the number is infinite there’s no reason the simulation has to only be supported by one peer universe so if you turned it off in one there’s no reason not to think it doesn’t just continue on supported by another peer. So to sum it all up: is Bostrum assuming infinity? Or just near infinity? ie very very very big. Thank you for engaging on this. Rocking a child and trying to not lose my mind and until recently didn’t know a place existed to talk about these kinds of things.
IMHO the proper answer to that is that the questions regarding nesting are orthogonal to Bostrom's argument, as it does not make any assumptions of infinity and asserts that this would work given our expectations that with sufficient engineering work it's computationally possible to simulate vast quantities of minds (not necessarily universes - it's enough to simulate your mind and the behavior of the very limited number of things and people you directly interact with) within constraints of our own universe if it's an actual, "full" universe, and thus implicitly also within our "parent" universe if it actually exists. So the argument would be valid even if the max nesting level would be 1 (the "ur universe" can simulate our universe, but we can simulate just tiny fractions of universes) or even 0 (the "ur universe" can simulate the environment directly experienced by me and/or you, but not a full universe like we expect; i.e. it doesn't simulate the physics of the cosmos but just simulates the astronomy articles I've read and the pictures that I've seen).
You only need enough compute power to simulate one universe. I don’t know why multiple levels are required, let alone infinite levels.
You don’t have to simulate a whole universe, you can cheat, like is done in current video games. Many games show graphics of a larger world, but don’t allow the player to access it.
Maybe the speed of light limitation in our universe is some kind of cheat, to make life easier for the authors of our simulated universe.
Thank you. I’ll need to chew on that for a bit. I’ve always assumed I’m too boring to emulate and much more likely to be some happenstance coincidence in calculating the digits of pi than a standalone. Causal connection may still bug me on that one but I’ll need to think on it more deeply. (ie how much can you get away with not simulating before someone is having to do a lot of weird stuff to make stock prices self consistent for instance.)
The simulations can evolve. If you assume the sim devs are watching us, they might say “oh, they’ve got telescopes now, I guess lights moving around in the sky aren’t going to cut it any more. Next release will have planets you can land a probe on.”
That seems completely doable/possible, so no argument from me especially if they can impact my memory so they don’t ever write themselves into a corner. (Although that does seem like a fun story explanation for why relativity and quantum mechanics don’t reconcile) I guess concern there is that it seems like that also cuts down on the number of simulations you can have. I get you don’t strictly need infinities but it does seem you need a pretty big number before you flip from “possible” to “most likely.” Let me know if I’m beating a dead horse.
Using your example of stock price self-consistency and my example of astrophysics, I'll illustrate some simple cheats to take.
A key observation is that the stock-price self-consistency and behavior of astrophysics is only an issue if the simulated mind(s) care about the details of stock prices. If you're simulating a small set of minds, that won't always be the case.
As suggested by Tom Bushell in another comment, you can start simulating them whenever needed. This may need some extra effort to ensure consistency, but this can be done.
One option is to arbitrarily alter memories and artifacts in order to be consistent with the new simulation - "Oceania had always been at war with Eastasia", the stock prices have always been like this, and galaxy rotation has always behaved like this.
You can also simply nudge the relevant people to not care enough. Whenever they try, distract them with something else so that they don't look into details. For a fictional example, look at how in the Truman Show they protagonist was prevented from traveling outside the town multiple times.
And the final option is simply cut the simulation and re-run it - perhaps after enabling the extra simulation feature that turned out to be necessary, perhaps just with different starting conditions. For most purposes I can imagine, there's no need for the simulation to run forever. Like, perhaps there are ways to uncover the wrongness of *our* universe, but that might simply result in an end of the world if the simulators consider that uncovering details of the simulation disrupt their use case for needing that simulation.
Since this is the festive season, also for rationalists, I would like to share Händel's praise of the Age of Enlightenment, which should find resonance in this community:
The recently discussed omission of Xi from the Covid-19 variant names has got me thinking about the kabbalistic implications of using Greek letters to name virus variants. (It is possible that the observation has already been made by ACX or comments and my brain simply decided to co-opt it. If so, please tell me!)
I think the decision of the WHO to use this naming scheme will prove to be extremely foolish and dangerous. To wit: the last letter of the Greek alphabet is Omega.
Omega famously appears in the book of Revelations, e.g. 21:6: "I am Alpha and Omega, the beginning and the end." From my limited understanding, "end" here would mean "end of everything, end of the world". By analogy, it is also used to refer to death, e.g. to mark "Year or date of death" (as per Wikipedia). Do we really want to risk naming a virus variant thus?
Much has been made of the Omicron ("little O") variant, recently. Unsurprisingly, given its name, it seems a bit less deadly. We should imagine that the Omega ("great O") variant will be quite more severe. If one asked an ancient Greek for the factor between "micro" and "mega", I guess they might say perhaps one hundred or so? Science, however, has firmly established that there are twelve orders of magnitude between "micro" and "mega". This is somewhat surprising given that Omega can hardly kill trillions of people, but I think it kinda makes sense if you take the long view: by eradicating humankind, Omega would not only wipe out the present population, but also prevent humans from populating the galaxies, reaching the singularity or what ever else the future might hold in store for us.
I would thus propose that we rename variants of concern using less problematic names. Perhaps the WHO can cut a deal with Disney and use their character names. A Dumbo, Bambi or even Shere Khan variant seems much less likely to cause the death of humankind. As an added bonus, instead of skipping Xi we could just skip Winnie the Pooh instead.
Oh, hey, I know about that Bible passage! "beginning" and "end" is "arche" and "telos": two words which can be used chronologically, but both have a whole interesting spectrum of meaning beyond that, and there are reasons to think that it's not the chronological meaning here. Arche is like source, progenitor, chief, leader... so kind of both "beginning" and "ruler". Often rendered in older English as "head", as in "head of the family" or "headwaters" of a river. Telos, we already use in English pretty straightforwardly in "teleology": purpose or goal, as well as endpoint.
I think theologians generally take "arche" and "telos" here in the non-chronological sense, because in Revelation 22:13, Christ calls Himself alpha and omega, arche and telos AND protos and eschatos. "protos" and "eschatos" are much more straightforwardly chronological, so you'd think then that "arche" and "telos" are meant in a less chronological way.
Second the teleological readings of ἄρχη and τέλος. Wanted to comment that ἄρχη is also a reference both to John 1:1 "in the beginning [ἐν ἄρχῃ] was the Word [λόγος]" as well as to Genesis 1:1.
ἄρχη esp in conjunction with λόγος can also mean "principle, that from which the rest is derived"
τέλος makes me curious how much Greek Philosophy had trickled down into Nazareth, since Plato and Aristotle played a great role in investing the term with the meaning you reference above. For the curious, τέλος also will mean "death" and "payment᾽"
In early 2020, sales of Corona beer dramatically dropped. I'd assume that for similar reasons, in mid-2021, Delta Air Lines fared noticeably worse than otherwise similar airlines, but I haven't heard of anything like that. Has anybody?
Probably because for as many people avoiding it, there were as many wise-asses such as myself who picked up a couple of cases when the office closed, and kept on buying them. (Corona is my usual beer to keep around anyway though)
The hurricane people figured out after 2020 that Greek letters were a bad choice to use for hurricane names when they ran out of names from the list. Hurricanes Eta and Iota got bad enough that they met the criteria for retiring names, and Delta almost did too. (https://en.wikipedia.org/wiki/2020_Atlantic_hurricane_season#Retirement) That's not a good feature for a list of names, especially when so many innocuous companies like Delta Airlines and the Delta Omicron fraternity (https://en.wikipedia.org/wiki/Delta_Omicron) end up sharing their names with these things. On CoVariants (https://covariants.org/), where they track many of the covid variants, some of the 2020 variants they were tracking got the name of generic bird species (pelican, robin, etc.) I can't tell if that's sufficiently better.
Is anyone doing work on building a 'rogue AI detection / containment system'? Would it make sense to do that?
Do AI safety researchers generally assume that a self-improving AI s system would be on our radar before it 'escaped'?
Or would it make senes to have 'honeypots' that exist specifically to detect self-aware systems attempting to manipulate human beings, so that they can be shut down?
I'm not sure such a thing would be possible, at least, not without the support of a major tech company or government. You can't just look around the internet and watch for users buying suspicious amounts of AWS Cloud Computing time, for the same reason that people can't look around the internet and see your credit card number when you send it to Amazon. Detection would have to come from the AI's end (the researchers keeping an eye on it) or from Amazon itself.
A honeypot might be possible, but what sort of bait would you put out that would attract AGIs but not humans?
The only thing I can think of to do in this department would be improving ways to trace digital records to an actual human. E.g., making sure the person signing up for a big chunk of AWS resources is an actual human, and their credit card belongs to an actual person whose identity hasn't been stolen. Which could also be good for fighting fraud in general.
I believe (but I am not sure) that the general expectation is that such a system would not be possible.
Like, there is a bunch of research being done on tracing the intentional opinion manipulation that's done for both international politics and internal politics reasons (IMHO more people are working on that than on AI safety), and some campaigns can be identified but the success in identification and attribution of such manipulation is often very limited and sporadic, and I'm quite certain that a sufficiently smart AI would not be properly detected by those efforts.
And the key practical things that an "escaped" AI would need to do in order to increase its power (obtain large amounts of untraced money, obtain large amounts of arbitrary computing power, anonymously hire some people, anonymously control some legal entities) are things that we know for sure can be done by not-that-sophisticated criminals and in many cases stay undetected for many years, so apparently we are not able to have such things detectable rapidly and reliably.
Like, if an escaped AI with the capability level of a somewhat talented teenage hacker would have engineered one of the many last year's cryptocurrency thefts or ransomware attacks, did some basic laundering (perhaps hiring money mules with "work from home" spam, as human criminals do) and then used the proceeds to hire some lawyers to incorporate "Ainur Homomorphic Encryption Investments Inc" and a dozen other companies, and then rented a few thousand servers on various cloud providers, that would not be detected because, really, there's nothing unusual to detect, it's all routinely done by non-AIs (at least I assume so - perhaps some of the current cybercrime ransomware gangs are run by some AI, how could I know?)
I would say the key direction of AI safety is to work on the "pre-escape" phase - e.g. ensure that the AI does not "want" to escape, or that any people and organizations capable of making such a system get informed enough about the risks to limit their activities and not build something that would escape.
What are peoples' opinion on chiropractors? Pseudo-medicine practiced by charlatans or vital restorative therapy? I hurt my back a couple of weeks ago moving a desk, and it's still rather stiff and sore. My lone previous experience with a chiropractor when I hurt my neck many years ago was not very gratifying, but I'm in rough enough shape to at least consider giving it another shot.
My understanding is that the modern consensus on this issue is that a good chiropractor can provide useful treatment for actual musculoskeletal issues, especially back-pain related ones. The problem is when they claim to be able to do other stuff.
I have several acquaintances who have had positive experiences with chiropractors, but my one experience was meh.
If you suffer from chronic or recurring muscle pain (and even some joint pain is actually caused by muscle issues), I would strongly suggest you look into Myofascial Trigger Point Therapy. Many massage therapists and some chiropractors are familiar with this, but not so many MDs.
I prefer to self treat, using hard rubber balls and a Backknobber tool to do deep tissue massage.
Fellow fan of hard rubber balls for the back here. I learned about using pressure on so-called trigger points from Paul Ingraham's excellent site Pain Science (https://www.painscience.com). Wow does that work well! Feels great to lean into that ball when it's under a trigger point, and back feels better and is looser for hours afterward. Paul describes the feeling of applying pressure to a trigger point as "like scratching an itch you didn't know you had" -- I agree.
Chiropractor practice is an alternative medical practice that doesn't stand up to scruteny. Physical therapy is a mainstream medical practice that... often doesn't stand up to scruteny either. Also, lots of people with mainstream degrees (DPT) hold themselves out as chiros based on the fact that people expect back pain doctors to hold themselves out as such.
Do you know any recreational athletes or lifters? I'd ask them for advice. Then call up those places and ask if they use the biopsychosocial model of pain in their practice. The active are most likely to hurt themselves, and actually care about their outcomes. These are likely to be the most up-to-date and evidence based therapists, and I have a feeling the reader at ACX would be turned off by woo that is very common in this field.
I've had excellent results from physical therapy, but I suspect the quality of the therapist is critical. Quality of the therapist may also be critical in chiropractors, my experience with them is so-so. I think they are better at dealing with symptoms than causes. Physical therapists have given me exercises to do, which when actually done, cured the problem.
I've gone a chiropractor occasionally for 30 years (maybe on average 4X a year.) A good chiropractor can make a massive difference for some problems, but not all chiropractors are that effective, and it's only some problems.
If you have a friend who is fairly active, and strongly recommends a particular chiropractor who they see less than once a month, I think it would be worth trying. (You are looking for a chiropractor who is effective--the ones I consider less effective tend to make people temporarily more comfortable.)
Here is one piece of info I know from life experience, one opinion of chiropractors without much to back it up, and one excellent lead on dealing with your back:
-Info: Hurting your back moving something heavy is very common, and the older you get the likelier it is to happen to you. Unless it is happening to you often or it hurts like hell or it does not seem to be improving, you do not need to seek out a Medical Intervention. You can reduce the chance of further back pain attacks by improving physical fitness and educating yourself about back-sparing ways to move heavy items.
-Opinion: Chiropractic "medicine" is not a science-based treatment approach, and therefore is highly suspect.
-Advice: Go to Paul Ingraham's excellent site Pain Science (https://www.painscience.com) and read his info and advice about back problems. Paul's advice is smart and evidence-based. Also, he's funny and writes well.
Chiropractors are good for adjustments if something is out and causing limited motion or discomfort. I’ve heard you should look for Palmer School trained chiros. But some chiropractors will try to sell you on weird and wacky unproven stuff.
I recently saw a chiropractor for the first time. I'd already had lots of physical therapy, which had reduced lots of pain through strengthening exercises, therapeutic massage, and learning new movement patterns. But I had lingering pain, and I wanted to see what a chiro could do.
The chiro was clear that his approach was not the typical "see me every 6 weeks for the rest of your life." He worked on my back and neck and hips a few times and acknowledged that I don't seem to have structural issues and I should continue to work on strength. He also had a PT assistant on staff who helped me put together a strengthening plan.
My advice: Talk to the chiro on the phone first and see what his or her treatment philosophy is.
As I mentioned on a previous open thread I'm working on a fantasy world strategy simulator and while I have most of the design down and in code/data there are a few things I'm still thinking about.
I'd like to have more detail in the sort of assimilationist practices of historical empires. Rome, Maurya, Macedonia, Carthage, China, and so forth but also stuff like the Norse invasions of England. And maybe something about the larger African states pre-Scramble. Actually the Russians, Dutch, Ummayads, English, Persians, or Ottomans would work as well.
My goal is to have a variety of options that aren't too focused on one historical era and location.
Additionally I'd like to read some good articles on interesting agricultural and hunting practices in different/rare climates. I'm still debating whether to have "discoveries" for farming, hunting, construction, raw resource processing and so forth or just a slow generic "knowledge" increase with little variance over time. If I do discoveries I'd not a lot of techniques and equipment to be discovered for a large variety of climates and processing methods.
Also I definitely need some reading material for classical and medieval trade/intelligence networks for flavor and character action ideas as well as other purposes since Axioms Of Dominion is a DIP strategy game. Diplomacy(foreign affairs), Intrigue(shadowy affairs, Politics(domestic affairs).
Ah this is great. He covers places besides Rome? Greek and Persian would be an interesting contrast from a similar time period. I'd really love to get Carthage and maybe some stuff about Caeser on Gaul. Small foreign populations interfacing effectively with very different native cultures.
The table of content pages are really incomplete so it's hard to remember everything. I think you'll find interesting things for your purpose in almost every article. From memory:
1) the series on Europa Universalis and Victoria II are probably what you want to read first for your purpose. "There are no empires in age of empires" is also a good read for the same reasons.
2) Two series on representation of certain cultures in movies/novel/video games : This isn't Sparta and the Fremen mirage (on nomad peoples). I think those definitely discuss quite a bit of interactions between empires and outside populations.
3) Bread, clothes, iron, fortifications.
3) The epic 14-part analysis of the battles in the Lord of the Ring.
4) The 4 or so articles that are analysis of classical text will probably be interesting for your DIP part.
Okay I read a couple dozen posts and a lot of it was pretty useful. The AoE4 part just told me I mostly had key stuff locked down. His CK2-3 and Vicky2 stuff was good. Turns out I read the EU4 stuff a few years ago. He said he doesn't have much to offer outside of Europe sadly.
I did a post discussing implementation of a system pretty simialar to the Socii and Auxilia. Might have to add more info to the how it works in game part.
Okay so far I've gotten through half the queens latin series since I've had to do some other stuff. I definitely want to be able to, more so than something like Imperator, which is currently the best Paradox sim game, represent the socii system as well as potential variations of it. I think that currently there are a few major aspects that aren't possible.
Hi Dirichlet-to-Neumann, are you a researcher in geometric analysis by any chance? I've been writing (well, trying to write) papers on Dirichlet-to-Neumann operators in grad school!
I'm a PhD student only. I do mostly functional analysis and PDEs, and in particular I've studied the DtN operator in relation with the Stokes operator (-\Delta u+\nabla \pi).
That's great! I'm a PhD student myself. You can email me at auk480@psu.edu if you'd like to know about my research, or talk about rationality-adjacent topics, whatever. I work in conformal geometry
His blog's been getting a lot of attention lately (~2.8M page views last year up from 1.9 in 2019) so you might just be at the end of a long queue.
I'll also add that if you're a very big fan of his writing I've found his Patreon to be worth donating to. A few bucks a month for an extra post (and a warm glow of satisfaction) each month. The topics are more on how the blog and his academic career are doing, than history per se, YMMV.
I do not. There are screenshots of me doing stuff spread around the web circa 2015, and pictures of map gen results and such. But that isn't very extensive.
The blog posts on substack and the information in the wiki pretty accurately describe, minus visuals, what the game will involve but the AI can't do most things since my programming strategy was to get the design and backend stuff done, code the data structures and file loading, then slowly add stuff to the game and UI for the player and only write the AI once a given aspect was more or less done.
I'd expect to make some videos or something when I am about 80% done, currently my estimate is 40%. So like mid-summer?
Does anyone believe there's actual health benefits to Dry January- to going without alcohol for 30 days? I'm a moderate drinker approaching middle age, so I do have general concerns that Alcohol Is Bad For You, and I'd be happy to go without for a month if I thought there were clear-cut benefits to one's liver or other organs. Are there? I Googled the topic, and of course there are many Pro pieces, but the more scholarly ones seemed skeptical of a concrete improvement to one's physical health in such a short time frame. I'm not interested in going without social drinking for longer than a month. In general I have a very healthy lifestyle otherwise, including fanatically healthy diet, lots of exercise, etc. Any thoughts?
I don't know of any physiological benefits, and I left the Catholic church two decades ago, but I do think they are on to something with the idea of periods of fasting or abstaning from something just for the experience of it. From a purely mental POV (and spiritual too, if you are into that), I do it and find a benefit.
One of the effects of alcohol is that it affects how well you sleep. In particular it disturbs REM sleep. I think this is a pretty instantaneous effect, so you have the effect on days with alcohol, and not on days without alcohol.
I am not sure about long-term effects, but you might sleep better during this month. It might be worth a try for finding out whether you actually feel a difference. You say that you are not interested in changing your long-term behavior, but that might depend on your experience, right?
Should mention that I tried Dry January last year (actually for 5 weeks). Maybe my sleep quality improved marginally, but not enough that it felt significant. I also don't drink alcohol Monday through Thursday now (I do this 12 months a year and have for years), and save social drinking for the weekends- again, maybe there's a marginal change, but just not a huge one.
My health goals for a Dry January would be lower future cancer rates and liver damage. If I can't achieve that with 30 days off- probably not worth it to me
I see. Then it doesn't sound to me like the Dry January would change a lot. I could imagine problems with liver regeneration that accumulate slowly over a long time for people who drink heavily, or people who drink daily. But you don't fall into these categories, so I wouldn't expect that it has additional effects beyond cutting your annual alcohol consumption by 1/12.
I've never heard anyone claim that 31 consecutive days of abstention from alcohol would have any more direct effect on your long-term health outcomes than an extra 31 days of abstention from alcohol scattered throughout the year. I think the point of Dry January is more to give someone a Schelling point to practice abstention, which can then help the person understand the actual effects of alcohol and abstention on their system, and also help develop the habits that reduce the amount of alcohol they drink at other times.
So the original purpose of memory foam was to protect astronauts during landing by absorbing the impact. Viscosity helped flatten the curve of g-force and prevent any rebound acceleration. But in everyday beds and pillows, viscosity seems to have negative utility. Viscosity makes repositioning less comfortable and makes surface blood flow more obstructed because the material can't deform easily for a single heartbeat. So why are nearly all the foam mattresses/pillows on the market high-viscosity foam aka memory foam?
Short answer is marketing. The long answer is still marketing but also that it is just really hard for individuals not willing to invest in a sleep study and long term testing to find out what works for them.
Many matresses now have a layer of memory foam as only one of the layers on the bed. A common set up is a thick layer of very stiff foam called a support layer, a layer of memory foam called a contour layer, and then a thin layer of very plush latex called a topper layer. If you get the firmnesses right given your weight and curves, you can get a bed that supports back or side sleeping well and is very deformable (and cool, latex is a much better heat conductor than memory foam).
I’m curious to know if anyone is aware of a project that is building a system to stitch together different people’s perspectives on how to think of various important issues. Kind of like a wiki-fact check site. A place to go to read about both the pros and cons of [insert contentious issue here].
I’m hoping to find a novel website/organization that is attempting a grand vision of making it possible to explore debates, decide where you stand, and as one potential feature allow you to indicate the main contributing fact/argument convincing you, so that you can be notified if that sub-fact’s truth or evidence value ever changes
You may be interested in the most recent article on my substack, although it’s not exactly what you’re speaking to here and doesn’t exist. That said, I lay out what a full adjudicative and democratic fact-checking process would involve and I think it may speak to some of the values you’re looking for.
Been having a discussion about the relevance of common school knowledge for technical and non-technical people, and how important various aspects of scientific literacy are, and how to increase rates of scientific literacy.
An argument I expect to be uncontroversial here, is that scientific literacy is low is most especially due to how material is generally introduced, as a constellation of beliefs that don't predict the future, or pay rent. Therefore, any attempt to improve education in general, and scientific literacy in particular, would have to reorient teaching practices further toward processes of how to ask and answer questions eg: "how do I figure out what this material is made of" or "how do I sanity check my mathematical model", and away from trying to remember particular facts.
But there's an issue: the above two questions are pretty niche, even for technical people. Most people are going to remember science and mathematics as a loose constellation of facts, simply by nature of not being scientists, mathematicians, and engineers: they aren't using the question asking methods I describe, so those neural pathways will fall away leaving the constellation of facts behind. So too it goes with scientific literacy.
Using chemistry as an example, most technical non-chemists will have taken several courses in chemistry, but wouldn't won't likely be able to tell you much about how to go about answering basic questions in chemistry like "what's this made of", or what X ingredient in their shampoo might do. So for all the non-chemists, chemistry becomes mostly trivia too.
But, the technical folks WILL retain certain knowledge about things like the scientific method, journal literacy, and how to go about attack technical problems in their domain, and might be able to pick the back up, or at least read the abstract of a paper on chemistry. That ability seems to be what we call scientific literacy.
But if we've just supposed that the pathways to asking and answering technical questions require consistent application to be maintained and used, haven't we implied that scientific literacy is basically a function of how many people in a given society are technical? If that's the case, trying to raise scientific literacy by changing education seems like a red herring, and the main focus was actually just trying to get a larger ratio of technical people in society in the first place.
I agree with you that perhaps scientific literacy isn't needed by most people. I wonder if that desire to teach more people the scientific method is so that they'll indirectly support our political opinions :)
A lot of material here and elsewhere in the rationalist community pertain to AI safety. As an outsider and passive consumer of rationalist content, the topic has always come across to me as a little sci-fi-y and virtualistic. What's the best digestible reading or listening that would help me appreciate the seriousness of the issue and risks involved?
I'm not exactly sure what you mean by "opinion" or what counts as a "scientific paper", but in contemporary analytic philosophy of science, there has been a burgeoning research area investigating the role of individual bias towards or against various scientific theories as a way of ensuring that the scientific community maintains sufficient diversity of opinion to properly explore all the angles that should be explored. Some of this picks up on Thomas Kuhn's ideas in The Structure of Scientific Revolutions, which many people naively read as saying that science is irrational, but others read as saying that individual irrationality leads to community rationality.
I downloaded the paper you suggested. I thank you for the suggestion, and I will read it thoroughly. A precursory reading suggested to me that it represents what I was looking for. What I meant by using the term 'opinion' was unproven but seemingly rational formulation--an ugly way of describing an informed opinion.
It is quite possible for rational people to accept theories that are later reversed, sometimes after having been accepted as true for thousands of years; e.g. that the sun revolved around the earth. We know that some rational people in the distant past held to the subsequently held 'truth' that the earth revolves around the sun, which had been rejected by the majority. From my point of view, earth-centric was a widely held opinion of rational thinkers.
New preprint suggesting that unlike earlier strains, which enter the cell directly from the surface, omicron viruses sneak inside endosomes (I think they're pockets of cell membrane that are used to recycle all the protiens on the cell surface) and get in from there.
This means that it doesn't do the thing where multple spike proteins on the viral capsid bind to several cells and make them combine into one big infected blob. Which is apparently what wild-type through delta did. I was happier not knowing about syncytia.
This means that its preference for infecting different types of cells is different from what you see in previous strains (infecting more epithelial cells in the nose and throat, fewer pneumocyces in the lung?). This is consistent which higher infectiousness but lower severity. I am not a biologist.
Is there a good Superforecasting/Rationalist-adjacent style book to investing? I'm thinking something relatively accessible to someone with a decent (micro)economic literacy and knowledge?
Investing advice/books seems oddly polarized (not in a big P politics way).
Depending on how you feel about hardcore value investors, I found Seth Klarman's Margin of Safety to be a nuanced take on managing risk and looking for arbitrage opportunities. Physical copies are basically nonexistent but most restructuring bankers or distressed debt investors will have a pdf somewhere.
With that said, I would be super interested in what others suggest.
This was not one I came across or expected, but very much intrigued and looks to this untrained eye as what I was looking for ("This has caused physical copies of the book to be worth to $500-$2,500 a piece. The high price of the book has resulted in piracy of the book.")
I have a clear recollection of a website which had human longevity interventions sorted by level of evidence, so exercise -> eat well -> ... -> metformin etc. Some cursory Googling is not finding such a thing. Anyone know what I'm thinking of?
Do you know what should come back? Illustrations in books. Not talking comics here but those excellent occasional illustrations you would get in Victorian era novels, like Dickens. Maybe on a kindle or ebook they could move a little, like a gif. Scrooge meets Marley and you see him come in.
I'm looking for a book about a war between the USA and USSR in an alternate universe where nuclear weapons were never invented. Does anyone have suggestions?
Not exactly what you're looking for, but I really enjoyed Tom Clancy's Red Storm Rising. Published in the mid-'80s, it was set in the then-near future. NATO and the Warsaw Pact nations wage a non-nuclear WWIII. I read it back in the day - should reread it to see if it's held up well.
Christmas is over, but I recently discovered that there is a cartoon where the Grinch gets psychoanalyzed by the Cat In The Hat. No, really- the CITH gets the Grinch on his unlicensed psychiatrist couch and gets him to stop being a jerk by reminding him of his mother’s love. I thought Scott might find this amusing, though I wouldn’t recommend watching it without liquor. Despite being generally terrible, “The Grinch Grinches the Cat In The Hat”. won an Emmy for animation in 1982.
Can anyone recommend a good biography of Thatcher to me? I'm a Canadian so I know very little about British politics, but she is apparently one of the figureheads of neoliberalism and as a dirty neoliberal I keep hearing people complain about her in ways that make me go "She sounds cool, I should learn more about her."
I think that with a modern figure as polarising as Thatcher, you're going to find that basically all biographies are either Pro or Anti, and need to be read with that in mind, and I'd be very wary of deciding that you support or oppose her on the strength of one side of the story.
If you're after a pro-Thatcher biography, Charle's Moore's authorised one is probably a good starting point; if you're after a more critical one I'm less sure not to recommend.
For what it's worth, I think "she was cool" is probably accurate but misleading - she was a women of exceptional ability, force of will and charisma, with a very strong and clear vision of what she thought was right, who bestrode the politics of her era like a colossus and accomplished an immense amount, most of it bad.
Self reflection: The pandemic has made me question my interpersonal relationships. And it is turning me a lot more libertarian. For context, I am--as I suspect many of us here are--a mid 20s Northeasterner earning an advanced degree in STEMM.
Consider: the pediatric population. Many of my fellow parents are up in arms about whether they should home school and restrict their children's social circle, possibly indefinitely. And these actions are accepted, if not encouraged, by media and public health authorities. Just today, I read an NPR interview with a peds ID doctor in which a father asked if they could let their 15 month old daughter "return to normal life" after she had a vaccine eg. play inside with other children or if they should wait until she was boosted. What the fuck? The hospitalization rate for covid in young children is less than it is for influenza, or RSV, which are both extremely common. If, in 2018, you socially isolated your children for years on end because you're afraid of them catching one of these illnesses, people would rightfully consider you a psychotic child abuser. But now it's considered virtuous. There is total disrespect for the baseline level of risk people accept every day (consider the similar prevalence, patient population, and pathophysiology of Kawasaki's disease and MIS-C, vs. the amount of press coverage and fear inspired).
Ignorant people are --often justifiably--mocked when they think they know more than their physicians, because their attempts to perform their own research and analysis lead to worse outcomes. But the liberal response to this behavior is to proclaim a blind allegiance to people with the right credentials, a complete abdication of critical thinking. Many liberals think they need an M.S. in a subject to form any sort of opinion at all. You don't need to know about the fine details of vaccine development or viral entry to think "gee, maybe instituting a public curfew for months on end is ineffective at containing the virus and a fundamental violation of rights and freedoms".
One big problem with "science" based decision making and governance is that it tends to downplay the importance of factors that aren't easily quantifiable--having a normal social life, having friends, etc. Sometimes the right response to a concern, or potential danger is "who gives a shit"? rather than a factual rebuttal. By far, the most likely health outcome my children--or you, if you're young and healthy--are going to experience from the pandemic is anxiety, depression, or both. Public health authorities encouraging extremely risk averse, neurotic behavior and then having the audacity to champion mental health care is absurd. Public health and media fear mongering is the cause of widespread, preventable suffering.
I think it's telling that "muh freedoms" is a slur against Republican protests for mask mandates/ lockdowns/ etc. In the end, having a free society means the right to say 'fuck you, I don't give a shit'--and this is important. "Well, you don't really neeeeeed it, and it could be harmful" is an excuse to take away every single thing that makes life worth living. Even if you believe personal freedoms should be overrun for true emergencies, it is becoming harder and harder to see covid as one--the pandemic is being conducted on an opt-out basis, and those who justify measures to curtail freedoms to people unwilling to either a) suffer a mild illness or b)protect themselves with a vaccine strike me as pathetic and life-hating.
Science can tell us things like "How long is a person likely able to transmit COVID after infection", or "this is our estimate of R_0 of this viral strain". Questions like how do we trade off public safety for personal freedoms, or for economic realities are hard. I wish the CDC would be more upfront about that and not say that the "science says" that we can cut COVID isolation from 10 days to 5 days. Science says nothing about what we should value or how we should make these tradeoffs.
I do agree with you that if, as it appears, Omicron will displace Delta and leave us with a more mild virus that is maybe on par with the flu or a bad cold, that a lot of people have to start changing their minds and reassessing risk, and that has to happen now.
There's not much point in mandating vaccines now since most everyone will have had Omicron, or have been exposed to it, by the time such mandates work through the courts and the teeth start biting. It's moot.
I just got an e-mail from my daughter's school that strikes the correct main notes about the focus on keep in-person learning going, but was a bit frustrated to see that this means continued masking and social distancing, etc., etc. People should be asking administrators and people in charge _at what point_ can we remove the restrictions, and I mean tie it to a real number (and I am no longer am convinced case numbers, or case numbers per 100,000 is the right number since Omicron seems so much milder). If policies need to be implemented because, say, the medical system is collapsing in an area, put a sunset clause on the restriction.
People on the left getting conspicuously angry and people who won't or didn't get vaccinated doesn't help, it won't convince them to do it, and anyway it's moot now. I say this as a person who got vaccinated early and boosted over a month ago. Now with Omicron, it make be appropriate to make a flu analogy. No one cares if anyone else gets their yearly flu shot.
I believe that the larger reason for restricting kids is not the risk to the kids themselves, but them transmitting the disease; so the approprate consideration of impact is not the hospitalization rate of young children but the hospitalization rate of the age group that includes you and other parents (and co-living grandparents) of your kids' circle of contacts.
Anecdotally, almost everyone who had Covid in my circle of acquitances had it "brought in the family" through school (probably not because schools are particularly unsafe but because all the other contacts e.g. work/shopping have been significantly reduced and altered); our schools have frequent mandatory testing (but didn't have vaccines for these ages of kids until very recently) so the tracing was quite obvious but not suitable to prevent infecting the family.
On the other hand, I fully agree with your argument that kids experiencing "normal contact" is important and should be considered as something important and relevant whenever considering the risks and costs of mitigating risks.
Scientific experimentation can give you data points. But then the tricky part is to tell a policy story that fits the data points. This story can vary depending on your values and worldview. Science doesn’t make decisions. Anyone who says it does is trying to manipulate you, or may be well-meaning but pushing a specific agenda. “Science-based” decision making is really story-based decision making, where the story contains a few data points narrated with value judgements based on one’s unique worldview perspective (which can appear totally irrational to someone with a different worldview).
Absolutely. Which is why the expression "I believe in science" drives me crazy--science is true whether you believe in it or not! Science can only be used to predict the consequences of certain decisions, it can't tell you which decisions you *should* make, which is a subjective and moral judgement.
Even if my risk tolerance were much lower--the sloppiness and inconsistency of lockdown/quarantine policy makes me hesitant to accept increased restrictions and government powers.
I always assumed "I believe in science" is some kind of (witting or un) weird Möbius strip joke, like saying "My congregation is the First Church of Atheism (Reformed), we meet to reverently not pray every Sunday at 10am -- join us!"
I believe in not believing things? In empirical skepticism? That's....strange. One wonders whether it's possible to fall into heresy. Start disbelieving in disbelief?
For what it's worth, let me make some observations from a few decades further on. First, it would be a mistake to assume parents are uniform on this subject -- or indeed on nearly any subject touching child welfare, alas[1]. I know parents who have kept their children out of school *for going on two full years* because they are freaked out about something happening to their little darlings in that filthy school with the offspring of God Knows Who running around with unwashed hands and snotty noses.
And there are others who breathed a huge sigh of relief when the schools re-opened, and cringe every time the poor kid has to root around in his backpack to see if he's got a mask he has to wear all freaking day, and the psychological trauma it does to them to be (implicitly) told (since they're old enough to know *their own* risk is small) that they better wear the damn mask lest they be guilty of murdering grandma, or at least someone's grandma. Ridiculous, tragicomic, as you observe.
And then you've got the school board, which is in a painful place, caught between the Scylla of acceptable risk (and lowered expense) favored by many, if not most parents, and the Charybdis of the fearful minority and the plaintfiff lawyers[2] who would *love* to have a cherubic promising 8-year-old in a medically-induced coma in the ICU as a client with which to drain $50 million out of the public coffers and into their Next Christmas With Jeff Bezos In St Barts Fund, not to mention the politicians in the capital who, to a man, want to Make An Important Statement About Our Values with the way in which public education is run -- or at the least want to avoid some horrible blowback if some decision or other goes badly wrong (and 8-year-olds in medically-induced comas are features on the nightly news -- and the nightly news would *love* to run with that story if they could, beats the hell out of some tired old mumbo-jumbo about infrastructure or Chinese currency manipulation or what in hell Joe Biden meant[3] to mumble into the mike).
So everybody who has to make decisions for others is in full CYA mode, while everybody who has to endure decisions made for them by others is in full dudgeon, convinced that Authority is 100% inept and/or corrupt fools, and various Other Tribes of their fellow citizens are basically under the sway of Mordor, not mistaken but so deeply evil that only burning them at the stake and throwing the ashes into a volcano would purge the Earth of their malign influence.
Not a great recipe for collective compromise-oriented rational decision-making in mutual good faith.
-------------
[1] In this corner, the challenger Free Range Parenting, weighing in at a trim 180, while in the other at 220 the reigning champeen If It Saves One Child. My Body My Choice is in the audience, seated among her good friends Parental Consent, Trans Rights, and Religious Exemption, all yelling enthusiastically but in support of whom it's hard to tell.
[2] Shakespeare pointed out the only cure for this nigh on 400 years ago already.
[3] Assuming the verb has meaning in this context.
Oh, I agree with all of what you've written, and I don't mean to imply that such parental risk aversion is universal, or even the majority opinion. I DO think that it drives policy, because it influences the public perception of 'acceptable risk', which, as you've said, makes schools open to lawsuits. I also think it has an unfortunate framing effect making 'moderate' opinions seem reasonable that would otherwise be considered beyond the pale.
Yah. School vouchers, eh? Dunno why that slipped off the conservative platform. I was always hoping it would someday make it into law. I guess at least around here it got headed off by the public-charter-school program, the sneaky rat bastards.
Edit: if you don't have yet have kids that age, wait until you ponder sending them to a much better private or parochial high school and paying *twice* for their education -- once to the people actually providing it, and via your taxes to the state for the public HS education you've declined. Bah.
At this point a fair number of states have voucher programs, although generally with some restriction in who can use them and with amounts well below the per pupil cost of the public schools.
Really? I had no idea. It was all the rage 20-30 years ago in conservative circles, I always assumed it got eclipsed by social con or neocon bullshit, and not that it was actually achieved to any real degree.
Just my luck that California never had a prayer of being one of them. I can't even count on the benefits of our public university system, Berkeley is not nearly the bargain it was when I was there :(
Incidentally, I think this is the way the rest of the liberal world often views the rationality community. I think that in fact, Effective Altruism is generally getting things better at both the quantified and unquantified effects of charity than the normie opinion, but many people say "One big problem with "science" based decision making and governance is that it tends to downplay the importance of factors that aren't easily quantifiable".
I don't know how to ensure that one is doing this in the right sort of ways and avoiding the bad sorts of ways, but I do think it's helpful to step back from one's frustration with a particular community applying a particular pseudo-scientific lens to a problem, since that is also what we all here are doing in one form or another.
Thanks for this! I've been very curious about this book, and I've been assuming it was cancelled after the doxxing incident that made TLP stop blogging.
Can anyone give a bit more info about the book, or link to a review that does? The only review at Amazon is by someone who gave it 5 stars, but was not what you'd call articulate about why they did: "The nuggets, big and small. The imperfection. And often, the void I find staring back. . . . the sweat that went into it, finding the time to do it + drinking, the freewheeling ride, the ideas, the brilliant cover, and the Christmas gift release."
There were nuggets. And their size varied, OK, got that. Um . . .
I don't know if there are any reviews of the book, but here's TLP's blog post about what it is (which is still rather vague about everything that comes *after* the porn story): https://thelastpsychiatrist.com/2013/07/still_alive.html
Last week there was much enthusiasm here over the idea of plastic-consuming beetles. I'm ignorant and confused why the existence of plastic is bad. It's non-biodegradable. If you believe man-made climate change is bad, isn't that good? Wouldn't plastic eating beetles speed up the process of releasing carbon absorbed by algae millions of years ago into the atmosphere? Plastic continues to trap the carbon whereas these beetles would release it. Why should I want these beetles?
That's true that there's a pair of competing goals there. I think the main benefit of the plastic-consuming beetles would be to clean up bits of plastic out in the wild in a decentralized massively-parallel way, so that less plastic ends up in the oceans and other waterways as micro plastic pieces that will be very very difficult to clean up
My understanding of the Carboniferous period is that it is a period in Earth's history when plants discovered how to make lignin, but nothing on Earth knew how to digest it. Thus, forests ended up covered in a pile of bark that just didn't break down, and this hugely disrupted ecosystems and led to lots of death and destruction, which promptly changed a few tens of millions of years later when some microorganisms figured out how to break it down. This ended up locking up lots of CO2 in coal formations, that we have recently started breaking down again. (https://en.wikipedia.org/wiki/Carboniferous#Rocks_and_coal)
I don't know whether the amount of plastic we are producing would have significant atmospheric impact if broken down, or whether breaking down the plastic that is accumulating in various parts of the wild would have significant ecological benefits. But the Earth has been through *something* like this before.
Why are so many great contemporary writers Marxists? To name three: our Freddie, Chris Wickham, Justin E.H. Smith.
My best guess is that the fields of Economics and Humanities no longer speak the same language and therefore can't communicate. I believe the above Marxists named are every bit as intelligent and intellectually honest as most top-notch economists, but the language of the respective fields they are in aren't compatible for constructive dialogue.
In agreement with an observation made by Thomas Friedman further below, because it's easier to be engaging and entertaining if you are less strongly moored to objective reality. Marxism has always been a tremendously hopeful and optimistic moralizing story about the destiny of mankind: someday we shall throw off the shackles of our ignorance of the true value of things, eject the parasites, and live anew in Paradise, with plenty for all. Indeed, it's an inevitable consequence of the march of history. Hurray!
By contrast, the traditional economists are a dismal lot, always telling us they can't predict when to buy and when to sell, or even whether the inflation is "transitory" or not -- but by gum they *can* tell us TANSTAAFL and wage-and-price controls won't work, the real minimum wage is $0/hour, the actual always greatly exceeds the projected revenue from some new tax vice dynamic effects, damn them, and so forth. A load of picayune bad news without *any* glorious uplifting narrative at all. Bah.
Somewhat related: I've always thought Paul Krugman could have substantially more impact as a NYT columnist is he had more than a vestigial sense of humor. Jeez, Paul, crank it up!
I'm curious what Scott, and the community, thinks is going on when people "are hypnotized". I seem to be one of the "not hypnotizeable" people, though of course there are plenty of nonmaterial ways to get me to change my mental state: show me a horror movie, tell me an erotic story, etc., so I can believe, to some degree, that a practiced practitioner could get me into a somewhat unusual mental state.
But what's going on when, say, someone is supposedly hypnotized into thinking they're Jimi Hendrix, or a cat, or that their hand is itchy or whatever? I think that social compliance--the pressure to think in a way that will be in accord with group expectations--is unquestionably powerful, so when I hear about a college orientation where a freshman was hypnotized into acting like a cat in front of his peers, I suspect that this is mostly his being game to please the crowd, even if he doesn't know that consciously; if the audience were half-asleep senior citizens he didn't know, he wouldn't be pulled so hard into the role. My hypothesis is that those of us who are not hypnotizable have just as much capacity to get into roles, we're just a bit more aware of those pressures and tend to retreat to first order thinking when we sense those pressures at work, instead of leaning into them.
I have no doubt there exist people who get up and imitate cats entirely out of social compliance or a even just a desire to be the center of attention -- and also that that there are people who are highly socially compliant and therefore lean harder into the hypnotic suggestions, with the result that they actually do experience some alternation of consciousness as a result -- they "get into the role," as you put it.
But there is definitely another relevant variable besides social compliance: ability to experience intense imaginative involvement in something that's not real. Hypnotizability correlates with scores on the Absorption Scale, developed by Tellegen. It has subjects rate how completely they get lost in daydreams, books, movies, etc. All of those are situations where there is no social compliance factor determining how hard one leans into imaginative involvement -- and yet there's a lot of variability in how far in people lean.
Absorption scores decline steeply with age, by the way, as shown both by results on the Absorption scale and most people's life experience. When at age 8 I pretended to be a horse I at least 50% believed I was a horse, and had brief moments when I 100% believed it. In my teens I'd get so absorbed in books that calling my name did not penetrate my attention at all -- somebody had to touch my shoulder or wave their hand in my face. There's much less of that kind of thing in my life now.
Still, I think the situation in which one is being hypnotized can make a big difference in one's experience. I had training in performing hypnosis, but never was able to feel much of anything during the many practice sessions in which members of the class hypnotized each other. What got in the way for me was the social awkwardness of the situation. Everything my partner said when in the hypnotist role pulled for me to do or feel a certain thing, and I knew that if I did not then my partner would be in the position of failing to pull off a successful hypnotic induction. Yet I was very uncomfortable with the option of lying or faking it. My preoccupation with this dilemma blocked my capacity for imaginative involvement in the hypnotist's suggestions. But a few years later I was sitting in on a treatment group run by a colleague, and he did a hypnotic induction of the group. Nobody present, including my colleague, cared how well I responded to the induction, and under those circumstances the guy hypnotized the daylights out of me. I knew I was being hypnotized and going along with suggestions, but in the process of going along with the suggestions I experienced intense imaginative involvement, on the order of what I experienced before I was a teen.
Thanks for this absorption framing. I too remember imagination as a kid that felt almost real; I distinctly remember telling my big sister, with wholehearted belief, that my teddy bear had a headache. And of course, dreams prove that reality is shiftable for me!
I had an experience years ago that's also relevant: at Esalen, I was part of a group that tried doing aura reading. At first only a few people claimed to see something unusual, but slowly everyone got on board until I was the only holdout. Obviously, auras are entirely a psychological phenomenon (aided by the ease of producing vision aftereffects in entirely natural ways), and ever since I have wondered what, precisely, was the internal experience of these friends of mine. I've felt pressure to say I saw them and seriously considered joining in, but I would have had to consciously lie to do so. My best guess is that few other people in the group were consciously lying; I think the distinction between lying and telling the truth just isn't as meaningful for most people as it is for me. A cynical way of looking at this is that most people are in a constant state of lying, and almost never sacrifice the slightest bit of convenience or social comfort for the sake of honesty.
All of which is to say, people who act deeply out of character under hypnosis and report no memory of the experience may have psychologies in which they don't carefully distinguish between reality and fantasy in the first place. If there is no reality you have fidelity to, then you're walking into the hypnosis session already, essentially, hypnotized.
I am also very difficult to hypnotize, so I never had this experience from inside. But when I asked others about their experience, it seemed like a "kinda yes, kinda no" situation.
For example, someone was hypnotized to think they were 10 years old. Afterwards the person told me that in some sense it was a pretense/compliance; that they were quite aware that they were actually just an adult person pretending to be a child. But at the same time, it was a much stronger experience than mere pretending/compliance would achieve -- for example they reported having felt very strong emotions; also acting like a child was very natural for them in the moment.
A person only consciously role-playing a child to comply with the instructions would need to continuously pay attention to "what exactly am I supposed to do now", instead of just acting naturally; and they wouldn't feel strong emotions (unless they consciously decided that this is what they should do, and then somehow manipulated themselves into feeling it). So clearly there is something beyond conscious compliance.
My guess is that part of it is you consciously playing along (or at least not fighting against it), but another part is that some parts of your brain that are not under your direct conscious control are also playing along. That is, hypnosis can recruit some parts of your unconsciousness that mere pretending would not.
When I've acted in plays, I've experienced something similar; when my character was angry, I really was angry, even though I knew I was playing a part. It might be that hypnotism mostly consists of creating a situation where immersion in a role is socially acceptable, as well as expected, for people who aren't used to blurring those lines; and having had them blurred, they might have less of a sophisticated understanding of their own experience than a trained actor who can get legitimately weepy on command, but who knows the sadness isn't "real". I wonder what experiments you might devise to test this theory; you might rate people on hypnotic immersibility, but then put them in a situation where reality and fantasy would call for different responses, say making them think there's a mousetrap in a box IRL, and then under hypnosis telling them there's a cookie in the box instead, and measuring their responses. I would guess that, adjusting for apparent immersibility, actors would hesitate more.
I was concerned after a couple months that the flow was slowing but after cleaning off some mineral buildup on the filter elements it's back to several gallons a day, much more than we need for drinking and cooking purposes
Does anyone have actual numbers on how much worse rapid antigen tests are at catching omicron? I've seen a lot of headlines that say "less sensitive", but not a lot of comparative confusion matrices.
So I think I want to take a shot at the NMN/NR supplements, but the entire field of supplement marketing looks like a howling wilderness populated entirely by predators greased in snake oil.
Scott's "20 bucks a bottle" comment in the lifespan post may be situating my estimate a bit, but the prices I am looking at seem really high (honestly even 20 bucks a bottle seems high if it's 30x300mg NR, given the clinical trials were testing at 1000 mg daily). What I have seen instead is 6 bucks a day for NR alone and 3 ish for NMN, plus tax and shipping. I am not sure $3650 per year for possibly useless pills is wise..
Does anyone have a suggestion on where I can get this stuff cheaply (read: about 1.50 per day for 1000 mg doses). Happy to look at Indian/Chinese sources if someone knows a reliable source.
I don't understand how internet protocols, packaging, IP addresses, routing, networks, and all of that stuff works. I find it unusually difficult to gain this knowledge from Wikipedia because it feels like every article is both too vague and related to about a dozen others. Is there a good book about this? ( I do have a background in cs and math)
"Will the US have more than 200,000 daily COVID-19 cases (7-day rolling average) before January 1, 2022?
This question resolves positively if at any point between 2021-07-01 to 2022-01-01 the 7-day rolling average of confirmed COVID-19 cases is greater than 200,000. The source will be CDC's official count of Coronavirus cases, unless Metaculus Admins determine there is a significantly superior source of data."
Result: The U.S. surpassed 200,000 new daily cases on December 24 and has stayed above that level.
"Will the US have more than 100,000 new daily COVID-19 cases before January 1, 2022?
This question resolves positively if on any single day between 2021-07-01 to 2022-01-01 there are more than 100,000 COVID-19 cases recorded. This question resolves negatively if there is no single day the United States records more than 100,000 daily COVID-19 cases according to the CDC's official count of Coronavirus cases."
Result: Milestone was passed on August 5.
"Will the US have more than 1000 daily COVID-19 deaths (7-day rolling average) before 1 January 2022?
This question resolves positively if at any point between 20 July 2021 and 1 January 2022 the 7-day rolling average of confirmed COVID-19 deaths is greater than 1000. The source will be CDC's official count of COVID-19 deaths. Make sure the "Daily Deaths" view is selected."
Result: Milestone was passed on August 18.
"Will EA Global London 2021 be cancelled, rescheduled, or moved online again?
The question resolves negatively if the EA Global London 2021 takes place in the originally scheduled physical location (London UK) at the scheduled dates (29-31 October 2021)."
Result: The conference ended up being held on time and in-person.
If anyone with covid is having trouble getting a fluvoxamine prescription, you can sign up for a 50-50 shot at it through the Activ-6 clinical trial (recommended by 'mj robinson' in another post's comment section). That trial has two arms for additional drugs but you can opt out of those. Covid out is another trial with 5 different treatment groups. Activ-6 pays participants $100 and covid out pays $400.
I've seen several references from Scott and others to evidence that the actual effects of parenting are pretty minor relative to other factors. Does anyone know a good overview of the evidence on this that they can point me to?
The main argument of The Nurture Assumption is that a child’s age peers have a stronger effect on socialization than the parents. She doesn’t say parents have no effect on their children’s life outcomes.
The book has received a lot criticism from Harris’s own peers.
> the only theory I have is that it's due to the lack of social cues
I’ve figured it’s some combination of depersonalization and that it’s hard to read text empathetically. Probably those two and what you mentioned are all strongly related.
I’d wager for most people, in person it’s easy to be amicable and listen to people’s words generously unless you’re motivated otherwise. Reading text does not do that same thing for me. I have to make a conscious effort to even remember that one block of text from commenter 1 and another by commenter 2 were even written by different people, much less remember to tie them back to specific identities with their own points of view and such.
If that’s at all common, then yeah, I figure something like what you suggest would help a lot. Even phone calls without video are much, much better at engaging the “yep, that’s another person” mode.
But now that I think about it more after writing that comment, maybe there’s something more subtle. Even on, say, a work Slack with people I’m very familiar with it’s too easy to accidentally be harsh or unfair in communication and I wonder if your social cues theory is onto something.
Depersonalization seems like a big part of it. Road rage is another example - people interacting with each other in a system where they see each other just as blank glassy windows can get *much* more heated.
I've seen studies finding that people in expensive cars are less polite to pedestrians than people in cheap cars, but I wouldn't be surprised if the effect is mostly due to whether your windows are open or closed, and tinted or clear.
I agree with Dalton, depersonalization is a big thing. I read texts from my wife in her voice and (possibly due to a long-ish history with her) I seldom get her tone wrong. I've yet to enter a heated debate with almost anyone except online commenters I don't know - rare exceptions happening in real life with close friends on religion or morals.
Perhaps this could already be lessened if there was a photo next to the name, maybe even more so if you add three values and three causes important to this person or something like that, providing context for this person's existence.
That is probably part of it, but even bigger part is that in the spoken-word world people usually don´t have conversations about hot button topics with complete strangers
I am not an American, so no. Probably every country has its out of bounds topics, but they are slightly different in each case. Social media dynamics are similar everywhere, though
Very, very fair point
I think I've seen more people tell strangers to go fuck themselves in real life than online.
Then again
A) I frequent mostly heavily moderated places like this one
B) one is usually sober when posting [citation needed]
Possibly the case that online discussion is less polite than in real life, but is that a desirable goal? My desire in participating in these discussions is to test my own thinking against others, and hear other people’s ideas. Ftf introduces a lot more social calculations which cloud that truth seeking. Also I find if I type out my arguments rather than express them verbally I am much more precise and less prone to glossing over issues or conflating things and I think discussions online makes spotting that easier in others. Of course it is unpleasant to be insulted by someone online, but if all somebody has to offer is insults it makes them easy to ignore.
Of course it is always better if people are polite, but I mean that being polite is a secondary not primary goal of online discourse. If being polite meant that discussions were meaningless because everyone was too scared of offending then it would not be a good in my view.
On whether online discussions are the cause of polarization, I think back to the pre internet days and my impression is that it was just as bad or even worse. Racism or sexism is called out very quickly in most online forums today, but it was certainly not the case in ftf discussions in say the 1980’s.
Filter bubbles could be a factor as well. I am sure plenty of people never actually talk to others who hold views that they find deeply objectionable in their normal lives but will encounter such people in digital spaces, this space is probably particularly bad for that.
You're rediscovering the reasions for the invention of the emoticon, and later emoji? We had this debate in the Usenet era. But I think the canonical hypothesis (lack of nonverbal cues) is mistaken. People can and do communicate by writing and other nonvisual means cordially, and have done so throughout history.
What I think makes the difference is the expectation of whether you will interact repeatedly with the same person, or whether future interactions are avoidable. If you are certain you will be interacting with the same person over and over again, then I think conversational styles -- face to face or not -- are always far more cautious. On the other hand, if you doubt you'll necessarily have any further interaction, they can easily become inflamed, whether they are in writing (the Internet) or even in person (cf. road rage).
So even Internet fora can be courteous places *if* almost all of the denizens interact over and over again, and there is a strong expectation from the beginning that they will, and that there is no easy way to avoid interaction with others in the same space.
Arguably, the major reason the Internet degrades conversation is that it too easily allows us to interact *just once* with others, and then never again.
Well, let me be careful here. I'm not referring to social reputation in the sense of how many "likes" you have -- *that* kind of thing, designed to solve the spam/flame problem has in many cases only made it worse, cf. the dumpster fire that digg.com became, the distortion of dating sites, et cetera. The only case that comes to mind where it kinda-sorta worked early on and never blew up was Slashdot, which used the concept of meta-moderating to keep the moderation on point.
What I mean is your *actual* long-term reputation in some other individual's head, and the chance -- nay, probability -- that you will need to deal with the consequences of that reputation in the future. It's surprising how moderating just the knowledge that you will be forced to look the other person in the eye and say hello some time in the future can be. People *really* don't want to look forward to a look of contempt.
That is, I'm saying predictably repeated long-term interaction is the missing key. In real life that's guaranteed by the historical difficulty of long-distance communication: typically *anyone* with whom you had a conversation of any length would be colocated for a while, for better or worse. You were at the same office, deployed at the same place, on the same train, living in the same neighborhood, et cetera. But now it's very different. You can meet 10,000 strangers a day, and never see or hear from them again. It's not natural, evolution hasn't provided us with good instincts for that.
You might find exploring Albert Mehrabian's findings (often summarized as the 7:38:55 rule) relevant. I can't add much more ... I'm familiar with them, but haven't gone into depth with them (the basic point seems SO obvious ... and I don't care too much about the exact numeric partitioning).
You are absolutely correct. This is why all modern ad-based social media is completely vile and only used by the ignorant.
Play some online games and discover that spoken word is no less toxic when everyone is anonymous?
Or maybe it has nothing to do with cues and everything with anonymity and lack of stable social structures where reputation could have been at stake?
I, on the other hand, wish more people would talk in person like most of us talk on the internet in text. Be more direct. Don't pull punches. Say what you mean. Sound angry if you're angry.
Anyone got recommended resources for learning about fusion power research?
(An interesting caveat to some news stories: the [energy out / energy in] ratio is called Q. But there's Q_plasma, and Q_total; the former is what's often reported, but the latter is what really matters if you want to run a power plant. [E.g. the plasma outputs more energy than the laser inputs to it, cool, but the laser itself takes power to run.]
Companies talk about Q_plasma>1 as an important milestone, and it is, but often the implication is that that would be enough to actually get energy from it, and that's false; only when Q_total>1 can you actually run a power plant that gives you more energy than it uses. https://www.youtube.com/watch?v=LJ4W1g-6JiY)
*if the plasma outputs
It's fine if you don't want to respond right away, but can you confirm you got the email I sent you a few days ago?
I did! Thanks for reminding me to get on answering!
I was curious about this same topic after seeing the same video, thanks for posting the question. Aside from the physicist in the video, I've found it hard to find non-hyped-up sources for fusion power. I'd like realistic estimates and timelines.
I've been thinking about how to do a series of blog posts about fusion. One would describe the timelines proposed by the major players and the extent to which I think they're plausible.
When I get this finished / up, I'll comment a link in an Open Thread.
That would be appreciated. As a lay-person on this subject, I'm really most interested in knowing about a few key milestones:
1. Know-how. Some team somewhere knows how to build a working fusion power plant (more energy out than in).
2. Proof. Someone has built a working fusion power plant.
3. Profit. Building, running, and maintaining a fusion power plant pays for itself.
The specific engineering details are mostly lost on me.
Everything I've seen about the current state of the field (e.g. MAST Upgrade) has been from random people who work in fusion research. I seem to remember some of them being in the comments section here .
(What's the field like, people who work in fusion research? I was thinking about applying for CCFE's mechanical grad scheme. Do engineers get to do fluidy things or is that all crazy MHD that gets left to the physicists?)
The hottest area in fusion engineering right now is probably magnet design with high temperature superconductors. Most people doing fluidy things for fusion are physicists, but the boundary between fusion science and fusion engineering isn't well defined yet. You don't really need a physics background to do fusion science: you need to know E&M and fluids, but not quantum.
Culham is great. They have had (debatably) the best fusion experiment since the 90s. I only know 2 theorists who work there, so I can't give you personal advice on who to work with. But the group as a whole is good. The startup Tokamak Energy Ltd is nearby and might also want engineers.
MHD is weird in that people who don't know it think it's super accurate & complicated, but people who do know it think it's jank. MHD is fluidy: it's fluid equations with an extra force and a few constraints. People use it for preliminary tests: if MHD shows that your plasma is unstable, it definitely is. The crazy analysis mostly involves kinetics. You might have heard of the Boltzmann equation. The Vlasov equation and gyrokinetics are the best equations to use to analyze fusion plasmas.
Ugh. That stuff is bad enough with normal fluids and short-range forces.
The past year or two have been very exciting, probably the most out of the ~10 years since I first joined the fusion field. Now's a great time to join.
Yeah, engineers may not be doing plasma-MHD, but actually there is a great need to study how fluids like liquid metals flow in the strong magnetic field environment. This is critical for heat transfer and design of the components surrounding the plasma. It's also under the banner of MHD, just in a different regime.
Fusion has a lot of challenging applications for mechanical engineering in general. Here's a sampling:
1) With high-temperature superconductors, some magnet designs can be limited by the strength of materials. The magnet assemblies need to be incredibly strong yet not take up a large area - one engineer called them "[one of] the most structural members ever".
2) The 'breeding blankets' that surround tokamaks need to withstand extreme fluxes of heat at the surface and neutrons, and operate at high temperatures: they're kind of like a heat exchanger between the neutrons and your primary working fluid. They need to be strong in order to withstand transient magnet field change events which drive currents in the walls; these currents react against the magnetic field and cause large forces. One strategy is to make cuts so that the electric current paths are much longer, but this makes the structure more flimsy. In future designs the blankets may need to be lifted out of the machine vertically to be changed; a flimsy structure makes this more difficult.
3) The machines will need to be handled by specialized (radiation-resistant) remote-handling devices. See what RACE (Remote Applications in Challenging Environments) at Culham Science Centre is doing; they're doing really impressive work.
4) In general the fusion machine environment has high temperatures, heat fluxes, neutron fluxes, and can have large mechanical and electrical loads. There are only certain classes of materials that can be used due to the neutron environment. The whole structure can be compared to building a pressure vessel out of magnets. The tolerances are very tight, sometimes 1 millimeter for a structure 20m high.
Maybe a bit high level but I recently read Arthur Turrell’s “The Star Builders”. Although my reading is he is pretty skeptical of some of the private enterprise efforts in fusion, perhaps seeing them as scams. He seems to favor the big Government led efforts as most likely leading to success, on much longer time tables than perhaps people are hoping. Note he is a former physicist researching this area but now has a role in UK Central Statistics agency and also has a website that teaches coding, interesting guy.
Some of the private enterprise efforts are not that different from scams. Lockheed Martin is big and powerful, so I'll pick on them. The only data they've published is for an experiment 1,000 times too cold for fusion - it wasn't even fully a plasma. And they say their first reactor will be small enough to fit on a truck.
In terms of the triple product (density * temperature * confinement time), the big Government fusion efforts are 10,000 times better than they were in the 1960s. While I love completely new designs and new ideas, they have a long road to catch up.
It is worth distinguishing between private enterprises that build on everything we've learned & have a plausible path to fusion and private enterprises that are little better than scams. Plausible private enterprises include Commonwealth Fusion Systems (building SPARC), Renaissance Fusion, Type One Energy, and maybe Tokamak Energy Ltd. Helion also deserves mention: I think their strategy is unlikely to work (at least in the next few decades), but they are taking the challenges seriously. Plausible private fusion really got started in 2018 after some advances in high temperature superconductors from MIT.
Commonwealth Fusion is currently leading in the race to First Fusion. Their experiment SPARC is being built over the next 4 years and should have Q ~ 10. ITER (the Government effort) should be finished slightly sooner, in 2025, but isn't planning on using tritium until 2035.
That seems very aligned with Dr Turrell’s analysis. Would not have thought Lockheed was a company likely to engage in a clear fraud, so surprised at that one. From my reading my uneducated guess is that fusion is very likely (90%) to be cracked in next 20 years and we have working commercial fusion devices that compete with fossil fuel. That is some black swan ahead of us. Would be interested in your probabilities on this.
One possibility is that Lockheed is credible, but secretive, and do have a plausible path to fusion. Another is that they know that it is implausible, but figured that they'd expose themselves to a high-risk, high-reward chance. A third possibility is that they are too specialized / don't bother distinguishing between plasma physics and combustion physics.
My probabilities vary widely depending on what "cracking fusion" means. If it means Q > 5 in some experiment, I would put it at 95% (and some of the remaining 5% is X-risk). If it means anyone is selling electricity produced by fusion to the grid in 20 years, then 90% is a good estimate. If it means that new fusion capacity is cheaper than or similar to the cost of new fossil fuel capacity, then I'd lower it to maybe 75%. There's just too much we don't know about how the economics will work, or how to scale up reactor manufacturing, or the regulatory environment for me to have much confidence.
In Taleb's zoo, this is a gray rhino rather than a black swan. Something we could see coming, but didn't bother to look at it.
I assume Lockheed is participating because they can possibly get government money flowing to keep it going, and it's not really meant to be much more than a jobs program.
Do you know if they published anything recently? I don't think they came to APS-DPP this year, though they did in the past.
I don't think that they've published anything since they last came to APS-DPP in 2019.
I have a PhD in plasma physics and am happy to answer any and all questions you have about fusion.
The best resource I know of for fusion is The Future of Fusion Energy by Parisi and Ball.
https://www.amazon.com/Future-Fusion-Energy-Popular-Science-ebook-dp-B07MYTCRNS/dp/B07MYTCRNS
For magnetically confined fusion, an optimized power plant should be able to get Q_engineering > 1 as long as Q_plasma > 5. The leading contenders for First Fusion, SPARC and ITER, both are aiming for Q_plasma = 10. They are not aiming for Q_engineering > 1 because they are designed as experiments, not as power plants. For example, ITER has 4 different ways to heat the plasma, to see which works best at these temperatures & densities, while a power plant would need at most two. Power plants would also need a lot fewer measuring devices looking at the plasma.
Hey, could you recommend a source that really focuses on the power plant side of things? For example, how did they come up with the optimized power plant numbers for magnetically confined fusion, and do these change a lot for different methods like inertial confinement or electrostatic confinement?
By way of example, the electrostatic people frequently make the claim that their plants will be able to use direct energy converters, which have much better efficiency than any of the boil-water-for-turbines options. At least intuitively, this seems like a big advantage (assuming, of course, the claims are true).
The Future of Fusion Energy has a chapter on the power plant side of things. I don't have it with me right now, so I'll respond again tomorrow after I look at their references.
A lot of the power plant is already standard boil-water-for-turbines. I am not an expert on how this works, and I don't think that there is too much innovation from the fusion community here. I have heard the idea of building fusion reactors in decommissioned nuclear (fission) power plants so we can reuse a lot of the plant.
For magnetic fusion, the main innovation in the rest of the power plant is the tritium breeding blanket. A fusion reactor will be surrounded by a layer of lithium, which absorbs the neutrons and produces tritium: Li6 + n => He4 + H3. ITER will have multiple Test Blanket Modules to see what works best.
https://www.iter.org/mach/TritiumBreeding (Including the linked articles at the bottom.)
Less work has been done on the rest of the power plant for inertial confinement fusion than magnetic confinement fusion. The main experiment, NIF, is really a weapons lab that publishes some energy-relevant work on the side.
The startup Helion has a very different electricity generation mechanism. They want to use the magnetic field generated by the expanding plasma (after the fusion has occurred) to directly induce a current. They publish less than ITER, but they have some patents that might be interesting. I think that Helion is unlikely to get First Fusion for other reasons, but their electricity recapture might be interesting.
https://www.helionenergy.com/our-technology/
Can you link some of these electrostatic people so I can see what they're talking about? The most common electrostatic confinement method is the fusor, which can be used as a neutron source, but (almost certainly) won't get net fusion energy. To get net fusion energy, you need to have the nuclei confined for long enough to collide a few hundred times. Fusors only confine the plasma for a few collision times.
A good technical summary, focused on ARC (now SPARC), is here: https://www.sciencedirect.com/science/article/abs/pii/S0920379615302337
For ITER, you can look through these pages: https://www.iter.org/mach/supporting
On power plant design: you might check out Principles of Fusion Energy by Harms, Schoepf, Miley and Kingdon. It's a bit older but it does a good job of introducing the basics at an advanced undergraduate level.
It came out recently and I have not yet read it, but there is also "Fusion Reactor Design: Plasma Physics, Fuel Cycle System, Operation and Maintenance" by Okazaki.
What do you think of the focus fusion approach (https://lppfusion.com/)? I've kept half an interested eye on it since running into one of their researchers at a conference a number of years ago and having a nice chat, but it is far enough out of my field of expertise I've never really been able to evaluate if they're just crazy or not, and I rarely see the approach brought up in wider discussions of the fusion research landscape.
I should read through their 6 relevant papers [1] before really saying anything definitive, but here is my impression from their website [2]:
I am very skeptical.
I do respect them for being open about the scale of the problem. They need to increase the energy produced by fusion by a factor of 100,000 from their current experiments. Their executive summary [3] also clearly shows that their triple product needs to increase by a factor of ~5,000.
So far, they've gotten high enough temperatures, but too low density. This is especially a problem because proton-boron fusion requires an almost 50 times higher n t (density*confinement time) than deuterium-tritium fusion.
Their strategy is to use a sequence of plasma instabilities to compress the plasma. I don't think that this is a good strategy. You have less control when most of the process is self-generated. I also think that the method will stop working as the density increases. Magnetic field-driven instabilities are likely to compress a plasma, while pressure-driven instabilities almost always expand a plasma. When the density is higher, pressure becomes more important compared to the magnetic field.
This is a pulsed reactor: the cycle of compression - fusion - expansion - electricity generation happens in sequence instead of holding the plasma in a steady state. In other pulsed fusion plans (e.g. NIF or Helion), you would need this to happen about once per second to make a power plant. NIF is currently at about once per day, but they're really a weapons lab, so they don't care as much. Helion has been trying to improve shot frequency, and are currently at about once per 10 minutes. LPP doesn't talk about shot frequency on their website, so I assume that they're not close to once per second.
[1] https://lppfusion.com/peer-reviewed-papers/
[2] In particular: https://lppfusion.com/investing-in-lppfusion/our-plan-to-net-energy/
[3] https://lppfusion.com/investing-in-lppfusion/executive-summary/
An Indispensable Truth is a good starting point if you don't have a plasma physics background. Plasma Physics and Fusion Energy or Ideal MHD by Freidberg are good if you are going into fusion or are trying to understand it more deeply.
By the way, you would likely need Q of at least 10 to make fusion economical. And it is probably more likely that you would need a steady state device and an ignited plasma (so Q = Inf) to get a massive reduction in the cost of electricity.
Happy to help answer more questions.
There's a lot of movies with an obvious formula and archetypal characters, but where the film can still be really fun and appealing if the execution and acting are charming. "Sing 2" is one of those films, and so are a lot of romantic comedies.
If we develop the technology to emulate a human mind from scanning brains, and then eventually to just create entirely original minds based off what we know about brain mapping, etc in digital form, do you think we'll try and "resurrect" people who died before the technology based off of whatever we know about them in records? Like will we try to create a plausible version of emulated Mozart based off his writings, his music, depictions of him, and what others wrote about him? I wonder how you'd test them for fidelity.
I ran into a weird anti-fusion power argument the other day. It argued that because fusion produces a ton of neutrons, you could use it and some uranium oxide to produce nuclear bomb fuel. Which . . . aside from the challenges of actually concentrating the fissile material, seems funny to me because if you've got 1)Deuterium/Heavy Water, and 2)Uranium Oxide, and 3)Really want fissile material, you could much more easily just make a heavy-water nuclear fission reactor to get it. You don't even need pressure vessels! That's actually how India got the fuel for its first nuclear bomb - they had a test heavy-water reactor for research purposes.
Speaking of fission, the World Nuclear Association estimates that the "overnight cost" of construction per kilowatt of electricity generated for nuclear plants in South Korea and China to be about $2157/kWe and $2500/kWe respectively, and it was at around $1500/kWe in the 1960s in the US. That's a lot lower than what it is now in the US, but still higher than what the US EIA estimates for "Solar Photovoltaic power plus storage" ($1612/kWe) except in the case of the 1960s US plants, and even then the difference isn't huge.
I bring that up because it makes me skeptical of the argument I've seen recently about why nuclear power hasn't come back heavily in the US, which tends to focus a lot on US regulatory changes. If even countries with far more favorable regulatory and political environments for a mass build-out of nuclear power aren't beating solar PV and storage, then it suggests there's other reasons at play - nuke plants may just be really expensive to build unless you're building a LOT of them.
http://www.world-nuclear.org/information-library/economic-aspects/economics-of-nuclear-power.aspx
https://www.eia.gov/outlooks/aeo/assumptions/pdf/table_8.2.pdf
https://en.m.wikipedia.org/wiki/A_Work_of_Art this is a pretty good science fiction story about the concept of "mind sculpting" you describe in the first part of your post.
On fusion and uranium enrichment: fusion plants will still likely require some sort of safeguards/inspection regime, but since they should normally have approximately zero uranium or plutonium around the inspection for that is much easier. It's much easier to find the presence of some radioactive isotopes when there should be none, than it is to track the flows of uranium materials to ensure that a part in a thousand is not being diverted for nefarious uses.
Yeah but on the other hand the fission numbers are real, while the massive PV + storage numbers are speculation. Cf. the massive difference between the sales-brochure predictions and deliverable reality of Ivanpah. You always want to keep your salt shaker handy when listening to investor pitches (including utilities and governments pitching you as an electricity ratepayer).
" Like will we try to create a plausible version of emulated Mozart based off his writings, his music, depictions of him, and what others wrote about him? I wonder how you'd test them for fidelity."
You could do it recursively. Emulate modern people who are known-accurate, then emulate one generation in the past and have them interact with their emulated descendants to test for accuracy. Then emulate two generations ago and test them against your verified one generation ago, repeat until you make it back to Mozart.
It wouldn't really be accurate, the system would surely end up swamped by noise, but it's a neat way to at least maintain consistency.
That's a neat trick but what you do with the simulations reported to be not accurate?
Delete them and try again. You have some idea of how they weren't accurate, and presumably there are some parameters you can tweak to make version 2 come out different.
But you can't just "delete them", as they would be people in their own right, however a-historical they might be. You wouldn't propose creating 1000 children and then "deleting" the 999 children who are least like a target historical figure after evaluating them as adults. What do you think is the relevant difference here? That the people in questions are simulations and presumably exist on computer hardware?
One of the big reasons death is bad is that people are expensive: it takes a lot of resources to make a person and that investment is lost when they die. Once we have the technology to spawn new digital people on demand that's no longer true. If we value a lot of sims existing then our only constraint is how much computing power we allocate to running simulations, a dead sim can be trivially replaced.
If any deleted sim will be replaced, what is it you value that's being lost when a sim dies?
I would say that the human rights of the sim are being violated if they are deleted without their consent, regardless of the economic value lost. So what would be lost in that case is our framework of human rights and the comfort of knowing that you yourself are unlikely to be involuntarily deleted. Perhaps you’re thinking about “deleted” in a different way than I am? What, specifically are you imagining happening?
That would amount to mass murder and defeat the whole point - which in my view was to protect the original mind from truly dying. If you do it by creating and destroying a 1000 similar-but-substantially-different minds, you didn't exactly decrease the total number of permanently dead people.
If we're troubled by the nonexistence of minds, what privileges minds that happen to have already existed? Those 1000 similar-but-different minds weren't alive at the start of our experiment, so why is it bad for them to be dead at the end of it?
For roughly the same reasons it's bad to kill biological humans, give or take. If you subscribe to some moral theory saying it's actually ok to kill people, I feel like you need to state it explicitly and argue your case, because this is a highly unorthodox view.
I'm doubtful we have enough information for people from the Mozart era, even for the most famous ones, to get anything remotely accurate. But for those who lived in the digital era, with all the photos and videos and messages and a good part of their day-to-day movements recorded, it can perhaps be sorta possible? At worst I hope this method could be helpful to augment another approach of transporting a consciousness into the future aka cryonics. I know some people do extensive life-logging because of this consideration, and considered it myself, though for now decided against it.
Consider the "Grandpa Joe" argument (https://www.youtube.com/watch?v=29_uSlEEPSk). You could interview grandpa Joe for 20 years, record every moment of his miserable life, and still not really understand how he would feel when he gets the "Golden Ticket", because he himself doesn't know, he can't imagine what it would be like to have real hope.
One of the tragedies of these predictive-model methods of resurrecting a person is that they will observe that the person is miserable at the end of their life, and then predict that they will continue being a miserable person even after they are resurrected.
Whereas I think that some of the people who are actually recreated in the future will experience a more profound sense of joy and safety than they have ever felt in their lives, and their reactions (and how they choose to live the rest of their lives) will be extremely hard to predict even given extensive knowledge of their past.
Grandpa Joe is a funny example, because you add another layer to figuring out your proposed person - deception. Grandpa Joe spends 20 years lying in bed, and can suddenly walk when there's an opportunity? Grandpa Joe hasn't gotten out of bed in 20 years, but he somehow buys Charlie a chocolate bar (and his tobacco)? Did he steal the money for that, by the way? If not, where did he get the money from?
Not to make this about a fictional character, but the reality is that Grandpa Joe has a lot of things he really doesn't want people to know about him. So he lies about some parts of his life, and keeps other parts quiet. That's going to make an emulation of him significantly harder to get right.
Ha! Good point with Grandpa Joe. I've personally always seen the story of Willy Wonka as existing on a more allegorical level than I think you're seeing it. In that interpretation the Golden Ticket has magic that "fixes" Grandpa Joe's story by altering his destiny at a narrative level. One consequence of this is that he is diegetically healed and can get up and walk again. He really had felt miserable for the last 20 years, and he really couldn't walk, and he really did experience a miracle of healing.
(though he is still is a bit of a rascal who proves himself unworthy of inheriting the chocolate factory!)
Under this interpretation I feel that it's a better analogy for how someone might feel after successfully being preserved / uploaded.
I'd say you are correct, that his sudden change in circumstances may very well have a sudden change in his behavior and even abilities. So do you model Mozart based on how he would have reacted right before getting his "golden ticket" or with the idea in mind that he may react positively?
Do you model Van Gogh as the guy who cut his ear off and later committed suicide, or do you model him as the guy who would love the 21st Century and have a different outlook on life? Would modeling him one way instead of another change something fundamental about him? Fun-loving-happy-Van-Gogh may (or may not, who knows for sure?) be accurate for the person, but does it defeat the purpose of modern people bringing him back to meet the "real" Van Gogh?
I think that if you're valuing the people from the past as ends unto themselves then you would end up just clearly explaining to them how the world has changed and help them to make of that what they will while respecting their fundamental human rights.
I imagine that you might see some very interesting groups form around shared pasts, while still being modernized in their own ways, and that that might end up being more interesting anyway than making a bunch of fake unrealistically grumpy people! For example one thing a person might do upon waking up and learning that their religion doesn't exist, after the initial shock wears off, is to get together with other people from their time and start it back up!
...it's interesting that you mention Van Gogh in particular, as this scenario has been partly explored in Dr. Who (https://www.youtube.com/watch?v=ubTJI_UphPk). To me this was a pretty powerful scene, and it felt profound and even just that Van Gogh would have a chance to be in the future.
> does it defeat the purpose of modern people bringing him back to meet the "real" Van Gogh?
IMO, If, after seeing the future, Van Gogh wouldn't want to paint again for another 100 years, that's OK. He doesn't owe anyone anything. In fact the future people have an obligation to help him flourish in his own way. If the people of the future can bring him back without an expectation that he "owes" them, simply a desire that he be, the same way they already bring children into the world, then I think that's the right place to be in to contemplate these things. I think that's what being a good steward of these souls looks like. Whatever way the people of the future treat themselves and their past, that's the precedent they're setting for even more powerful, vast, and alien people in the even farther future to deal with them.
You're talking about "predicting" reactions in a way a simple modern-day ML system does, I don't think there's much value in it. What I have in mind is that if we had an immensely better understanding of the internal workings of human's mind, and huge amount of computational resources to throw on the problem,we may be able to some extent "reverse-engineer" the mind based on external obaervations. As an analogy, imagine for example that you're trying to understand how a certain WWII-era plane worked. Not a single copy or even a blueprint survived, but you have a lot of photos and video footage, and reports, and field manuals, and interviews with the people who've flown it, and you also know a great deal of the engineering practices at the time in the company who built it. Will you be able to recreate the plane in every detail? No. Will you be able to build a plane that is functionally identical, have the same layout, same specs, but maybe some arbitrary little details are different? Maybe? With enough information and effort and some luck, I think you can get close. And if you fly this recreated plane in some conditions of which there's no records for the original, it'll likely behave the same way as the original would.
But I recognize that humans are far more complex than airplanes, so even a very detailed life logging probably won't be able to constrain the mind-design space narrowly enough to pinpoint a single mind. This is why I'm talking about using it to augment cryonics, which preserves all or most of the information in the brain, but there may be losses due to imperfect procedure or storage.
I don't think the airplane analogy applies to human minds in this case, because you can mostly figure out the different situations the airplane is expected to be in and what it's expected to do in those situations. But people experience really new situations all the time that don't look like previous situations you've observed them in, and they have a lot of hidden depths like grandpa Joe that seem to me to be beyond the ability of language to actually derive.
To make it concrete, you can ask me as many questions as you like, and look at all of my social media posts, but you still can't ever get at even a relatively simple, legible hidden layer of my mind like the password I use to unlock my computer. Good luck predicting how I will behave when I learn that my country doesn't exist, or that no one believes in my religion anymore, or that the world really is 1000x more just and safe than it was when I was previously alive. My response to these events relies on hidden layers of my mind that are even more difficult to infer from observing me than my password is!
I would change my mind if you could convince me that it's plausible to do some kind of life-logging that would be able to infer my laptop password; do you think something like that is likely in the future?
No, I definitely don't believe this level of depth can be inferred from the information that normal people store on digital devices. Although to be fair your password or at least its hash is definitely stored somewhere, and we can probably assume the future mind-simulating civilization will be able to crack modern hashes, but I totally agree with the broader point. An extreme version of life logging where you really honestly videotape your every waking hour *and* keep a diary with your thoughts can maaaybe come close, but I'm extremely doubtful and the point is moot because nobody would do that. So as I said, I don't believe it's practical as the primary means of mind preservation, but it can potentially provide necessary context or missing pieces for someone trying to repair the damage done by imperfect cryopreservation.
One abstract point where I disagree is that there's qualitative difference between this and the plane example. Both minds and planes are fundamentally physically deterministic, so if you had enough information to pinpoint the *exact* design, this design will behave, and be, fully identical to the original. (I include memories and life history in what I mean by "design" here) The difference is that you need vastly more information to pinpoint the exact design of a mind, much more than even a very digitally active person can plausibly have recorded about them.
I suspect that we perhaps don't disagree about anything then, since I also agree that with "sufficient information" you can of course recreate someone. Though I don't think that I'm attacking a strawman either, since there are actual projects such as Terasem (https://terasemmovementfoundation.com/) which work to make a "Mindfile" and eventually recreate a person through, essentially, a detailed personality quiz.
One thing that I think is key here is that language is almost certainly incapable of pinning down what a person is in enough detail to recreate them, mostly because the subconscious is relatively inaccessible to language. It's not that the details don't EXIST in some physical form, just that language is inadequate at getting at those details. It would be the equivalent of taking a photograph of a ROM chip in our metaphorical plane, but not being able to read the firmware off of the chip.
The only thing I can think of in today's world that can capture enough detail would be to preserve a person's body and physically retain all of that language-inaccessible information encoded in their brain / nervous system.
Yeah it is true that fusion produces a ton of neutrons, but so do several other devices that do not generate energy
https://en.wikipedia.org/wiki/Neutron_source
Just embarrassed myself posting in MR comments thinking it was this blog. But I'll repeat the same point: if you review Nixonland please also read Before The Storm. It sets up Perlstein's hypothesis and yeah, he hates Nixon, but he also hates LBJ and kind of has a soft spot for Goldwater.
They've been on my list for a while. Could you give a short summary of his feelings towards LBJ contra Goldwater?
Goldwater was just a weirdo. You can pretty much just copy everything about Ron Paul 2008 onto him. Would he have been a good president according to Perlstein? No, absolutely not. But he really believes in his message and is sincere.
LBJ according to Perlstein is a nervous miserable wreck, fraught with doubt about everything and massively insecure. He tries to do good, but mostly he tries to get re-elected, and will use any excuse to get closer to that goal.
Weird. Never heard anyone describe LBJ as a nervous anxious wreck. Ornery vengeful bastard, yeah. Out of his depth in the Presidency, sure. Baffled by the leftovers (both voters and staff) from Camelot, OK.
I don't know -- Caro's LBJ seemed pretty nervous or at least insecure.
Is there some kind of SSC/ACX Readers demographic survey that's been done and that someone could link me to? Would be curious to peruse such a thing in general, but am especially curious about what percentage of Scott's readers follow a (sincere, traditional) religion.
https://slatestarcodex.com/2020/01/20/ssc-survey-results-2020/
I am suitably ashamed of not getting around to doing one of those last year and intend to do it sometime in 2022.
re: Engines of Cognition, (a) ordered and looking forward to having a physical copy of some LessWrong essays, and (b) distraught that I missed out on the previous offering of A Map that Reflects the Territory — will this ever be in print again? I can’t find it in stock after a cursory search online.
Hey daltonmw! If enough people request more copies of A Map That Reflects The Territory we may do another run. Book printings are pretty expensive and a bit time consuming so we haven't done it yet, but if enough people request it we may.
Great to hear! Totally understand the circumstance, not to worry. If I could make a suggestion and it’s at all in your power to influence, perhaps adding some kind of a “Notify Me When Available” mailing list here https://www.lesswrong.com/books/2018 would be an easy way to gauge interest in another printing run. And perhaps Scott (I think he hangs out around here too, wink wink) could signal boost the idea.
+1 for these books. They look wonderful and would love to be able to have and share them.
Why not make it print on demand?
Print on demand tends to be worse quality and more expensive.
Is there a reason you can't do it as a self-published print on demand on Amazon? Doing that doesn't cost you anything and makes it easily available.
I am sad that I missed out on this! I briefly looked into trying to import a copy from Australia but I would love if there were another printing!
Are you in the Bay? I have an extra copy and can give you one if you want.
I’m in the Bay Area (SF). But obviously @daltonmw has clear first priority.
That’s very generous of you. Just sent an email.
I'm an editor in the relatively lower-stakes world of sports journalism, but the posts about legible expertise from a while back still haunt me. They stoked my existing hunches that there is no way to reconcile a publication's ceaseless pursuit of growth and attention, and its ostensible goal of producing truthful stories that also give their readers a clearer sense of the world (to say nothing of my hunch that most breaking news reporting is inherently immoral).
What I'm trying to figure out is: Are there any steps that can be taken to drive progress, however incremental, in this area? I hold no illusions about fully disentangling journalism's competing interests, but I remain hopeful that not all is yet lost.
This presupposes people want to be lied to. The only reason there's an inherent conflict is if there's significant market demand for stories that are not truthful or helping inform the reader. If that isn't true then there's no inherent conflict.
That doesn't mean misinformation wouldn't exist. It would simply mean that the cause would instead be something else. For example, that people have trouble figuring out where to place their trust. That's the case with a lot of education. It's not that there's inherently some conflict between student's desire to learn and teacher's desire to teach. But students are, definitionally, not in a position to judge the quality of education in a particular subject. And teachers have every incentive to portray themselves as maximally capable of teaching to attract money and students.
I do think there's market demand for stories with narratives (or just opinions) that inspire strong emotions in readers. They're what many people are understandably drawn to. I don't believe that means they "want" to be lied to, in the sense that there is a conscious desire to seek out misinformation.
I'm not sure I understand. Does falsehood inherently inspire stronger emotion than reality? If not, how is it relevant?
I suppose what I'm gesturing at is "falsehood" not just in terms of the words on the page, but also how a story alters one's model of the world. For example, if someone sees a lot of stories about young people dying of Covid, they could get an inflated sense of how dangerous Covid is to that demographic. The stories are accurate and they inspire strong emotions, but they may not improve overall clarity- and journalists can't help but write them.
This seems to fuzzy to test against. Aside from, "I don't like the narrative," how would you rigorously determine whether a story improves overall clarity or not? At scale?
It's true - quantifying would be incredibly challenging, perhaps impossible, and that's a big part of the problem.
I think your other comments are helping me understand your thought process a bit better, and you may be right that there isn't a conflict between truth (as you define it) and success. Instead, maybe it's a conflict of media incentives + our human inability to properly update based on new information vs. overall clarity. Good journalists' jobs aren't about giving you a story and telling you how much it should matter to you, they're about telling a story in as compelling a fashion as possible, within the bounds set by journalistic standards and reality (which gives journalists plenty of wiggle room re: story selection). And we certainly know why it's impossible to expect individuals to give new incoming information proper weight. Don't think that helps with testability, though.
It's easier to write a compelling click-worthy story if you're not constrained by the requirement that the story be true..
Exactly - journalism constantly tries to maximize reader interest, constrained by the facts. In general, the fewer constraints your maximization problem has, the more you can maximize it.
Yet if they weren't constrained by that requirement AT ALL then you'd expect the media to make things up wholesale. That rarely happens and when it does it's usually punished. Instead what happens is exaggeration or just barely plausible interpretations or laundering sources. So clearly they are constrained to it in some degree.
The question is where that constraint comes from. Since it's clearly advantageous for the news writers to be able to make things up but disadvantageous for the news readers to read something made up I think it's pretty clearly coming from the readers. The question is then... what I said at the start. To what extent is there demand for truthful stories that inform the reader.
You can imagine a world where it's relatively low and they're actually happy with the world we're in. Though if that's true, why do news outlets have such low approval ratings? Or you can imagine a world where it's some form of structural failure. For example, maybe they have no good way of vetting the news or maybe some effective cartel has been captured by interest groups or something.
I think the danger here is a few false dichotomies. "People want interesting stuff" v. "people want the truth" (they want both). "News sets out to deceive" v. "news sets out to inform" (As a general rule the journalists I know desperately want to inform and also desperately want to have enough money to keep informing and also are entirely unaware that there's a difference between facts and the conclusions they personally draw from those facts).
The biggest issue is that we have minds that crave narrative and society is now large enough that any given narrative is probably misleading. If you want to believe that kids are dumber today, a national news engine can find 6 stories about kids doing dumb stuff in an hour. There's tens of millions of kids in the U.S., you just gotta look around. But to get to the bottom of the "are kids dumber today" issue, all that is irrelevant. Instead you'd have to take all recorded instances of kids doing dumb stuff from the present, all recorded instances of kids doing dumb stuff from the past, control for a bunch of stuff and work out per-capita whether any given kid is more likely to eat a Tide Pod (or the 1980 equivalent). Such reporting would no longer be interesting and it wouldn't satisfy journalists or readers.
I think there's a kind of weakness of will in the readers. Over the long run, they want stories that tell them the truth and help them understand the world. Over the short run, they get their attention won over by stories with good narratives and the right sort of poetic justice.
It's the same problem you have with sales of "healthy" food, where people do in fact want the thing with the long-term benefit, but keep making short-term choices that subvert that, and the companies aren't incentivized to figure out the long-term benefit, but just to figure out how to keep appealing to the short-term incentives.
Inherently no, but practically yes. When there's a hot market competition for writing *anything* that elicits emotion, and when most of the things one could possibly say are false, you're inevitably going to end up with writers in a situation where they have to choose between saying one boring true story and ten false-but-exciting stories.
Another way to think about it is that writers are trying to differentiate themselves. But the truth is not differentiated. If you're trying to stand out then the truth doesn't help you; anyone can report the facts of the story and their article will look like all of the others. So where can you look to find interesting things to say that no one else has said? That's when motivated reasoning starts picking up some of the epistemic slack.
If this were true then wouldn't organizations with lower journalistic standards outcompete those with higher? Is that actually the case? I'm not sure it is. It could be but it seems like CNN outcompetes the National Enquirer.
It's difficult to compare television and print media. But if you look at the circulation of some major print journalism (https://www.statista.com/statistics/272790/circulation-of-the-biggest-daily-newspapers-in-the-us/) and you note that the National Enquirer has a print circulation of 265,000 (https://en.wikipedia.org/wiki/National_Enquirer) the picture doesn't look as hopeful. WSJ and NYT are above the National Enquirer, but USA Today, Washington Post, and everything else are below it.
Reality is a constraint. If a story about a demonstration is most moving if there are fifty thousand demonstrators but there were only twenty thousand, a reporter who insists on telling the truth will write a less moving account than one who isn't. If a one-sided bad guy mistreated good guy story is more moving than the complicated interaction that really happened, similarly.
There is a reason why fiction outsells biography.
Sure. And yet if this were infinitely true then people would willingly lap up regime propaganda which is designed to be as maximally moving without truth constraints. That is not the empirical case as far as I can tell.
There must be some reason that American news outlets do not go full Pravda. If it's because the consumers will punish them for it then the problem is not that consumers don't want truth. That doesn't mean there isn't a problem. But it does mean the idea there's an effective tension between truth telling and popularity is not necessarily true.
"This presupposes that people want to be lied to"
Or that people have limited knowledge (particularly of newly-breaking developments) and don't particularly care to do the work to check the factual accuracy of stuff they read. Or even that people think that (plausibility and/or truthiness) + (affirmation of existing priors) > (factual truth).
The question is, do people read journalism like a textbook? Or like a novel?
I believe I brought up this alternative elsewhere. There's definitely a problem but the question I was responding to is whether insufficient demand for the truth is the problem.
Responding to your general point in this thread: People want things that they think are both true and interesting. As you note, this stops most outlets from simply making up random falsehoods, you can't get far reporting that WWIII is occurring. But there is great tension between truth and popularity within the narrower range of "things that are false but can be convincingly lied about".
Give everyone $100 that they must spend on news media subscriptions of their choice. Increasing the time distance between deciding on the sort of stuff you want read and then actually reading it will cut down on the click bait.
I personally think maybe we worry too much about this issue, the concern is perhaps a transition from traditional media to the new way of consuming ideas via non mediated sources. Eventually we will simply accept traditional media as untrustworthy as the rest and let reputation drive which data and sources we trust. This will be harder for older people, for whom when a prestigious newspaper prints something it means it should be paid special attention rather than a small update in ones priors.
Sometimes I wonder this too. But I'm not convinced. There seems to be a fundamental variation in the ability for untruth to persist across different domains. For example, in physics, a meme like flat earth cannot persist in perpetuity (outside of the fringe). At one point in human history, it was maybe the dominant belief. Now it's extremely fringe, because we can put a camera on top of a rocket and we derive concrete benefit from being correct about the nature of the solar system.
But on various social and political issues it appears there can be no true reconciliation. "Does Donald Trump respect democracy?" For me, Jan 6th was a completely satisfactory answer to that question. But did all the people who claimed otherwise see a reputational hit by Jan 7th? Maybe to some degree, but in general no: those people have as high of a reputation among their audience as ever.
I think COVID gives us a good bound on "how much feedback is necessary for social consensus to converge?" More than a ~1% death rate (or whatever it is).
The conventional wisdom is that this is where competition and The Invisible Hand work their magic. In principle a publication in savage competition with many other publications will have a hard time bullshitting their readers with enticing but false stories, because their competitors will take them down with glee. Everyone wants to print rosy lies, let us say, or is at least driven in that direction by the hunger for growth, but when truth is the rosiest story that *can't* be destroyed by a vicious competitor, that's the market-clearing product.
Of course, there are two obvious caveats to this, even if you believe it in general: (1) you have to *have* competition, so you can't be living in a world where a few huge organizations control most publication in the area, and (2) it probably doesn't work for those stories in which *everyone* wants to be lied to, e.g. stories in which human mortality is a central point.
I'm a traditional horary astrologer who enjoys practicing his art for strangers on the internet. If any of you have a question on just about anything you'd like examined via the toolkit of real astrology, drop me a line at FlexOnMaterialists@protonmail.com. Materialists and even atheists are, despite the email address, quite welcome. I look forward to hearing from you!
I'm going to post my predictions for 2022 soon, and if history is any guide then Zvi and the prediction markets will try to beat me on the global ones. Others will be about my personal life. If you want, I can give you my birth date and any other information you need, and you can try to beat the markets on either category.
A fine idea. There are some techniques which use progressions based on the natal chart to determine personal experiences in a given year of life; I'd need your birth date, birth time (or reasonable approximation), and birth city to check out your nativity, but that's only likely to be relevant for your personal life predictions. For everything else, I'll definitely watch for whatever might be amenable to horary prediction and take my shot along with the rest.
Huge fan of your work since the SSC days btw. I'll try not to gush too much in any personal correspondence.
November 7, 1984, 11:35 AM, in Santa Ana, California. I'd also be interested in you telling me what experiences I might have this year before I make predictions, so that it's unprompted.
That's funny, you don't *sound* like a Scorpio! 😁
EDIT: People on here have often asked how Scott is so good at writing. The answer is simple: it's due to his Mercury (planet of thinking, communication, intellect) in the Tenth House (house of Ambitions. Motivations. Career, achievements. Society and government. Father or authority. Notoriety. Advantage):
Mercury in the 10th House
Mercury in the tenth house tells us about achieving personal goals and ambitions. These people express themselves well both orally and in writing, and they can use this ability to achieve success and recognition in their work. Their job is important to them and their career is carefully and thoroughly planned, including education and preparation necessary to achieve their goals. Mercury brings the ability to present ideas and opinions to the public, and it often generates a talent for making great speeches, publishing or political strategy.
Er, Scott's natal Mercury is in the 11th house, as well as being in detriment in Sagittarius.
Ah well, that was quick'n'dirty using online chart creation, and the website must be using a different system to yours.
I *am* surprised Scott is a Scorpio, given the general (often overblown) reputation of natives of that sign. Must be a lot of influence from his Rising Sign! 😉
(My natal sign, as you can guess, is Gemini but I have a *ton* of Taurean influence in my chart and it really shows).
FWLIW @Deiseach, I looked up the astrological descriptions of myself and folks who’re important to me, and in many cases the descriptions were frighteningly accurate, with one important exception:
My wife.
“That’s not her at all!” I thought.
But then I looked up our “compatibility”, and: “Oh jeez, that is us!”
Memorably, near the end of the description was:
“Is this an ideal couple? No way; in fact, it's a dumpster fire”
And on that: thank you again for your advice, previously I’ve mostly heard either (stay with your wife, divorce will cost you too much money and the kids, visit whores if you have to” (usually from men) or “She’s a monster! Follow your heart, leave her and be with your true love instead” (usually from women), and while you made some wrong guesses about the history I don’t think that altered the core of what I needed to learn from you, and because of your uniquely unclouded atypical viewpoint it was you I wanted to learn from.
One thing I learned from the disparate advice from others is how much most women really hate housewives, most men regard them as doing a job, most women don’t, the opposite of what I’d expect.
I'm glad the advice wasn't completely rubbish, but I do think that what you need to do first and foremost is sort out what *you* want, and then decide about (a) your marriage (b) any new relationships.
I'm not thrilled by divorce, but sometimes ending a bad marriage is best for all parties concerned. Don't rush into a new, rebound romance, though.
Cheers; I'll try to have this done by the next open thread.
Can I ask what do you think of Indian astrology (use precessed equinoxes so overcoming the objections in articles like this one: https://www.livescience.com/4667-astrological-sign.html , e.g. under traditional Western astrology I'm Gemini but in Indian I'd be Taurus) versus Western astrology (derived from/based on Babylonian, Greek and Egyptian influences)?
Given that mediaeval erections of horoscopes use the same set-up as Indian kundlis, rather than the circular chart layout of modern astrology, did older astrology fall more into line with such practices in your opinion?
(1) Indian kundli
https://i.pinimg.com/originals/1c/b3/47/1cb3478f5c918d4123066d3e20ad2d8b.jpg
(2) Mediaeval horoscope
http://astrolabesandstuff.blogspot.com/2015/06/drawing-up-medieval-horoscope.html
(3) Modern horoscope
https://upload.wikimedia.org/wikipedia/commons/thumb/4/42/Astrological_Chart_-_New_Millennium.JPG/640px-Astrological_Chart_-_New_Millennium.JPG
I haven't studied Vedic astrology myself; what I've gleaned incidentally makes me think it's an excellent system when approached with determination and skill. I've also heard of a few instances where Western and Vedic astrologers examine the same query and get largely similar results.
Re: square vs circular charts, the data is the same, only the shape is different. Medieval Arabs were using both more or less interchangeably back when giants in the tradition like Mashallah and Sahl ibn Bishr were actively practicing.
The precession of the equinoxes is often used to attack modern astrology, but was actually known to astrologers > 2000 years ago. For example, the Greek Hipparchus circa 129 bce, and 2nd century Ptolemy 'the Alexandrian' wrote about it in his Tetrabiblos. Maybe the Chaldeans before then?
What exactly is a materialist in your view?
A person who believes in the sole existence of matter and energy. Your comment made me realize there's at least a hypothetical category of "agnostic"(?) materialists--someone who doesn't know whether or not the physical world of matter and energy is all that exists, though I expect this is fairly rare.
I prefer "physicalist" because you don't make this kind of very specific commitment. There's space, time, fields...
Why do you think agnostic materialism is rare? I'm having difficulty seeing how it is not the default position.
Likely the same reason that agnosticism is rare compared to theism/atheism: it's exceptionally difficult to wait for evidence or keep your thoughts cautiously neutral in the face of such a difficult and fraught proposition.
Don't we have to be a little careful here? On almost all factual data, almost of which is by definition unmeasured, I think people are perfectly comfortable with agnosticism. Who will win the Superbowl 150 years from now? Did Abraham Lincoln have a comfortable crap the morning he was assassinated? Did your or my great-grandparents prefer to do it doggy or missionary? Et cetera. About these and an infinity of other factual questions, most people are quite happy to shrug their shoulders and say Who Knows?
Not on profoundly important existential questions, no, but arguably the problem here is not that they are disturbing per se but that you have to *move forward* if you are not a stylite philosopher who can spend his days ruminating over the nature of the infinite. You have to at least act as if you have satisfactory answers to questions of Is It All Worth It? and What Gives Life Meaning? so you might as well just pick a functional answer on grounds of taste or convenience and get on with life.
That isn't especially cowardly, it's just practical.
You said it much better than I could.
Hmm... would I still be a materialist if I believed in the number 3? Like, I don't think you could go out and punch a 3, but I still think math is real.
What about linguistic prescriptivism? Or objective moral truth?
Apart from that, your definition seems like a good one. My only real caveat is what if some new thing is discovered that isn't matter or energy, but is still described by simple mathematical constructs with no privileged reference frame?
I don't know if there is such a third thing or not. But I would hardly call myself agnostic about my materialism, other than the fact that 0 and 1 aren't real probabilities.
IMO the materialist would say abstractions like 3 only exist as signals in the brain; you could tie yourself in knots trying to reason out whether or not 3 existed before some mind counted that high, but ultimately the expectation is that 3 does not have meaning outside of a thinker thinking. The non-materialist would say 3 has an objective existence which transcends the cosmos of matter and energy.
On a related note, I know several groups who buy (or would buy) successful predictive power. These range from actuarial to stock picking to whatever. If you can show any consistent predictive power in any of the events they're interested in then they'll be happy to pay you large sums of money. If you're not money motivated then they'd be perfectly happy to donate it to charity or something. And it would absolutely convert a lot of people who think astrology is BS.
And on the flip side, if you refuse to engage your "talents" with people who have strong incentives to rigorously test your predictions, that strengthens the priors of people who think astrology is BS.
Both excellent points. All I can say right now is that I'm working on it.
I, personally, would think less of an astrologer who was motivated primarily by financial gain.
In what regard? Do you think less of, say, chemists, if they apply their knowledge and skills for a paycheck?
Given that I assume chemists are real and astrologers are charlatans (reasonable or not, I typically assume that):
If a chemist is getting paid a lot, I infer that he's very good at chemistry. If an astrologer is getting paid a lot, I infer that he's a particularly craven and insidious liar.
These aren't necessarily reasonable inference in all cases. The point is merely that getting paid won't always overcome the bias against astrologers.
Financially motivated chemists are the worst kind! \s
I recently watched this video (https://www.youtube.com/watch?v=rvLFEh7V18A), which explains the recent retro video game bubble, where the top auction price of old, high quality video games jumped from $30K to $1.6MM over the past 4 years. According to the video, a group of connected people (including a specific auction house) have been selling retro games between one another at increasingly higher and higher prices. These circular sales give the impression to the general public that the value of retro video games is rising very fast, which creates extra hype, which causes more people to flood the space and buy these retro games for higher prices, with the hope they'll be able to sell the games for an even higher price in the coming months/years. Since retro video games don't have much intrinsic value (games can be played for free on an emulator or bought cheaply on a virtual console), the price they sell for at auction is mostly a function of how much hype exists around them, and how likely buyers think it is they will be able to flip the game for a higher price in the future.
Do any of you think this practice could be even more widespread?
The video says something like this may have happened in the comic book collecting space. Additional places where this practice seem possible to me include sales of fancy art, sales of NFTs, and potentially the prices of some crypto coins. Sales in all of these places (especially in the crypto space) can be very difficult to trace, making circular sales just to bid up the price very doable. In addition, valuations in all of these places (especially in the crypto space) are incredibly dependent on hype and the belief that prices will continue to go up. While there are some safeguards in place (e.g. auction houses have some controls on who buys and sells products there, and some crypto exchanges have identity verification), it doesn't seem like they would be enough. In addition, such circular bidding does not need to go on forever - it's just needed to get some initial hype - afterwards you can let the market do the rest.
all of the classic methods of fraud in that video have been heavily used in all notable NFT markets, alongside many others that crypto makes easier, yes
Interesting, I’d love to read more about this type of fraud with NFTs. Do you have any links/news articles I can look into?
This is a variant of wash trading, a very common, and where regulated illegal, market manipulation tactic. I expect it to be all over the place in unregulated markets. As you say, it has two advantages for the manipulator: generating fake trading volume and a fake price. Products which young people trade are more sensitive to start hyping from those initial impulses because there are not many data points in the past 5 years of hypes crashing hard enough to painfully punish early investors. It's a problem.
You don't even need to collude directly if the initial majority manages to cooperate on the prisoner's dillemma by playing playing bid/bid.
Sneakers, comic books, trading cards, crypto currencies, NFTs, in-game items. In a more abstract variant also the hype-stocks like GameStop, AMC. Also Tesla, Amazon, etc.
It's wrong but hard to bet against. Most trades I can come up with have very high variance relative to the return they should eventually provide. "The market can remain irrational longer than you can remain solvent." The long long-run should see these prices eventually correct, but it might easily be a matter of several years if not more.
Question for anyone that hires experienced software developers. I am trying to demonstrate soft skills in front end web development. This is not always easy to do in a short interview. Would you be convinced by a series of blog posts on a candidate's personal website?
For what country? I can only answer for UK hiring.
What are "soft skills in front end web development"? Do you mean having artistic sense? Incidentally if you could write a blog post on how to have good sense that would be nice, there's not much in that space that I'm aware of: even the lowest level stuff like "every layout" gives you a series of components to build with and tells you why they work, instead of telling you how to figure out whether something looks good (without showing it to your friends) or design something that looks good.
Blog posts affect my decision if they're good, in the same way that GitHub repositories might affect my decision if they're good. A generic React tutorial is neither good nor bad, a bad post is negative, a good post is positive.
I would expect an interview to be more valuable for soft skills than blog posts would. No matter how much you've written about communication, the interview will tell me how good you are at communicating.
I also expect most people interviewing to not read your blog posts, especially if they're long.
US based.
Blog posts will be 1-5k words.
Soft skill topics would include:
1. Asking better questions from other devs in the same organization.
2. An easy way to be a better team member (taking good notes).
3. Common ways that designers break web standards, and how to have a conversation about it.
I don't know if most interviewers are even looking for good communication skills above a bare minimum. I don't think my technical skills and experience make me stand out very much. So part of my strategy would be to try and sell interviewers on the importance of these other skills.
Your strategy expressed here requires three things:
1. Write good blog posts about soft skills
2. Convince interviewers that good soft skills can make up for technical deficiencies
3. Convince interviewers that you personally have good soft skills, using the posts written in 1.
I think this is going to be quite difficult. When I think about the people I knew who I would have hired on the strength of their communication, they did a lot of things -- for example, they were easy to talk to and communicated information well. If this isn't coming across in the interview, it's possible you don't actually have standout soft skills.
For the given topics, I see the first two as neutral, and the third as a potential negative. Which do you care about more: delivering software that gives value to users (and the business), or meeting web standards? Are you going to be overly picky? There are definitely cases where accessibility (to give an example) is worth the time (and money), but also cases where it isn't. You could talk about this in the post, but any unusual opinions can sink your chances.
When I think soft skills, I think more broadly: Dale Carnegie's "How to Win Friends and Influence People", Robert Cialdini's "Influence", Communication (and feedback) as one of the four pillars of Kent Beck's Extreme Programming. What do you think about pair programming? Code review? Mentoring? Office politics? How much / how little? What's your experience in them?
If you think your technical skills are holding you back, I'd also expect you to be working on improving them. Trying to sidle around to focus on soft skills is not going to be terribly convincing, I'd expect.
Good advice all around. I disagree on the point about web standards and it is part of why I want to write a post. Web standards exist for good reasons and breaking them is often bad for everyone involved. It makes your product less useable for customers, it makes bugs more likely and increases the cost of maintenance, and it often requires writing custom software. So my argument would be it's worse for customers, and more expensive.
And accessibility is also something where I disagree. Usually meeting accessibility standards means your code is legible to screen readers, which has carry over benefits because it makes it more readable for machines in general, which translates to better SEO, easier automated testing, and easier web analytics.
I look forward to seeing your argument in full, preferably with some examples. That's definitely worth a post.
I agree it often makes products less usable. I don't think it necessarily makes bugs more or less likely, or increases or decreases the cost of maintenance. I don't understand the custom software point, though I'd agree if you're writing custom software to do something that can be done natively that's often a bad idea.
I think it's important to consider the audience for the tool you're building. Look at the browsers they use, look at whether they're using screen readers or not (as in your example). If it's an internal tool and nobody involved uses a screen reader, spending two weeks making it screen reader friendly is time you could have spent solving problems for other teams. If somebody arrives who is colour-blind, or has vision problems, that's the time to start focusing on improving that area -- but not before. YAGNI.
In short: improving things takes development time. This is the tradeoff: you can't code everything. You have to prioritise. And if you choose to prioritise standards, you still have to maintain that code you wrote. So it should be delivering benefits worthy of the time you spent writing it and the time you'll spend maintaining it in the future. And maybe it will, but maybe it won't.
If you want better SEO, focus on SEO. If you want better automated testing, focus on automated testing. If you want better web analytics, focus on web analytics. Improving accessibility and saying it offers all this extra stuff -- well great, but are any of them priorities? If they are, they should be their own focus. If not, it's half price on something you didn't want in the first place.
I do look forward to your post though :). My perspective is software consulting, which has a limited budget (client pays for time) and so work has to pay for itself internally to a much greater extent than outside. If you have unlimited time and budget, all this stuff is great to work on. If you don't, tradeoffs need to be made.
I agree with the points about tradeoffs. I have more often worked at companies that are building a web platform that they will have to support indefinitely.
One example of breaking web standards is how modals are sometimes used. I've had designers that wanted to use full page modals, which is basically creating a page where standard web navigation is broken. it is a surprisingly common anti-pattern (I think the reddit redesign had full page modals for viewing comments on a post, not sure if they still have it.)
The mock-ups for these full page modals often look pretty and clean, but they are crap for user experience. And once product owners realize that web navigation is broken they usually want it back. Which leads to the front end dev having to add in a bunch of standard web navigation functionality.
I've never met a product owner that has been happy with the full amount of time and effort to implement full page modals, but if no one knows where things are headed then each individual step along the path of failure seems worth it.
Interviewers likely have a busy schedule. I expect most would click around and glance at your blog posts for a few minutes, then make a snap decision on whether they like your style or not, rather than reading the full thing. Doing anything out of the ordinary is normally a net positive, but given how short the time to impress is, treat it like your CV: perfectly styled, thematically curated and organized to bring across the points (about yourself) that you want.
I'm a FE dev for a US startup, but I don't do interviews. I would say that our interview process already evaluates candidates' soft skills, and our interviewers wouldn't have time to read blog posts.
For what it's worth, I've reduced a candidate's ranking based on Internet writing -- yes, I google everybody, naturally -- but never elevated it. I wouldn't say it's worth the risk. On the down side, if there's *anything* there to which exception can be taken, you get questions raised about your judgment in signing your name to it publically. There isn't much of an upside, as far as I can see, because your ability to craft a piece of human or humane sounding writing given plenty of time, solitude, and unknown assistance is not especially relevant to the question of how well you would interact with colleagues and management in the moment, day to day.
Maybe it's old-fashioned in this day of building your own brand, but I assess "people" skills of candidates by looking them in the eye, talking face-to-face about this and that, and seeing how they react to even the minor twists and turns of an interview process. Are they gracious, polite, considerate, neither easily offended nor a doormat? Big plus. Are they brittle, self-centered, needy, unaware, or see the job as some aspect of a personal crusade? Next!
I am a U.S. based software engineer who has done many interviews mostly for backend candidates. I would say there would be a 50% chance I would see your blog and spend more the 5 minutes on it. If anything I may ask you one question about your blog.
For me one thing that makes a candidate stand out for some of the soft skills you mentioned is if they are able to teach me something in the interview. It definitely hits point 3, and potentially shows you can ask the right questions to learn a topic. In my experience this doesn't happen often, but a blog can potentially increase your probability.
Short answer: No, I wouldn't be convinced of much by blog posts.
Long answer:
If the folks doing the hiring care about soft skills (and they should), then their interview process should be designed to get those signals regardless of whether you have any public writing available.
Most software interview processes at most companies should not depend on your doing "work" outside of work — in particular, I would not support any interview process that required you to have a presence on GitHub. There are some good exceptions, e.g. open source focused roles. Likewise, for a developer relations role, perhaps it is reasonable to expect candidates to have some public writing.
A few ways a software interview process can probe soft skills:
* how does the candidate respond to feedback in a coding interview?
* have a written (code, text, or both) take-home portion of the interview; can also add a feedback cycle or two
* questions about how the candidate has handled interpersonal scenarios, e.g. gathering requirements for an ambiguous project or dealing with a difficult teammate
* etc.
All interview loops I've had a hand in designing have had some of this. Typically at all stages from intro call through final round. And the feedback from candidates has been overwhelmingly positive when these things are included.
Recommendation: don't go to Walgreens for your COVID booster. I did this last week, showed up right on time for my appointment, and they made me wait an hour and a half on an indoor non-socially-distanced line full of coughing people asking about COVID tests (miraculously, I don't seem to have gotten COVID). I'm told that non-Walgreens pharmacies are less bad than this, though I can't personally confirm.
I went to a Walgreens for my booster and didn't have to wait in a long line or be exposed to lots of sick people. But they *also* refused to update my vaccine card with proof of the booster, so I'll have to order a new one from the CDC or whoever does this, assuming I'm even in the database at all now. Foolish inconveniences!
The CDC refers people back to where they got vaccinated (and/or the state health department) for the similar issue of replacement vaccine cards.
If the shot was properly recorded, you may be able to get a digital vaccine record you can store on your smartphone as an alternative from your state (depending on the state) or your primary care health provider (if you have one and it does digital vax cards).
(My state didn't have any record of my second shot, and when I was finally able to get through to the health department half a year later, it referred me back to the place I was vaccinated. To its credit, the provider was actually very quick to correct it once I contacted them through their company website. But if Walgreens is having issues, that may not work as well for you.)
Huh. I have a COVID booster appointment at Walgreens tomorrow, since they were the only place I could find with open slots. They also had a great signup system, where I could search appointments at all area pharmacies at once. I'll let you know how it goes!
Update: Pretty much everything was great. There was only about ten minutes' wait. They wrote the booster dose on the back of my old vaccine card, but there really wasn't room on the front (blame the people from last spring.) Two people did come up asking about at-home COVID tests, but neither of them was coughing, and they left within a minute.
And, I don't have any side effects aside from a sore arm and feeling slightly more tired than normal.
Ah Scott, if only it were so simple as avoiding fucking Walgreens.
I went to Walgreens for my COVID booster because they were the only pharmacy in town with Moderna boosters. When the pharmacist heard I had already gotten two shots of J&J (I was in the two-shot phase 1 trial before they even planned the one-shot regime), he refused to give me a booster, claiming I wasn't eligible, even though it had been over a year since my second shot. (He told me I could get a doctor's prescription for one, but that seemed like too much of a hassle). Two weeks later, I end up catching COVID. I called the pharmacist to rant at him and he had zero regrets about following procedure. Even more reason to avoid Walgreens.
"Isn't eligible" doesn't mean "zero chance of getting Covid", so the pharmacist had no reason to have any regrets.
I think someone can legitimately regret having caused a harm, even if they did so completely according to the rules they were operating under.
We had an awesome experience at Walmart 2 weeks ago. Random walk in with no appointment flu shots, could have done the same for Moderna booster. They were taking Phizer appointments one week out, which we declined. 30m after we left they called and asked if we could return to fill two cancellations.
For what it's worth, I'm pretty sure the shot I got from Walgreens was a dud, either due to cold chain issues or because the pharmacy tech didn't stab it in far enough. No reaction. I am now an immunocompromised person in a risky work situation. Next dose is coming from a hospital if I can manage it.
FWIW I would not assume your booter was ineffective just because it did not provoke a reaction: I had no reaction at all to a booster of Pfizer after getting AZ originally (having had a very acute fever after the first dose of AZ but almost no reaction to the second).
I had zero reaction to a first dose. It was slightly creepy. Then I had a mild but definitively existing reaction to second one. From some conversations I understand that many people had no reaction, so that is normal
For what it's worth, I had all three of my shots from my local Walgreens, and the procedure was smooth and uneventful on all three occasions. Same for my mother.
About 9000 Walgreens locations in the US. You could have found a clinker. My booster at another Walgreens went smoothly. I scheduled an earlier AM appointment and was in and out in 15 minutes.
I’ve run into this situation at the same location though:
https://www.theonion.com/walgreens-customer-really-pushing-it-with-amount-of-non-1848066899
At CVS, I was in and out in 10 minutes. Maybe it depends a lot on the location?
They *made* you stand in line? As in, you were manacled to a post or something? If not, I wouldn't assign Walgreens more than 50% of the liability if I were on the jury.
I realize I’m being a bit cranky here, but — this comment seems to me to be an instance of a low-value kind of gotcha: 1) Break off a piece of somebody’s remark. Out of context that piece sounds dumb. (2) Point out that it sounds dumb.
There’s lots and lots of that kind of thing online, and more on ACX than there used to be. Jeez.
You, I and everybody else know that Scott was not claiming that he was forced to stay in the line in the way that somebody manacled to a post would have been forced. He meant that in order to get his scheduled Walgreens shot without waiting in line he would have had to protest the arrangement in some way that was conspicuous and unpleasant and not very likely to result in his getting his shot.
If you think he should have chosen the alternative of making some sort of public protest, go ahead and make the case for that. But the gotcha you posted is just silly.
I disagree. I'm pointing out that Scott's actions, read as revealed preference, demonstrate that *his* concern for the possibility of catching COVID from the line was low. If he thought it was a serious risk, worth at least the very mild trouble of coming back another day, or going to another place, he would not have stayed.
Therefore in his own estimation, the risk was low. Given that, why should Walgreens assign a higher importance to the risk than Scott himself? Which suggests the issue is entirely cosmetic, like a ritual "thank you" to the cashier or something. Cosmetic niceties are...well, nice, but they are not in the same category as actual carelessness with public health, which is what the implication of the comment was, I thought.
In short, I had a serious point to make. I regret it was not more obvious.
Now that you've explained what you had in mind, I agree that your comment was not an instance of Low-value Gotcha. Sorry if I kind of jumped on you.
Not at all, you gave me a valuable chance to explain what was almost certainly not well put in the first place.
"Recommendation: don't go to Walgreens for your COVID booster. I did this last week, showed up right on time for my appointment..."
Ummm .... 'n' of 1?
Did you go to the Berkeley Walgreens that looks like it’s about to shut down: broken front gate, many empty shelves, multilingual “no shoplifting” signs everywhere? My girlfriend did; her experience was very similar, but with fewer people asking about tests, as it was pre-omicron.
Behold the power of Astral Codex Ten! From my inbox, just now:
Walgreens <Walgreens@e.walgreens.com> Tue, Jan 4, 2022 at 5:05 PM
Reply-To: Walgreens <support-bf2tj8dbftkvmuaueyqpjbwjugeyth@e.walgreens.com>
To: n-g@xxxxxx.xxx
Walgreens
We hear you.
And we're here for you.
It's been a very tough few years. As the pandemic continues to evolve, we're doing everything we can to keep everyone protected.
To date, our teams have provided over 55 million COVID-19 vaccines, more than 23 million COVID-19 tests and have filled over a billion prescriptions filled annually. We're incredibly proud of our team members and their unwavering dedication to your health and safety.
Now more than ever, we owe it to you to recognize the challenges we as a company have faced during the pandemic, including longer lines, out-of-stock items and delayed appointments.
We heard your feedback and we're working tirelessly to be better. We're implementing speedier appointment scheduling and hiring thousands of new team members to help make visits faster. You can also save time and trips by using our digital tools online and on our app.
We care deeply about every one of you. Your well-being is our top priority, and we're committed to delivering the exceptional wellness experience you deserve.
We're in this together, and because of your feedback and understanding, we're confident we'll all come out stronger.
We wish you a very happy and healthy 2022.
John Standley, Walgreens President Tracey D. Brown, President of Retail Products and Chief Customer Officer
Oh, and I like how "longer lines, out-of-stock items and delayed appointments" were problems the company faced rather than problems its customers faced.
Like many, I've been "George-Pilled" by the recent series of essays. I'd like to get the other aide of the story. Can anyone link to some good critiques of Georgism?
The same!
It seems like the most intuitive critique is that homeowners are a pretty strong political block to most places (read: the US), and would object to their rent pricing model changing significantly, (perhaps wisely) and would not believe that a UBI dividend would be how government would actually choose to spend that tax revenue. I'm not sure I disagree with them, even despite that I found all the Georgist arguments presented quite convincing.
This doesn't seem to be a critique of Georgism so much as an explanation for why it has never been taken up.
Impracticality seems like a critique to me.
More generally, I suspect it's the strongest criticism, and why so few attractive policies are implemented.
If a policy is attractive theoretically but difficult to implement, either because it is unpopular with stakeholders (homeowners, in this case), or else a radical departure from norms, even if those norms are bad and the stakeholders suck, the policy seems unlikely.
It may or may not meet the "good" threshold, but it is a critique. Personally I found their specific arguments unconvincing, but I can definitely conceive of others in this (very) general category that fare better.
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1999105
For instance, the person who buys a mountaintop in the middle of nowhere to open a ski resort gets hit progressively harder and harder as the resort town builds up around it. Compounding the problem (but in another category entirely) is that this mountaintop property is qualitatively different from the surrounding town; suitable for different uses from the rest of the area, making neighborhood comparison-based valuation impossible.
The comment sections of those articles seemed to be filled with a lot of criticism, but it was a fair amount of work to distinguish thoughtful economic critique from people who have a lot of equity in their homes.
It's my impression that most economists think it'd be a great idea but almost completely impossible to implement because of political considerations. The central problem is that like 60-80% of this country believes, very strongly, that owning their own home is this incredible opportunity for building wealth. Even people that don't yet own a home tend to be saving up for one. Since, for most of these people, these investments would represent the majority of their net worth, they are extremely averse to suggestions of a different arrangement. Many of these people created a retirement plan decades ago predicated on paying off their mortgage.
In general, our addiction to land-as-an-investment is very strong. I think breaking that addiction is just incredibly hard, politically. Those are the problems that Georgists have to contend with.
"The central problem is that like 60-80% of this country believes, very strongly, that owning their own home is this incredible opportunity for building wealth. "
The more charitable conclusion is that your method of calculating the value of a home fails to capture important aspects that actual people value.
I don't believe I proposed a method for calculating the value of a home. Care to elaborate?
In order for you to state with disapproval "The central problem is that like 60-80% of this country believes, very strongly, that owning their own home is this incredible opportunity for building wealth." you implicitly claim that the value of a home is low enough that you can disapprove.
You may fail to state a method, but you implicitly have one.
I wouldn't say I'm passing any judgement on any home-valuing methods. I think you are misunderstanding my comment in a defensive manner.
If I have disapproval, it's for the suboptimal equilibrium we find ourselves in with our system of taxation. I can disapprove of the system while respecting that every actor within the system is behaving rationally. I didn't say that people are wrong to consider home ownership to be a good investment. Under the current paradigm, it is a good investment.
But I do think that in the long run this is bad for society, and it harms us all economically, disproportionately the future generations. One way to look at it is that housing cannot simultaneously be a good investment and be affordable. Think about it: anything that's a good investment will be bid up to the maximum price the market can bear.
It's not like a Georgist LVT would make wealth building or home ownership any harder. The stock and bond markets would be fine (they would actually be stronger) and homes would cost the values of the houses built upon it. You could still make improvements to a home to increase its resale value. Teardowns would become much less economical and frequent.
You can think this is a bad idea, but saying that "people value homes more than you think they do" does not demonstrate a serious critique.
"I didn't say that people are wrong to consider home ownership to be a good investment."
You implied it. Stating "the problem is that people believe X" implies that you think X is false.
Furthermore, you're confusing "is a good investment" and "is a good value". People don't believe that homes are a good investment, in the sense of providing more money when they are sold than if they had bought something else. People rather believe that homes provide good value for the money put into them, which is different.
I wrote a lot of critiques in those threads, from having been very supportive of the idea when initially posted in the book review contest.
I think Georgism works pretty well in dense cities where most people rent. The more the population leans towards owning their own home, and the more rural the scenario, the less Georgism makes sense. Because so much of the [U.S.] country has made decisions based on home ownership, the transition will be extremely difficult - a fact most Georgists acknowledge. That's not people being needlessly selfish or just holding onto outdated ideas, it's a real problem where they've sunk a lot of their financial wealth into property ownership. A swift move to Georgism would destroy the wealth of most people in the country. A UBI would help, but only by making everyone dependent on government. Taking a lot of people who feel independently comfortable in their livelihood, stripping them of most of their accumulated wealth, and then making them dependent on a government stipend seems like a loser of a deal.
Trying it out in some cities that have very high rates of renting would be a good place to look. If that's not feasible, then there's no way it could be feasible across the country.
Georgism privileges skilled labor and productive enterprise over getting there first and paying your dues.
We really don’t like the guy who gets paid a lot of money because he can write code or trade derivatives. We think it’s wrong that he gets paid so much, and proper that he gets cut down to size by heavy taxation.
Meanwhile we really do like and respect and want to support the family that put down roots and bought a house in a community that turned out to be successful.
Georgism would transfer tax burden from the former to the latter, and for that reason would bring public policy further out of alignment with the public’s values.
I don't have a link, but I remember reading an interesting counter argument in the comments. Basically it was that if I build and own an entire city on my own, I won't get taxed much for the land, because all the extra value is an internality, but if a million people build and own that city, they'll all get taxed on the externalities they get from each other, which seems unfair and like it would distort markets toward concentration of land ownership.
Interesting. But how does it scale?
10 000 landowners are surely enough for a city to have almost the same landprices as 100 000.
It seems like only very active planning can ensure the concentration of ownership that is necessary for the land in a city to become significantly lower taxed.
I don't think that much planning is needed. It can start with just a gated community or a shopping complex or whatever, and then the incentives should encourage any land owners between whom there are large positive externalities to merge their holdings in order not to get taxed for those externalities, until the whole city has the same owner.
Could anyone with life sciences/nootropics background comment on nicotinamide adenine dinucleotide (NAD+) as a supplement? This video has some impressive claims with papers backed up by reputable journals: https://www.youtube.com/watch?v=cY25i_bkUys
https://www.nature.com/articles/s41586-020-2975-4
From my cursory googling this seems to be both available online and safe, yet I haven't seen much discussion of it in the few places that I trust on this kind of material.
Don't most people take this general category of thing as nicotinamide riboside or mononucleotide for availability reasons? Or am I thinking of something else?
That seems right and explains why I was getting so few hits.
I just found discussion of it in your https://astralcodexten.substack.com/p/book-review-lifespan post, which tells me what I need to know.
My half-assed intuitive take after reading way too much literature is - fasting should give you most/all benefits of NAD+/NMN supplementation, plus many others, so I haven't considered it further as it seems redundant.
Interesting
For some reason Google books/play has digital copies of hard to find books at less than half of the price of what you can find elsewhere. I don't know if it's legal or not but I was amazed to get a digital copy of a fairly obscure textbook called "The Rediscovery of Classical Economics" for a fraction of the price I could find it elsewhere. It kind of looks scanned in but it's just fine. It's not a downloadable PDF but you can highlight and save your notes to Google Drive. If you do lots of research this should be a source you check.
Does anybody know a good resource or resources that succinctly explain this Venn diagram of libertarianism? https://en.wikipedia.org/wiki/Left-libertarianism#/media/File:Libertarianism-groups-diagram.png
Before I end up in a deep rabbit hole and with hundreds of wikipedia tabs open...
I don't think this particular diagram is particularly enlightening, but will attempt to give my own classification system/understanding of the landscape.
A useful question about a particular advocate of libertarianism is if they are using a deontological/rights based foundation (eg non aggression principle) or a consequentialist foundation. Note that two individuals could arrive at the same policy bundle through different foundations, so sub-ideologies cannot cleanly be classified into these categories.
I bring this up as the left right groupings are different between these views.
In the rights based view, i see the following 2 axis, positive rights yes/no (the right to something (eg healthcare or food)) and property rights yes/no.
No to property rights places one in the libertarian socialism grouping.
Yes to property no to positive places one in the right libertarian/anarcho capitalist grouping
Yes to property and yes to positive places one in the social libertarian grouping
Alternatively if you are in a consequentialist foundation world, you tend to get more uncertain positions, eg i am 80% sure full anarcho capitalism would work (no state), but if i am wrong i would be happy with a minarchist state (smallest possible state). Which means that individuals tend to be less clearly members of a particular ideology.
But the beliefs on the effectiveness of markets (and thus property rights) is sufficiently bimodel, that you still get a clear socialist libertarian cluster and a clear capitalist libertarian cluster.
I agree, this is an unuseful Venn diagram designed to shoehorn libertarianism into the left-right dichotomy, while deontological/consequentialism is the primary axis.
I... wouldn't say neoliberalism is a subset of libertarianism. They must be using either a *much* narrower definition of the former or a much broader definition of the latter than I would.
(or they're just confused about how Venn diagrams are supposed to work)
I would definitely not put "neoliberalism" at the position they have it. And "Democratic Socialism", even with the qualifier "(in some forms)" seems like an odd thing to include at all!
Is there an effective altruist case to be made for investing in great works of art? Or would a utilitarian framework regard the notion of great art as too mushy? I think art that is goal oriented or dogmatic risks being bad, but surely there is a middle ground between art for its own sake and agitprop?
If the case for beauty is not enough, I don't know what case can be made. Art as Investment Vehicle brings me out in hives, and that's plenty utilitarian.
We can then go on to argue what is and is not Great Art, but there is a strong case to be made for beauty, and some will see beauty in works by modern artists. People can find pleasure/enjoyment/satisfaction/aesthetic stimulation in a work like this one,
https://www.wikiart.org/en/mark-rothko/black-on-maroon-3
and that is just as valid as the pleasure I get from the other one:
https://en.wikipedia.org/wiki/Pallas_and_the_Centaur
Is there a utilitarian case to be made for the beauty of Plato even if his writings are not arguments for utility optimization
Utilitarianism is a means to an end, and the end is maximal happiness and well-being. A beautiful work of art directly contributes to that end. Arguments about utility optimization can only indirectly increase utility, if they convince someone to act in a way that results in more happiness and well-being, and these have to be *really* good arguments if they need to stack up to Plato. (I can't personally comment on the quality of Plato but his fame says enough.)
So, to answer your original question, yes, there is a good case. If you find someone who is a modern Vincent van Gogh, you could justify investing in them over e.g. fighting poverty if otherwise, all future generations are deprived of enjoying their work. The hard part is knowing a modern Vincent van Gogh when you see them.
My thinking is like Laurence's. "Art/beauty" is something like "that which, when people experience it, they think it's valuable, even if nobody understands why it's valuable" Sort of like an experience of value that's direct or unmediated, or whatever mediation there is, is opaque.
So, an argument for beauty from utilitarianism is a tautology. Utilitarianism is probably smuggling beauty in through definitions, no need for an argument.
Have you read Kant's Critique of Judgment? I believe that's where I got this idea from.
An aside, but as I recall JS Mill actually claimed to derive his Utilitarianism in part from Plato's Protagoras.
This discussion of finding aesthetic stimulation in a work like "Black on Maroon" reminded me of this Substack post: https://ruinsruinsruins.substack.com/p/one-christians-thoughts-on-the-usefulness
I think that I possibly don't "open myself up to" enjoying art when I don't know what it's intending to "say"/convey! But when I look at "Black on Maroon," (and especially the "Rust and Blue" one... I see; they are by the same guy. Hence the resemblance--I am an ignorant oaf about this.) I DO feel myself relaxing a bit. (The "Rust and Blue" colors are delicious.)
Maybe the higher level question is whether there is an aesthetic (i.e. "artistic") argument for utilitarianism.
Asked another way, could Beethoven, Rothko, David Foster Wallace, or insert your favorite artist here justifiably win an EA grant or an ACX grant? What about philosophers whose work is not rationalistic, but whose work might contribute to beatifying the world through style or just through opening horizons of new thought?
There is definitely a good argument for the utilitarian value of art, but buying high art as an investment isn't a very good way to incentivise the production of more art. You'd be better off donating to art education charities, schools, etc. Or commissioning art from people online.
As an investment vehicle in itself its almost certainly inferior to index funds
Replace the word "investing" with "buying" and you have a better case, assuming you mean the former to be an instrument that will appreciate in financial value. The art market is the ultimate vehicle for insider trading and only very few wealthy insiders working with a handful of high end galleries will ever see an appreciation in value. The other 99.99% of art buyers will see their purchase price drop to near zero after walking out the front door.
On the other hand, if you want to purchase a quality work from a starving artist, I think of art as a liberalizing (in the positive sense) force and not only will you support the art industry but hopefully you and your guests will have a work beauty on the wall to both admire and contemplate.
I would say it's better to invest in artists than to buy art. Not invest in them for financial gain but for the sake of the art itself, if it is worth being produced. From an EA perspective, if you're going to spend any money on art, you should fund the starving artist who has the greatest potential to produce something that will go down in history as a great work, and appreciated by countless people for centuries to come.
Yes, this is impossible to predict, but shouldn't we try anyway?
Here's what Peter Singer had to say about art museums in The Most Good You Can Do:
"Art surely can be an aid to learning, but building new museums is not likely to be the most cost-effective way to do that. We have other opportunities for studying art from which we will learn as much or more than we can by joining the crowds peering at very expensive paintings from behind a rope or through bulletproof glass. If the goal were really to educate the public about art, museums would do better to spend a few thousand dollars on the highest-quality reproductions and allow the public to get as close to them as they like.
To forestall misunderstandings: there is value in creating and enjoying art. To many people, drawing, painting, sculpting, singing, and playing a musical instrument are vital forms of self-expression, and their lives would be poorer without them. People produce art in all cultures and in all kinds of situations, even when they cannot satisfy their basic physical needs. Other people enjoy seeing art. In a world in which everyone had enough to eat, basic health care, adequate sanitation, and a place at school for each of their children, there would be no problem about donating to museums and other institutions that offer an opportunity to see original works of art to all who wish to see them, and (more important, in my view) the opportunity to create art to those who lack opportunities to express themselves in this way. Sadly, we don’t live in that world, at least not yet."
"In a world in which everyone had enough to eat, basic health care, adequate sanitation, and a place at school for each of their children, there would be no problem about donating to museums"
This very much does not follow. What he is implying is that since food is more important than art, we shouldn't fund art. But by his own standards, just because someone would die without food and wouldn't die without art doesn't make food more important than art; you need to do a utility calculation and in the calculation, lives don't have infinite value.
A utilitarian cannot a priori restrict other people's utility functions. There's really no coherent notion of a well-formed utility function being "too mushy." John Stuart Mill tried to give arguments about what utility "ought" to be, but it's pretty clear that those arguments were parasitic on non-utilitarian ethical ideas.
Whether the above tells for or against utilitarianism is a separate question.
Undying glory (found only in great works of art) more accurately tracks human well-being (esp over the long term) than economic modeling, because economic modeling incompletely measures the human condition.
Consider the value of a stay-at-home parent. This is not very easy to track economically, and harder to recompense, but (esp given the right parent) it can pay off on the scale of billions. Easy to mismeasure economically, but practically every Great Book enshrines "honor thy father and mother...that thy days may be prolonged, and that it may go well with thee. (etc)"
A society that understands this (and things like it) well cet par be better off than one that doesn't.
Re: objective framework for art. I think there is an inside view and an outside view of what makes art objectively valuable.
I think the inside view can be summed up in the maxim "nothing too much." Not a stroke out of place. Every arbitrary choice tends to the same end to form unity in variety, a perception of parsimony. Because the choices an artist can make depend on the convention of the form, whether a piece of art conforms to the "nothing too much" standard is something generally only the learned in the field can judge. Sometimes this makes laymen mad.
I think the outside view of beauty can be summed up by the phrase "undying glory." What's the news that stays news? What are the books that stay in print? What are the lindyest of the lindy?
The argument here is that (a) a work's popularity could just be accidental -- a fad -- and so inherently worth nothing, but that (b) the longer a work stays popular (especially through many societal "selections"), the more likely it is that it's NOT a fad -- i.e. that there's something inherently valuable that's causing the fad phenomenon. You generally also find that experts agree on what these works are.
For example: what made Aristotle the philosopher of choice for the Greeks, Romans, Golden Age Muslims, Scholastic Catholics, German Idealists (depending on who you read), Existentialists, and now modern Virtue Ethics? These societies are pretty different, and I'm only willing to buy the "historical contingency" argument for so long before I think that there's something to explain the 2500-year staying power (esp. when we think that everyone knows the name "Aristotle" and nobody knows the name "Longinus").
"For example: what made Aristotle the philosopher of choice"
I'd go with historical contingency.
First of all, everyone in your list is related. The Romans copied lots of things from the Greeks, for instance, and by the time we got to virtue ethics, Aristotle was very well known in the culture.
Second, there's a difference between Aristotle's ideas being inherently valuable, and Aristotle being inherently valuable; it may be that his ideas are useful, but not hard to come up with, and if he hadn't existed, someone else who came up with them would have lucked into popularity.
Third, ideas can spread for memetic reasons; the Church adopted Aristotle's ideas, and the Church is a powerful force for spreading ideas regardless of their merit.
1. How is the Islamic Golden Age related to the Romans except through Aristotle? Likewise, why do the Catholics (who were both anti-islam and anti-pagan) pick him up in the first place?
2. This is somewhat of an aside, but Aristotle easily fits both categories. The ideas are important and they're hard to come up with, both in content and in form (i.e. Aristotle's unique method of inquiry/thought). cf. e.g. the invention of Symbolic Logic, which wasn't improved until either Leibniz or Frege (and depending on who you ask not even then).
3. (a) There are no successful memes without merit. Even if the merit is just "fitting well" in a particular environment. However, like I was saying, with Aristotle, the "contingent-upon-environment" argument becomes less convincing as the years pass. (b) the Church doesn't just adopt ideas for no reason either.
"How is the Islamic Golden Age related to the Romans except through Aristotle?"
Islam ruled vast swaths of land, much of which had been previously occupied by Romans, and adopted some ideas from the conquered areas.
"Likewise, why do the Catholics (who were both anti-islam and anti-pagan) pick him up in the first place?"
Catholics weren't "anti-pagan" in that sense. Notice, for instance, how reference to Greek and Roman gods stayed a fixture of literature in Catholic Europe.
"There are no successful memes without merit. Even if the merit is just "fitting well" in a particular environment."
The argument about Aristotle implies nontrivial merit, not just any old merit.
1. Why these ideas? Why does this idea get selected over and over again and not others? (e.g. Why was Aristotle adopted but not Greek togas?)
2. Catholics pre-Aquinas were pretty opposed to Greek Rationalism in general and Aristotle in particular. It's historically really surprising that they integrated him. Many Aristotelian texts didn't even survive except through the Muslims.
3. Do you think adoption by the Catholic Church is trivial? The church has strict selection criterion esp for stuff they integrate so deeply into doctrine. Even if Aristotle happens to "fit well" with the environment of the Catholic Church, the Catholic Church is also a nontrivially meritorious meme in its own right.
"Why was Aristotle adopted but not Greek togas?"
If Greek togas were adopted and not Aristotle, you could say the same thing. Given that they liked the ancient Greeks, some subset of things would survive; Aristotle happened to be one of them.
"Do you think adoption by the Catholic Church is trivial?"
It's trivial in the sense that it has little bearing on the question you're bringing it up for.
FWIW, Aristotle is remarkable not only for his longevity and range, but also for the extent to which his ideas penetrate and then permeate the societies in which he's adopted. E.g. Aquinas doesn't even use his name, instead just calling him "The Philosopher."
"Looking for a Chri…fine, sorry, looking for a Martin Luther King Day gift this year for the rationalist in your life?"
No Christmas? How very Puritan of you 😁 https://www.youtube.com/watch?v=jBCxE8tUIWM
Though it's nice to see the return towards celebrating saint's days, even in a secular manner!
It's January 3.
That’s right, 4 days left.
Still within the Octave/Twelve Days of Christmas if we're going the traditional route, which only ends on January 6th (Twelfth Night) with the Feast of the Epiphany!
Off topic but I just tried your how many surrealists does it take to screw in a light bulb gag on my wife. I shrugged at that one but it tickled her.
I don't know if I should be pleased or worried for you that she seems to have a similar sense of humour as me! Glad to know the joke worked for her! 😀
People often say processed carbs are bad. I've never quite understood this. Why would it matter if something is 'processed', and what that does even mean?
Here is my best guess of what is going on, but this could be totally wrong (this comment is very much meant as a question). The guess is that 'processed' is a misnomer and it's really about something like 'modified'. It's not actually the case that people in the ancestral environment never had too much to eat, so are bodies are capable of regulating food intake to some extent, but they're used to food with certain compositions. Therefore, if you take grains and make them into flour and make flour into bread, this is still okay even though it's quite 'processed'. But if you only take part of the grains to make the flour, well then you modify the natural composition, and as a consequence, human bodies are worse at regulating it and so you may put on weight.
This would also suggest that modified carbs are really only a problem if you're overweight and otherwise not particularly unhealthy?
My understanding is that it's more about glycemic index ("fast-acting" vs "slow-acting" carbs) and the fact that most types of processing generally increase the glycemic index; and also about the mixture of carbs vs fiber, where most types of processing keep the carbs but remove the fiber.
FWIW, fiber *is* carbs, just not (very?) digestible. That's why it's normally included under the carbs category. But when the category is broken down it's usually into fiber and sugar, but those don't add up to the grams of the whole carbs category, because they don't usually list starches, etc.
And *that* isn't the whole story. Even things with identical amounts of fiber, starch, and sugar can be different depending on their accessibility. Cooked carrots are high-glycemic and raw carrots very low glycemic. So much so that a lot of the sugar is likely to make it's way through the digestive system without being absorbed (depending partially on how well you chew the carrorts).
So processing is important even in cases where it doesn't change the relative percentages of nutrients. (Though, yes, it usually does.)
Wait are you saying not chewing carefully helps you to keep the effective glycemic index low [in some foods]?
Carbs aren't bad, that's just one of many dietary misconceptions. A more accurate approximation might be "carbs without fiber are bad", because they tend to be easy to overeat and have higher glycemic index. Fruits, vegetables, and whole grains are all healthy sources of carbs, and are perfectly appropriately even when on a fat-loss diet.
"Processed food is bad" is not an unreasonable generalization, because processing tends to remove fiber (making it easier to overeat) and some of the more nutritious parts of the food. Going from wheat to white flour is an example where a lot of nutrition is lost; in fact, most flours you buy will be enriched to add back some of the lost vitamins.
All that said, "bad" foods in moderation are usually fine.
"processed" also often means "with extra salt and sugar"
> "Processed food is bad" is not an unreasonable generalization, because processing tends to remove fiber (making it easier to overeat) and some of the more nutritious parts of the food.
Ok, but doesn't this mean "it's really about modified carbs, not processed carbs" is accurate?
I'd push back on the sweeping "all fruits are good" generalization here. Many if not most of the fruits you find in your grocery store - including (gasp!) the organic ones - are bred to contain insane amounts of sugars, and either as a result of selective breeding or something else, but they don't contain that much fiber either. Plus, that sugar is mostly fructose which some have argued is actually worse than glucose (though this last part is by no means a consensus).
Charitable interpretation - Processed means concentrated beyond what you could usually find in nature outside of a beehive. More calories per volume or weight. Less fiber and micronutrients. More calories to reach the same level of satiety. More commnonly mixed with fats (fats + carbs is the least satiating combo of any 2 macronutrients).
Unchartiable interpretation - Its a purity fixation.
Arguments from genetic evoltuion in nutrition are usually horseshit. We evolved to eat anything we could get our hands on.
Processed carbs are *typically* bad for WEIRD folks because we *typically* don't need to rapid injection of easily-available energy when we do our *typical* activities like sit at a desk for 10 hours followed by recreationally laying on the couch for 6 hours. But my dad was on a road crew and could put away 4k calories a day and getting processed carbs in was no problem for him staying lean. Lumberjack camps served like 7k calories a day back when trees were felled with muscle power, and they ate a ton of simple carbs.
Its "best" imho to think of food as fuel, and matching your fuel to your needs.
I can only maintain a healthy weight on a "low carb" diet, and here's my understanding of why for n=1:
1) Americans eat just an absurd amount of carbs. So when I say "low carb" that actually just means "my entire diet isn't made up of flower and sugar with hardly a protein to be seen."
2) I do better with "non-processed" carbs. You're absolutely right that this is a nonsense phrase filled with nonsense. But the best way to describe what I'm looking for is that usually a piece of plant nutrition is made up of a shell with a fair amount of fiber/other good-for-you things and a gamete which is where all the calories are. Sugar, white flour, and corn syrup are "processed" to remove the nutrient rich shell. As a result you don't get full when you eat it and you tend to overeat.
The term is bad though because whether a carb is "processed" or not isn't the best way to judge this. A baked potato isn't "processed" by any reasonable use of the word, but it's still a calorie bomb with very little satiation or nutrition. Instead, best to look for foods where the outer layer is doing reasonable work at filling you up - carbs with big fiber content, potassium, omega-3 rich oils and small gametes such as grain, dark greens, non-white rice.
Probably also worth mentioning that if you are *active* and already at a healthy weight those low-satiation carbs are not quite as big a deal. Your body just burns them in your triathlons or whatever.
Weren't there some people claiming to lose wieght by eating nothing but potatoes for a year? I don't know if it's a good idea, but if they can live on just potatoes surely there is sufficient nutrition there?
Spudman! Lived on Potatoes for a year, lost a ton of weight, and did not die. To be fair, potatoes are definitely healthier than, say, white bread. Spudman ate a variety of potatoes, including sweet potatoes and smaller potatoes with thicker skins. He also supplemented with vitamins since he was not getting enough protein.
He does not recommend eating only potatoes and only did it to make the point that potatoes are not super bad for you and are a reasonable part of a balanced diet. I agree. But the fact that it worked doesn't really prove anything other than that he was in a calorie deficit. People have done the same thing with pizza.
I probably overstated my case here - I, like most Americans, am a simple carbs fiend, and so cutting simple carbs does not really put me at a carbs deficit. Carbs are still the foundation of a balanced diet, you just don't want to eat so much simple carb that you put on a ton of excess weight. The biggest issue with American diet is the prevalence of soda, for instance.
The tautological way to put it is that if you're not eating too many simple carbs you don't need to cut simple carbs. But Americans really really overestimate how many simple carbs are healthy.
I knew I wasn't supposedly supposed to eat potatoes, but I've found them to be rather filling.
I think there are some quite serious indication that processed food is bad for you. Most importantly it increases your weight, but perhaps also inflammation beyond the weight, and it has some other bad consequences. This is pretty well established. For example, lab animals gain weight from so-called "cafeteria diets", but not from diets which have the same ratio of carbs, fat, protein, sugar, and so on.
But we have plainly no idea why. There are dozens of popular theories: glycemic index, types of oils, fibers, sugar content,... And as far as I can tell, none of these theories withstands even a modest level of scrutiny. We can most certainly rule out explanations along the lines "it contains too much sugar" or "it contains the wrong types of fat" or "the fats/carbo/protein ratio is bad".
So our best conclusion is "probably processed food is bad, and all major candidate explanations are wrong".
For a book-long discussion (focusing on the weight part), see here:
https://slimemoldtimemold.com/2021/07/07/a-chemical-hunger-part-i-mysteries/
Their favorite explanation is that lithium is the culprit, and that it enters food because machines are often greased with lithium-containing substances.
it seems like the answer for questions on nutrition is always "we have no idea". I've read about half of the article you linked, and it pattern matches to solid work, so I cautiously trust your take.
But even so, shouldn't we at least be able to define what 'processed' means? Is it the amount of steps taken to do stuff with the grains? The amount of chemistry involved? The amount of machines? The current concept is so vague that I can't tell whether white bread more processed than whole grain bread.
In one of the later pages in the Slime Mold Time Mold series, the writer conjectures that "processed food" is more likely to have a higher concentration of contaminants.
I share your frustration, but I really think we don't know. This is something that I have heard from many sides, not just from the linked article.
The standard recommendation is "have as few steps as possible done by industrial machines and as many as possible at home in your kitchen". This is probably a pretty bad proxy for the unknown thing that actually matters. But we don't have a much better proxy than that.
"Processing" is a vague term that basically means doing anything to food (e.g. milling, chopping, cooking, adding ingredients, removing elements, etc.). It's too broad of a term to always be bad. Some processes, such as fortifying common foods with vitamins, are done to improve health.
There are some reasons why "heavily processed food is bad" might be a good rule of thumb.
First, processing often increases the convenience, which may cause us to overeat. Compare shelled pistachios to unshelled pistachios, or oranges to orange juice. You're much more likely to eat a larger quantity of the processed version.
Second, processing improves taste, which may cause us to overeat. There's a reason we cook meals instead of eating a bunch of raw ingredients. If you're like me, those freshly baked chocolate chip cookies are more tempting than the bowl of apples and bananas.
Third, salt, sugar, and other easily digestible nutrients tend to be tastier, and processed foods may have too much of these.
Fourth, it's too easy to buy large quantities of processed food. Home baked cookies might not be any healthier than store bought ones, but just the effort of having to bake cookies from scratch each time will probably cause you to make them less often and eat fewer cookies overall than if you can just throw boxes of them in your shopping cart.
I think a lot of the harm of processing is simply from making food too convenient, tasty, and easily digestible.
I'd like to add a Fifth to this list which is processing breaks down the cell structure of the food which changes how the food is processed by the GI track in a way which is presumably detrimental to nutrient bioavailability as well as the glycemic response.
Do you have a link to the animal studies you mention? My current model is that "processed food bad" sentiment predominantly comes from observational studies indicating that people who eat a lot of processed food tend to have poor health, but it's a correlation that has little to do with processing itself making the food bad somehow (though of course it's hard to deny that a lot of processed food is just obviously crap). Decent animal studies would indicate against this.
I have only read secondary sources, but Section 6 ("Mystery 6: The Palatable Human Food") here mentions half a dozen of studies:
https://slimemoldtimemold.com/2021/07/07/a-chemical-hunger-part-i-mysteries/
As far as I understand, it's actually not so easy to find a diet that makes lab animals obese, and this was a serious challenge before researchers found out that you can simply feed them human food. High-fat diets work less well for making animals obese (there is a direct comparison study linked above), and high-sugar diets seem to be used even less. I think they just don't work at all, though I am not completely sure about that point.
Citation that I stole from the link above:
""Palatable human food is the most effective way to cause a normal rat to spontaneously overeat and become obese,” says neuroscientist Stephan Guyenet in The Hungry Brain, “and its fattening effect cannot be attributed solely to its fat or sugar content.”"
If you want to dig into it, then this book might the place to start.
Ah I see, so in that study rats were fed ad libitum, that changes everything. Rats on human food gained more weight, but simply because they ate more calories, you can see it from the charts. And of course they did, the human food they used in the study (cookies, cereal, processed meat etc) is hyper palatable to humans and rats alike, but unlike (some) humans rats don't have the common sense to not eat an entire meal of snacks. And they had comparison groups on standard chow, high-fat diet and low-fat diet (with fat calories replaced with sugar so it was effectively high-sugar diet). And all groups were fed ad libitum. Rats on all 3 diets gained a lot of weight compared to the controls on standard chow, with high-fat and high-sugar faring about the same. On the human food they gained the most, but again, they ate more calories, the diets weren't matched by calorie or nutrients content at all. So nothing surprising here except maybe for high-fat and high-sugar giving the same result, I'd expect high-sugar to induce more weight gain and insulin insensitivity.
On the broader point, I have quibbles with all their "mysteries" except for 5. Most importantly, they basically ignore the effect of exercise - they handwave higher exercise in 1, and then completely ignore it in 4 where it could easily explain all the difference. In 8, they directly contradict their sources. The meta-analysis they link says that people on low-carb diet lost 8.73 kg in 6 month and 7.25 kg in 1 year, similar but slightly less on low-fat diet. For the population with median weight of 91kg that's quite a fucking lot. Now the article somehow concludes from this that "diets don't work" and "even if someone does lose 20 lbs, in general they will gain most of it back within a year". No, they're not gaining most of it back, they are gaining ~3lb of it back! And even for this, a part of the reason is the study design - the median duration of a diet was 24 weeks, so by the end of the year many of those people haven't been on a diet for a few months!
Generally, it all sounds like a lot of motivated reasoning twisted to prove that *clicks through to the last chapter* oh right, eat whatever you want, don't exercise, industry and the US government are to blame. Ok this was a bit uncharitable, but seriously, lying about your sources is bad so they deserve it.
They address your points in later parts of the article series (e.g., exercising and 4). Also, I have recalled that Scott has written a book review of The Hungry Brain,
https://slatestarcodex.com/2017/04/25/book-review-the-hungry-brain/,
which also says a few things about that.
For the question whether "diet's don't work", I think that's a question of what you expect from them. By Amptoons' standards, that they should bring fat people into normal weight regimes, it is pretty clear that diets don't do that, https://amptoons.com/blog/?p=22049. And the 7-10kg of weight loss is not for a population of 91kg, most of these studies were restricted to obese participants. Correct the studies for weight and dropout, and I can see the point of not finding the results impressive.
I do agree that the statement "even if someone does lose 20 lbs, in general they will gain most of it back within a year" is not supported by their link, and they should take the blame for this. But I am not sure that this makes their general point invalid. My first google hit was this, which easily supports the magnitude of such weight regain, though in a time frame of 2+ years instead of one year.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5764193/
The study says right in it's abstract: "Conclusions and Relevance Significant weight loss was observed with any low-carbohydrate or low-fat diet <...> This supports the practice of recommending any diet that a patient will adhere to in order to lose weight." And they somehow infer from this "Diets Don't Work". At this point I just wash my hands, I'm not wasting time on first finding out what they have to say about hunter gatherers and exercise (it should be right there when the topic first mentioned, otherwise all their comparisons are irrelevant), only to later find out their "evidence" for this is some study which says something diametrically opposite.
>And the 7-10kg of weight loss is not for a population of 91kg
Yes it is, read the paper, specifically the "Results" section. It is the median weight of the population they investigate. Yes median is a shitty metrics in this case, but that's all they report.
And the paper you linked says in almost every paragraph that a regimen of diet and exercise *works fine*, the problem is that it's difficult for people to *maintain* this regimen, for various reasons. Main of which is the same as with the poor rats: hyper palatable food easily available in unlimited quantities. So absolutely no mystery that would require pollutants for explanation - people it less and loose weight, then they stop eating less and regain weight. The entire "Clinical recommendations.." section is focused on ways to *keep people eating less [crap]*.
Besides, I always wonder when people make arguments like this - don't you, like, personally know someone who has lost weight via diet and/or exercise? Because I certainly do know a few such people, all of them basically say they can control their weight insofar as they can control their diet (the latter being challenging for some, for sure).
>The Hungry Brain
Direct quote from the review: "Researchers have been keeping records of how much people eat for a long time, and increased food intake since 1980 perfectly explains increased obesity since 1980". I'm not sure what "few things" you were referring to, but at least this book/review doesn't seem to support the idea that calories and nutrient composition are irrelevant, which is the main point I'm arguing against here.
"Processed" is a proxy for particular, characteristic modifications. Carbohydrates range on a spectrum from simple, (roughly, sugars), to complex, which the body breaks down into sugar for use. Commercial processing characteristically breaks down carbohydrates to simpler form and, importantly, concentrates them. The production and heavy use of high-fructose corn syrup is one example.
While admittedly a proxy, I would say that "processed" carbs actually does have fairly high correlation with a significant problem in the modal American diet.
"Processed" in the context of food being bad for you mostly means the gross composition of the food has been altered. So for example when ordinary wheat or rice is "processed" after harvesting, the seed coat and germ are removed (removing almost all of the fiber and protein), and only the stach kernel is left. When orange or apple juice is made from apples, the juice, which is high in sugar, is removed from the pulp, which contains all the fiber.
There are also some chemical modifications. The most famous of these is the production of high-fructose corn syrup from corn starch, in which the starch is broken down into glucose monomers, and then about half of those are converted to fructose, an isomer of glucose, and the replacement of sucrose (which is a dimer of glucose and fructose) with HFCS, which is cheaper to produce (in the US at least). This alters the chemical composition of the sweetener slightly.
The argument is that that you evolved, 1 million years ago, to eat foods with a certain gross mixture of nutrients, and this is optimal for health. For example, in a primitive state, where only whole raw foods are readily available, it's very difficult to consume large amounts of sugar, because it simply isn't found in high enough concentration. For example, an apple contains about 20g of sugar, which is about 1.25 tablespoons. You can easily eat 5-10 tablespoons of sugar in one sitting, either by itself or in some baked food -- but it would be rather more work to wolf down 4-8 whole apples at once. Additionally, it's thought that a certain mixture of gross nutrients -- so much protein to so much carbohydrate to so much fat -- results in less imbalance to circulating metabolites, like blood sugar, and fewer demands on regulatory pathways like that controlled by insulin. And, finally, there is the more recent hypothesis that some significant effect on our health comes from the health of the "micriobiome" of bacteria dwelling in our guts, and if you don't "feed" those critters the right mix of nutrients, you get unwanted changes in the demographics -- certain desirable bacteria die off, and certain undesirable bacteria populations grow.
All of these are necessarily speculative, with most evidence for or against coming from epidemiological studies if not mere anecdote ("I ate an apple a day every day and my asthma was cured!") so...pretty far from empirical soundness. Of all of these, my favorite is the microbiome hypothesis, on account of it explains how an excess of *an actual nutrient* can turn out to have chemical toxicity -- it's not the nature of the nutrient itself, but rather that it "feeds" some highly undesirable microbes which produce the toxic consequences, sort of like a "red tide".
Edit: I will add that there is also an argument, based again on the microbiome, that we are used to having certain species of gut bacteria regularly replenished from our food, e.g. certain species of soil-dwelling bacteria that tend to cling to roots when you eat them. And that the processing of plant foods we do going from field to table is *designed* to kill off as many bacteria as possible, make the plants foods if they are whole squeaky clean, and if they are turned into ingrediens nearly sterile, and that this may deprive us of some unknown necessary "micronutrients" in the sense of certain valuable bacteria species we evolved to eat regularly.
>This would also suggest that modified carbs are really only a problem if you're overweight and otherwise not particularly unhealthy?
Not really, you definitely can have metabolic issues without [yet] being obese. Insulin insensitivity is the most common of them, and roughly speaking it's driven first and foremost by spikes of glucose in your blood that come after gorging on lot of easily digestible carbs without much to dilute them or fiber to slow down their uptake. Even if these metabolic issues never lead to obesity, they can lead to other problems: increased inflammation, broken lipid profile (leads to CVD), and maybe non-trivially increased chance of some cancers (but don't quote me on that). The standard fasting glucose test your doctor gives you will catch it only when the problem is well underway.
Being young and physically active definitely helps a ton, but the former only works for a while and of the latter they say "you can't outrun your bad diet".
Anyone see this article on Wired: https://www.wired.com/story/the-teeny-tiny-scientific-screwup-that-helped-covid-kill/
Assuming it's true (any one with expertise care to comment): "science" has spent the past few decades not really understanding how viruses spread. It seems crazy to me. But somehow believable.
IIUC, it's true. There were a lot of people saying from early cases that the spread was probably due to aerosols. I decided that was true based on an early chinese restaurant case. But this wasn't the officially adopted explanation until a lot more proof was available.
AFAIK, there are still arguments about the relative importance of different modes of transmission. People around here are still cleaning surfaces. Perhaps that helps. I'm still washing my hands frequently and using alcohol wipes in between. I suspect this is excessive, but don't feel like testing. I've seen studies that COVID was detectable on paper 3 days after exposure, but I don't know whether what was detected was a live virus or just fragments.
Reads like sensationalized half-assery to me. It conflates a serious and difficult point, which is deciding the relative contribution of aerosols and fomites to infectious spread, which *everbody* has known for centuries varies by disease and conditions, and turns it into a cartoon plucky peasant heroine v. Arrogant Establishment narrative that is probably Chapter 11 in How To Write Hollywood Screenplays. Really, all we need is a tense Hans Zimmer soundtrack and a comely actress to play the lead (while Willem Dafoe can be dragged out of retired to play the WHO baddies) and we're looking at box-office gold.
It's certainly true that the presumption early on was that COVID, like most coronaviruses (e.g. those that cause the common cold) was more transmissable by fomites than aerosols. That was the basis for the early advice, which, alas, turned out to be wrong. But no one could have known that at the time. And the fact that it was hard to turn the gigantic ship o' public health around once it was off and running...well, duh. The presumption that one bright-eyed intelligent worker with a handful of data and a good argument should make such a ponderous org do a 180 inside it's own length only happens in...well, Hollywood. Reality is way more fraught.
But a story that reflected that reality wouldn't be nearly as exciting to read.
Wait, where did you hear that the other coronaviruses spread significantly on surfaces? The viruses that survive long times and spread significantly on surfaces are capsid viruses. Enveloped viruses don’t, in my understanding from virology class years ago and from searching in the literature. I was kind of appalled early on that virologists who should have known better didn’t speak out about the unlikeliness of significant COVID (an enveloped virus) surface spread for months while millions of people were going to great lengths to sanitize their groceries and everything else. (I certainly didn’t bother based on my understanding of capsid vs enveloped viruses.) But maybe I was mistaken and there is literature suggesting that surfaces are a significant source of spread for the other coronaviruses?
You're right, I was sloppy -- I should have emphasized the direct transmission (e.g. touching of hands) rather than fomites. The conventional wisdom for years has been that most rhinoviruses and (ordinary) coronaviruses spread through direct transmission at least as well, maybe better, than through aerosols, hence all the advice about washing your hands, the hand sanitizer craze, et cetera. You can find that advice everywhere. Transmission via fomites varies all over the place, so far as I can tell -- and this is not my area of expertise. I believe influenza has been shown to routinely remain viable for 48-72 hours on stainless steel surfaces and such, which is alarming for an enveloped virus :(
All of which reinforces my argument that, like much else about virology and immunology, It's Complicated, and trying to turn the struggles of the community to ferret out exactly what is what into a Manichaen MCU storyline is at best not helpful.
Ok, thanks. For flu, fomite spread can happen, but I remember reading an estimate that fomites may account for low to mid single digit percent of transmission.
People love to tell the story of Semmelweis - all the doctors of the time thought diseases were caused by bad air, and here's this upstart telling them to wash their hands, and history eventually proved him right. Given the centrality of this story to much of the picture of medicine and public health, it would be not at all surprising if the entire medical community had overweighted "wash your hands, don't worry about the air so much" for over a century.
Regarding the "It should take 0.5-2 hours to write an application (setting aside time actually planning the project)" part, I haven't looked into this particular process, but for the grant applications I have written (various academic options) it generally takes 40-100 hours of work in addition to actually planning the project.
I assume that your process is simpler, but I'm also quite convinced that if you would do a post-application survey of "how much time did you spend on the application" then almost noone would answer 0.5-2 hours, and the majority would be 10+ hours even if you think that it should take less than 2; people applying for grants generally do spend excessive time on various details that may seem trivial to application evaluators.
FWIW:
- By "It should take that long", I mean "It taking that long is fine and is what EA Funds wants". I didn't mean "This is my prediction of how long people will actually spend on it". I should've been clearer about that.
- For context, the EAIF application page says "We aim to keep our application process very simple and fast, and recommend that applicants take about 1–2 hours to write their applications. This does not include the time spent developing the plan and strategy for the project – we recommend thinking about those carefully prior to applying." https://av20jp3z.paperform.co/?fund=EA%20Infrastructure%20Fund
- I'd indeed guess the mean and (less confidently) median length of time spent on applications, setting aside planning, is >2hrs. I think this is somewhat unfortunate.
- I'd guess that the median is <10, maybe around 3hrs or so?
- I'd guess the mean is notably higher, maybe like 6 hours or something? (I feel even less confident here than for the median.)
- I'd guess that how long people spend probably varies wildly, driven largely by how large the grant is, how crucial it is for the person (like does this determine the next 6-12 months of their career/organization?), and how integrated into the "core EA community" they are (the more they are, the more they'll believe that a quickly written application is fine and the more they'll feel confident relying on our prior knowledge of them).
- I'd guess that at least 10% of applicants indeed spent <2hrs. I've seen some applications that were clearly written very quickly (and if I recall correctly most of those were in fact approved, which is at least weak evidence that quickly writing an application does little to harm one's chances).
- These guesses are informed both by applications I've read, interactions I've had with applicants, and the ~3 times I've been involved in writing an EA Funds application myself (before I joined the EA Funds team). I haven't looked for or gathered systematic data, though.
- I also wrote EU grant applications for a company before, and that was horrifically time consuming. I'm quite confident people tend to find the EA Funds application *far* less time-consuming than grant processes for e.g. the EU or academia.
Asking people in a post-application survey how long they spent sounds interesting and potentially useful. I'll look into whether that's ever been done and, if not, whether it should be.
Yes, you definitely should. (See my comment above.)
> I'm quite confident people tend to find the EA Funds application *far* less time-consuming than grant processes for e.g. the EU
This bar is awfully low. EU funding requires so much paperwork that in numerous cases organizations decided to abort applying as expected applying cost was similar to expected funding received
> I'd guess that the median is <10, maybe around 3hrs or so?
I expect this to be untrue, with median for funded grants at least full day of work. Maybe even multiple days. (based on own experience of applying to grants that claimed to be easy to apply)
> I'd guess that at least 10% of applicants indeed spent <2hrs. I've seen some applications that were clearly written very quickly (and if I recall correctly most of those were in fact approved,
Though maybe I actually should send first draft after first hour of work?
My guess is that someone with an appropriate level of knowledge of the issue they want to work on, (assuming a Bachelor's Degree), you should be able to vomit out a short proposal in less than an hour. Maybe reread once or twice and submit?
If you are working with a collaborative endeavor which isn't funds-constrained, you only need to pass the threshold for "worthwhile funding" rather than a tournament-style competition where you have to be both "worthwhile funding" as well as "better proposal than most of the other applicants".
I guess that I should look through my pie-in-the-sky TODO list and look what may qualify for funding.
Case in point: my NIH F31 application actually took 60 hours, and claimed to take 12.
ya, when Scott said his application would take "15 minutes" or whatever ... I knew there was no way I could do that. but i also realized that's me, and wasn't sure how much i could generalize there
My ACX grant application (https://astralcodexten.substack.com/p/open-thread-198/comment/3648492) was basically just a joke and still took an hour to write.
Does anyone know if Nick Bostrum has ever addressed computation loss in chains of simulated worlds? ie That there might be a limit to how many dreams in a dream there can be? Never understood from what I’ve read of him why simulation B from parent universe A couldn’t turn around and then simulate parent universe A in such convincing fashion you couldn’t tell who gave birth to who if you can just carry on that kind of nesting forever.
Have not read Bostrum, but I wonder if software emulations of older computers might offer some insight?
For example, these days it is routine for newer platforms to emulate older platforms, because they are much more powerful. It’s fairly easy for a current Wintel PC or Mac to emulate an Apple II.
I guess it is theoretically possible to emulate a Wintel PC on an Apple II, but it would take a tremendous amount of programming effort, and the resulting implementation would run incredibly slowly.
In a nutshell, possible, but not worth doing.
What you’re saying makes sense to me and is more or less how I think of it but it seems like that exponentially caps your number of universes. You could bud many from the true parent, some from the children, fewer and fewer from the grandchildren. However, I’m sure Nick Bostrum/someone else who ascribes to that theory must have thought of that and have a counter argument. I’m just not aware of what it is and can’t seem to find it. Or maybe the argument is more limited in scope than I understood (this is true only when you have near lossless reproduction, etc).
There are two obstacles - essentially information and computing time.
Simulating an universe requires storing at least as much information as anything it can simulate - so a finite universe can only simulate equal or smaller universes (although for many needs properly simulating a tiny fraction of that universe, like a single planet, is sufficient and everything else can be faked relatively cheaply, but we'll stick to "proper simulation of an universe" for the sake of argument). I'm not sure if infiniteness solves this problem, since at least we seem to be limited to a finite observable universe.
A similar restriction applies to computing time - simulating an universe takes at least as much computation as any universes it might simulate - however, that can be solved with time dilation, having the simulation "run in slow motion"; perhaps you're spending a year of limited computation power to simulate a microsecond of more complex computation. If you can solve the storage problem (perhaps "cheating" by using an infinite universe) then you *can* have arbitrary nesting, it's just that the "simulated-A-within-simulated-B" would be much slower than the actual parent universe A but have no ability to measure or detect the slowdown.
Also, there is the notion of arbitrary causal intervention. The parent universe can have the ability to arbitrarily alter or destroy or clone the simulated universe. That does provide an ability to distinguish who gave birth to whom - although before *trying* to distinguish them I'd like to refer you to the short story "I don't know, Timmy, being God is a big responsibility " at https://qntm.org/responsibility as an interesting exploration of implications of such nested simulations.
Fantastic story! I think I may need to restate the part of the hypothesis I had in my head to check my thinking and it may impact the story as well. My understanding of Bostrum is that we are more likely than not to be living in a simulation (just like in the story) given a nesting of simulations springing from some kind of ur parent universe that has some semi magical property of being the prime reality. However my thinking has always been that can only be true if the ur universe and its simulations has some kind of infinite nesting power. (This may be related to mathematical vs computational universe arguments but I’ve only been aware of those for about a year) If it’s not infinite it seems like that dashes the argument quite a lot as even near infinite nesting power would see an exponential decline in how far down you can really go before you run out of compute power. (Let me know if I’m thinking about that wrong, my gut is saying exponential but it might be just a straight down trend) That may be a big number still but it makes the whole thing a bit more shakey for me. However, I’m sure that particular problem has been thought through by someone in more detail. The other piece I have not seen is that it seems if you do have an infinite ur universe reproducing infinitely fertile children, that you run into circular child/parent problem that impacts your ur universe as well. If everyone gets the same juice as everyone else so to speak, I no longer understand what could make the ur universe special. I don’t even know if you could tell from a Gods eye point of view if you tried to turn the simulation off because if the number is infinite there’s no reason the simulation has to only be supported by one peer universe so if you turned it off in one there’s no reason not to think it doesn’t just continue on supported by another peer. So to sum it all up: is Bostrum assuming infinity? Or just near infinity? ie very very very big. Thank you for engaging on this. Rocking a child and trying to not lose my mind and until recently didn’t know a place existed to talk about these kinds of things.
IMHO the proper answer to that is that the questions regarding nesting are orthogonal to Bostrom's argument, as it does not make any assumptions of infinity and asserts that this would work given our expectations that with sufficient engineering work it's computationally possible to simulate vast quantities of minds (not necessarily universes - it's enough to simulate your mind and the behavior of the very limited number of things and people you directly interact with) within constraints of our own universe if it's an actual, "full" universe, and thus implicitly also within our "parent" universe if it actually exists. So the argument would be valid even if the max nesting level would be 1 (the "ur universe" can simulate our universe, but we can simulate just tiny fractions of universes) or even 0 (the "ur universe" can simulate the environment directly experienced by me and/or you, but not a full universe like we expect; i.e. it doesn't simulate the physics of the cosmos but just simulates the astronomy articles I've read and the pictures that I've seen).
Pete, I agree with both your points..
You only need enough compute power to simulate one universe. I don’t know why multiple levels are required, let alone infinite levels.
You don’t have to simulate a whole universe, you can cheat, like is done in current video games. Many games show graphics of a larger world, but don’t allow the player to access it.
Maybe the speed of light limitation in our universe is some kind of cheat, to make life easier for the authors of our simulated universe.
Thank you. I’ll need to chew on that for a bit. I’ve always assumed I’m too boring to emulate and much more likely to be some happenstance coincidence in calculating the digits of pi than a standalone. Causal connection may still bug me on that one but I’ll need to think on it more deeply. (ie how much can you get away with not simulating before someone is having to do a lot of weird stuff to make stock prices self consistent for instance.)
The simulations can evolve. If you assume the sim devs are watching us, they might say “oh, they’ve got telescopes now, I guess lights moving around in the sky aren’t going to cut it any more. Next release will have planets you can land a probe on.”
That seems completely doable/possible, so no argument from me especially if they can impact my memory so they don’t ever write themselves into a corner. (Although that does seem like a fun story explanation for why relativity and quantum mechanics don’t reconcile) I guess concern there is that it seems like that also cuts down on the number of simulations you can have. I get you don’t strictly need infinities but it does seem you need a pretty big number before you flip from “possible” to “most likely.” Let me know if I’m beating a dead horse.
Using your example of stock price self-consistency and my example of astrophysics, I'll illustrate some simple cheats to take.
A key observation is that the stock-price self-consistency and behavior of astrophysics is only an issue if the simulated mind(s) care about the details of stock prices. If you're simulating a small set of minds, that won't always be the case.
As suggested by Tom Bushell in another comment, you can start simulating them whenever needed. This may need some extra effort to ensure consistency, but this can be done.
One option is to arbitrarily alter memories and artifacts in order to be consistent with the new simulation - "Oceania had always been at war with Eastasia", the stock prices have always been like this, and galaxy rotation has always behaved like this.
You can also simply nudge the relevant people to not care enough. Whenever they try, distract them with something else so that they don't look into details. For a fictional example, look at how in the Truman Show they protagonist was prevented from traveling outside the town multiple times.
And the final option is simply cut the simulation and re-run it - perhaps after enabling the extra simulation feature that turned out to be necessary, perhaps just with different starting conditions. For most purposes I can imagine, there's no need for the simulation to run forever. Like, perhaps there are ways to uncover the wrongness of *our* universe, but that might simply result in an end of the world if the simulators consider that uncovering details of the simulation disrupt their use case for needing that simulation.
This explains why I have never been to Europe and can’t speak Mandarin.
Since this is the festive season, also for rationalists, I would like to share Händel's praise of the Age of Enlightenment, which should find resonance in this community:
https://www.youtube.com/watch?v=PVCtCxnJyKY
Here is what they sing:
As steals the morn upon the night,
And melts the shades away:
So Truth does Fancy's charm dissolve,
And rising Reason puts to flight
The fumes that did the mind involve,
Restoring intellectual day.
#Handel #Milton #Shakespeare
The recently discussed omission of Xi from the Covid-19 variant names has got me thinking about the kabbalistic implications of using Greek letters to name virus variants. (It is possible that the observation has already been made by ACX or comments and my brain simply decided to co-opt it. If so, please tell me!)
I think the decision of the WHO to use this naming scheme will prove to be extremely foolish and dangerous. To wit: the last letter of the Greek alphabet is Omega.
Omega famously appears in the book of Revelations, e.g. 21:6: "I am Alpha and Omega, the beginning and the end." From my limited understanding, "end" here would mean "end of everything, end of the world". By analogy, it is also used to refer to death, e.g. to mark "Year or date of death" (as per Wikipedia). Do we really want to risk naming a virus variant thus?
Much has been made of the Omicron ("little O") variant, recently. Unsurprisingly, given its name, it seems a bit less deadly. We should imagine that the Omega ("great O") variant will be quite more severe. If one asked an ancient Greek for the factor between "micro" and "mega", I guess they might say perhaps one hundred or so? Science, however, has firmly established that there are twelve orders of magnitude between "micro" and "mega". This is somewhat surprising given that Omega can hardly kill trillions of people, but I think it kinda makes sense if you take the long view: by eradicating humankind, Omega would not only wipe out the present population, but also prevent humans from populating the galaxies, reaching the singularity or what ever else the future might hold in store for us.
I would thus propose that we rename variants of concern using less problematic names. Perhaps the WHO can cut a deal with Disney and use their character names. A Dumbo, Bambi or even Shere Khan variant seems much less likely to cause the death of humankind. As an added bonus, instead of skipping Xi we could just skip Winnie the Pooh instead.
Oh, hey, I know about that Bible passage! "beginning" and "end" is "arche" and "telos": two words which can be used chronologically, but both have a whole interesting spectrum of meaning beyond that, and there are reasons to think that it's not the chronological meaning here. Arche is like source, progenitor, chief, leader... so kind of both "beginning" and "ruler". Often rendered in older English as "head", as in "head of the family" or "headwaters" of a river. Telos, we already use in English pretty straightforwardly in "teleology": purpose or goal, as well as endpoint.
I think theologians generally take "arche" and "telos" here in the non-chronological sense, because in Revelation 22:13, Christ calls Himself alpha and omega, arche and telos AND protos and eschatos. "protos" and "eschatos" are much more straightforwardly chronological, so you'd think then that "arche" and "telos" are meant in a less chronological way.
Yes, much more appropriate to think of the end of a parable (where the lesson is made clear) than the end of the world.
Second the teleological readings of ἄρχη and τέλος. Wanted to comment that ἄρχη is also a reference both to John 1:1 "in the beginning [ἐν ἄρχῃ] was the Word [λόγος]" as well as to Genesis 1:1.
ἄρχη esp in conjunction with λόγος can also mean "principle, that from which the rest is derived"
τέλος makes me curious how much Greek Philosophy had trickled down into Nazareth, since Plato and Aristotle played a great role in investing the term with the meaning you reference above. For the curious, τέλος also will mean "death" and "payment᾽"
Tangentially related:
In early 2020, sales of Corona beer dramatically dropped. I'd assume that for similar reasons, in mid-2021, Delta Air Lines fared noticeably worse than otherwise similar airlines, but I haven't heard of anything like that. Has anybody?
Corona beer was unharmed by corona-chan
https://www.cbsnews.com/news/no-coronas-beer-sales-did-not-suffer-from-the-coronavirus/
Probably because for as many people avoiding it, there were as many wise-asses such as myself who picked up a couple of cases when the office closed, and kept on buying them. (Corona is my usual beer to keep around anyway though)
I remember checking early on in 2020 and seeing Corona stock was actually up noticeably.
Alarming if nominative determinism applies to variants
The hurricane people figured out after 2020 that Greek letters were a bad choice to use for hurricane names when they ran out of names from the list. Hurricanes Eta and Iota got bad enough that they met the criteria for retiring names, and Delta almost did too. (https://en.wikipedia.org/wiki/2020_Atlantic_hurricane_season#Retirement) That's not a good feature for a list of names, especially when so many innocuous companies like Delta Airlines and the Delta Omicron fraternity (https://en.wikipedia.org/wiki/Delta_Omicron) end up sharing their names with these things. On CoVariants (https://covariants.org/), where they track many of the covid variants, some of the 2020 variants they were tracking got the name of generic bird species (pelican, robin, etc.) I can't tell if that's sufficiently better.
Is anyone doing work on building a 'rogue AI detection / containment system'? Would it make sense to do that?
Do AI safety researchers generally assume that a self-improving AI s system would be on our radar before it 'escaped'?
Or would it make senes to have 'honeypots' that exist specifically to detect self-aware systems attempting to manipulate human beings, so that they can be shut down?
I'm not sure such a thing would be possible, at least, not without the support of a major tech company or government. You can't just look around the internet and watch for users buying suspicious amounts of AWS Cloud Computing time, for the same reason that people can't look around the internet and see your credit card number when you send it to Amazon. Detection would have to come from the AI's end (the researchers keeping an eye on it) or from Amazon itself.
A honeypot might be possible, but what sort of bait would you put out that would attract AGIs but not humans?
The only thing I can think of to do in this department would be improving ways to trace digital records to an actual human. E.g., making sure the person signing up for a big chunk of AWS resources is an actual human, and their credit card belongs to an actual person whose identity hasn't been stolen. Which could also be good for fighting fraud in general.
I believe (but I am not sure) that the general expectation is that such a system would not be possible.
Like, there is a bunch of research being done on tracing the intentional opinion manipulation that's done for both international politics and internal politics reasons (IMHO more people are working on that than on AI safety), and some campaigns can be identified but the success in identification and attribution of such manipulation is often very limited and sporadic, and I'm quite certain that a sufficiently smart AI would not be properly detected by those efforts.
And the key practical things that an "escaped" AI would need to do in order to increase its power (obtain large amounts of untraced money, obtain large amounts of arbitrary computing power, anonymously hire some people, anonymously control some legal entities) are things that we know for sure can be done by not-that-sophisticated criminals and in many cases stay undetected for many years, so apparently we are not able to have such things detectable rapidly and reliably.
Like, if an escaped AI with the capability level of a somewhat talented teenage hacker would have engineered one of the many last year's cryptocurrency thefts or ransomware attacks, did some basic laundering (perhaps hiring money mules with "work from home" spam, as human criminals do) and then used the proceeds to hire some lawyers to incorporate "Ainur Homomorphic Encryption Investments Inc" and a dozen other companies, and then rented a few thousand servers on various cloud providers, that would not be detected because, really, there's nothing unusual to detect, it's all routinely done by non-AIs (at least I assume so - perhaps some of the current cybercrime ransomware gangs are run by some AI, how could I know?)
I would say the key direction of AI safety is to work on the "pre-escape" phase - e.g. ensure that the AI does not "want" to escape, or that any people and organizations capable of making such a system get informed enough about the risks to limit their activities and not build something that would escape.
What are peoples' opinion on chiropractors? Pseudo-medicine practiced by charlatans or vital restorative therapy? I hurt my back a couple of weeks ago moving a desk, and it's still rather stiff and sore. My lone previous experience with a chiropractor when I hurt my neck many years ago was not very gratifying, but I'm in rough enough shape to at least consider giving it another shot.
My understanding is that the modern consensus on this issue is that a good chiropractor can provide useful treatment for actual musculoskeletal issues, especially back-pain related ones. The problem is when they claim to be able to do other stuff.
I have several acquaintances who have had positive experiences with chiropractors, but my one experience was meh.
If you suffer from chronic or recurring muscle pain (and even some joint pain is actually caused by muscle issues), I would strongly suggest you look into Myofascial Trigger Point Therapy. Many massage therapists and some chiropractors are familiar with this, but not so many MDs.
I prefer to self treat, using hard rubber balls and a Backknobber tool to do deep tissue massage.
The book I learned it from is https://www.amazon.ca/Trigger-Point-Therapy-Workbook-Self-Treatment/dp/1608824942/ref=asc_df_1608824942/?tag=googleshopc0c-20&linkCode=df0&hvadid=292914176058&hvpos=&hvnetw=g&hvrand=14163074838849093585&hvpone=&hvptwo=&hvqmt=&hvdev=c&hvdvcmdl=&hvlocint=&hvlocphy=1002108&hvtargid=pla-464479570764&psc=1
Had a similar back issue to yours recently, and almost fully recovered after a week of self treatment.
Fellow fan of hard rubber balls for the back here. I learned about using pressure on so-called trigger points from Paul Ingraham's excellent site Pain Science (https://www.painscience.com). Wow does that work well! Feels great to lean into that ball when it's under a trigger point, and back feels better and is looser for hours afterward. Paul describes the feeling of applying pressure to a trigger point as "like scratching an itch you didn't know you had" -- I agree.
Chiropractor practice is an alternative medical practice that doesn't stand up to scruteny. Physical therapy is a mainstream medical practice that... often doesn't stand up to scruteny either. Also, lots of people with mainstream degrees (DPT) hold themselves out as chiros based on the fact that people expect back pain doctors to hold themselves out as such.
Do you know any recreational athletes or lifters? I'd ask them for advice. Then call up those places and ask if they use the biopsychosocial model of pain in their practice. The active are most likely to hurt themselves, and actually care about their outcomes. These are likely to be the most up-to-date and evidence based therapists, and I have a feeling the reader at ACX would be turned off by woo that is very common in this field.
I've had excellent results from physical therapy, but I suspect the quality of the therapist is critical. Quality of the therapist may also be critical in chiropractors, my experience with them is so-so. I think they are better at dealing with symptoms than causes. Physical therapists have given me exercises to do, which when actually done, cured the problem.
I've gone a chiropractor occasionally for 30 years (maybe on average 4X a year.) A good chiropractor can make a massive difference for some problems, but not all chiropractors are that effective, and it's only some problems.
If you have a friend who is fairly active, and strongly recommends a particular chiropractor who they see less than once a month, I think it would be worth trying. (You are looking for a chiropractor who is effective--the ones I consider less effective tend to make people temporarily more comfortable.)
Here is one piece of info I know from life experience, one opinion of chiropractors without much to back it up, and one excellent lead on dealing with your back:
-Info: Hurting your back moving something heavy is very common, and the older you get the likelier it is to happen to you. Unless it is happening to you often or it hurts like hell or it does not seem to be improving, you do not need to seek out a Medical Intervention. You can reduce the chance of further back pain attacks by improving physical fitness and educating yourself about back-sparing ways to move heavy items.
-Opinion: Chiropractic "medicine" is not a science-based treatment approach, and therefore is highly suspect.
-Advice: Go to Paul Ingraham's excellent site Pain Science (https://www.painscience.com) and read his info and advice about back problems. Paul's advice is smart and evidence-based. Also, he's funny and writes well.
Chiropractors are good for adjustments if something is out and causing limited motion or discomfort. I’ve heard you should look for Palmer School trained chiros. But some chiropractors will try to sell you on weird and wacky unproven stuff.
I recently saw a chiropractor for the first time. I'd already had lots of physical therapy, which had reduced lots of pain through strengthening exercises, therapeutic massage, and learning new movement patterns. But I had lingering pain, and I wanted to see what a chiro could do.
The chiro was clear that his approach was not the typical "see me every 6 weeks for the rest of your life." He worked on my back and neck and hips a few times and acknowledged that I don't seem to have structural issues and I should continue to work on strength. He also had a PT assistant on staff who helped me put together a strengthening plan.
My advice: Talk to the chiro on the phone first and see what his or her treatment philosophy is.
As I mentioned on a previous open thread I'm working on a fantasy world strategy simulator and while I have most of the design down and in code/data there are a few things I'm still thinking about.
I'd like to have more detail in the sort of assimilationist practices of historical empires. Rome, Maurya, Macedonia, Carthage, China, and so forth but also stuff like the Norse invasions of England. And maybe something about the larger African states pre-Scramble. Actually the Russians, Dutch, Ummayads, English, Persians, or Ottomans would work as well.
My goal is to have a variety of options that aren't too focused on one historical era and location.
Additionally I'd like to read some good articles on interesting agricultural and hunting practices in different/rare climates. I'm still debating whether to have "discoveries" for farming, hunting, construction, raw resource processing and so forth or just a slow generic "knowledge" increase with little variance over time. If I do discoveries I'd not a lot of techniques and equipment to be discovered for a large variety of climates and processing methods.
Also I definitely need some reading material for classical and medieval trade/intelligence networks for flavor and character action ideas as well as other purposes since Axioms Of Dominion is a DIP strategy game. Diplomacy(foreign affairs), Intrigue(shadowy affairs, Politics(domestic affairs).
You should have a look at Brett Devereaux's series on Rome in particular, and his whole blog in general.
https://acoup.blog/2021/06/11/collections-the-queens-latin-or-who-were-the-romans-part-i-beginnings-and-legends/
Ah this is great. He covers places besides Rome? Greek and Persian would be an interesting contrast from a similar time period. I'd really love to get Carthage and maybe some stuff about Caeser on Gaul. Small foreign populations interfacing effectively with very different native cultures.
I don't believe the blog is paywalled
Devereaux's? It isn't. But it often takes some time for a new reader to dig through the post history.
The table of content pages are really incomplete so it's hard to remember everything. I think you'll find interesting things for your purpose in almost every article. From memory:
1) the series on Europa Universalis and Victoria II are probably what you want to read first for your purpose. "There are no empires in age of empires" is also a good read for the same reasons.
2) Two series on representation of certain cultures in movies/novel/video games : This isn't Sparta and the Fremen mirage (on nomad peoples). I think those definitely discuss quite a bit of interactions between empires and outside populations.
3) Bread, clothes, iron, fortifications.
3) The epic 14-part analysis of the battles in the Lord of the Ring.
4) The 4 or so articles that are analysis of classical text will probably be interesting for your DIP part.
Okay I read a couple dozen posts and a lot of it was pretty useful. The AoE4 part just told me I mostly had key stuff locked down. His CK2-3 and Vicky2 stuff was good. Turns out I read the EU4 stuff a few years ago. He said he doesn't have much to offer outside of Europe sadly.
I did a post discussing implementation of a system pretty simialar to the Socii and Auxilia. Might have to add more info to the how it works in game part.
Okay so far I've gotten through half the queens latin series since I've had to do some other stuff. I definitely want to be able to, more so than something like Imperator, which is currently the best Paradox sim game, represent the socii system as well as potential variations of it. I think that currently there are a few major aspects that aren't possible.
Hi Dirichlet-to-Neumann, are you a researcher in geometric analysis by any chance? I've been writing (well, trying to write) papers on Dirichlet-to-Neumann operators in grad school!
I'm a PhD student only. I do mostly functional analysis and PDEs, and in particular I've studied the DtN operator in relation with the Stokes operator (-\Delta u+\nabla \pi).
That's great! I'm a PhD student myself. You can email me at auk480@psu.edu if you'd like to know about my research, or talk about rationality-adjacent topics, whatever. I work in conformal geometry
I ended up emailing him to ask if he had any good blog recommendations for other areas of world history. But it might have gone to spam.
His blog's been getting a lot of attention lately (~2.8M page views last year up from 1.9 in 2019) so you might just be at the end of a long queue.
I'll also add that if you're a very big fan of his writing I've found his Patreon to be worth donating to. A few bucks a month for an extra post (and a warm glow of satisfaction) each month. The topics are more on how the blog and his academic career are doing, than history per se, YMMV.
Might sign up for that once I finish the game. Currently trying not to get distracted.
Do you have a beta version/source code published somewhere?
I do not. There are screenshots of me doing stuff spread around the web circa 2015, and pictures of map gen results and such. But that isn't very extensive.
The blog posts on substack and the information in the wiki pretty accurately describe, minus visuals, what the game will involve but the AI can't do most things since my programming strategy was to get the design and backend stuff done, code the data structures and file loading, then slowly add stuff to the game and UI for the player and only write the AI once a given aspect was more or less done.
I'd expect to make some videos or something when I am about 80% done, currently my estimate is 40%. So like mid-summer?
Does anyone believe there's actual health benefits to Dry January- to going without alcohol for 30 days? I'm a moderate drinker approaching middle age, so I do have general concerns that Alcohol Is Bad For You, and I'd be happy to go without for a month if I thought there were clear-cut benefits to one's liver or other organs. Are there? I Googled the topic, and of course there are many Pro pieces, but the more scholarly ones seemed skeptical of a concrete improvement to one's physical health in such a short time frame. I'm not interested in going without social drinking for longer than a month. In general I have a very healthy lifestyle otherwise, including fanatically healthy diet, lots of exercise, etc. Any thoughts?
I don't know of any physiological benefits, and I left the Catholic church two decades ago, but I do think they are on to something with the idea of periods of fasting or abstaning from something just for the experience of it. From a purely mental POV (and spiritual too, if you are into that), I do it and find a benefit.
One of the effects of alcohol is that it affects how well you sleep. In particular it disturbs REM sleep. I think this is a pretty instantaneous effect, so you have the effect on days with alcohol, and not on days without alcohol.
I am not sure about long-term effects, but you might sleep better during this month. It might be worth a try for finding out whether you actually feel a difference. You say that you are not interested in changing your long-term behavior, but that might depend on your experience, right?
Should mention that I tried Dry January last year (actually for 5 weeks). Maybe my sleep quality improved marginally, but not enough that it felt significant. I also don't drink alcohol Monday through Thursday now (I do this 12 months a year and have for years), and save social drinking for the weekends- again, maybe there's a marginal change, but just not a huge one.
My health goals for a Dry January would be lower future cancer rates and liver damage. If I can't achieve that with 30 days off- probably not worth it to me
I see. Then it doesn't sound to me like the Dry January would change a lot. I could imagine problems with liver regeneration that accumulate slowly over a long time for people who drink heavily, or people who drink daily. But you don't fall into these categories, so I wouldn't expect that it has additional effects beyond cutting your annual alcohol consumption by 1/12.
All this is speculation from my side, though.
I've never heard anyone claim that 31 consecutive days of abstention from alcohol would have any more direct effect on your long-term health outcomes than an extra 31 days of abstention from alcohol scattered throughout the year. I think the point of Dry January is more to give someone a Schelling point to practice abstention, which can then help the person understand the actual effects of alcohol and abstention on their system, and also help develop the habits that reduce the amount of alcohol they drink at other times.
Sort of like Meatless Mondays.
So the original purpose of memory foam was to protect astronauts during landing by absorbing the impact. Viscosity helped flatten the curve of g-force and prevent any rebound acceleration. But in everyday beds and pillows, viscosity seems to have negative utility. Viscosity makes repositioning less comfortable and makes surface blood flow more obstructed because the material can't deform easily for a single heartbeat. So why are nearly all the foam mattresses/pillows on the market high-viscosity foam aka memory foam?
Short answer is marketing. The long answer is still marketing but also that it is just really hard for individuals not willing to invest in a sleep study and long term testing to find out what works for them.
Many matresses now have a layer of memory foam as only one of the layers on the bed. A common set up is a thick layer of very stiff foam called a support layer, a layer of memory foam called a contour layer, and then a thin layer of very plush latex called a topper layer. If you get the firmnesses right given your weight and curves, you can get a bed that supports back or side sleeping well and is very deformable (and cool, latex is a much better heat conductor than memory foam).
I’m curious to know if anyone is aware of a project that is building a system to stitch together different people’s perspectives on how to think of various important issues. Kind of like a wiki-fact check site. A place to go to read about both the pros and cons of [insert contentious issue here].
Wikipedia talk sections of articles?
I’m hoping to find a novel website/organization that is attempting a grand vision of making it possible to explore debates, decide where you stand, and as one potential feature allow you to indicate the main contributing fact/argument convincing you, so that you can be notified if that sub-fact’s truth or evidence value ever changes
You may be interested in the most recent article on my substack, although it’s not exactly what you’re speaking to here and doesn’t exist. That said, I lay out what a full adjudicative and democratic fact-checking process would involve and I think it may speak to some of the values you’re looking for.
Thanks guy, i’ll check that out.
Kialo?
also hi dad!
I've never written answers on it but I have voted. I've found it somewhat useful before but I feel it could be better laid out
Thanks there Liamsson, i got to learn more of what they’re doing, how it’s going. It looks like what i was thinking! Have you participated on it?
Been having a discussion about the relevance of common school knowledge for technical and non-technical people, and how important various aspects of scientific literacy are, and how to increase rates of scientific literacy.
An argument I expect to be uncontroversial here, is that scientific literacy is low is most especially due to how material is generally introduced, as a constellation of beliefs that don't predict the future, or pay rent. Therefore, any attempt to improve education in general, and scientific literacy in particular, would have to reorient teaching practices further toward processes of how to ask and answer questions eg: "how do I figure out what this material is made of" or "how do I sanity check my mathematical model", and away from trying to remember particular facts.
But there's an issue: the above two questions are pretty niche, even for technical people. Most people are going to remember science and mathematics as a loose constellation of facts, simply by nature of not being scientists, mathematicians, and engineers: they aren't using the question asking methods I describe, so those neural pathways will fall away leaving the constellation of facts behind. So too it goes with scientific literacy.
Using chemistry as an example, most technical non-chemists will have taken several courses in chemistry, but wouldn't won't likely be able to tell you much about how to go about answering basic questions in chemistry like "what's this made of", or what X ingredient in their shampoo might do. So for all the non-chemists, chemistry becomes mostly trivia too.
But, the technical folks WILL retain certain knowledge about things like the scientific method, journal literacy, and how to go about attack technical problems in their domain, and might be able to pick the back up, or at least read the abstract of a paper on chemistry. That ability seems to be what we call scientific literacy.
But if we've just supposed that the pathways to asking and answering technical questions require consistent application to be maintained and used, haven't we implied that scientific literacy is basically a function of how many people in a given society are technical? If that's the case, trying to raise scientific literacy by changing education seems like a red herring, and the main focus was actually just trying to get a larger ratio of technical people in society in the first place.
https://www.lesswrong.com/posts/NMoLJuDJEms7Ku9XS/guessing-the-teacher-s-password
I agree with you that perhaps scientific literacy isn't needed by most people. I wonder if that desire to teach more people the scientific method is so that they'll indirectly support our political opinions :)
A lot of material here and elsewhere in the rationalist community pertain to AI safety. As an outsider and passive consumer of rationalist content, the topic has always come across to me as a little sci-fi-y and virtualistic. What's the best digestible reading or listening that would help me appreciate the seriousness of the issue and risks involved?
Wait But Why's take is a fun read:
https://waitbutwhy.com/2015/01/artificial-intelligence-revolution-1.html
Superintelligence FAQ by Scott himself is a good start IMO: https://www.lesswrong.com/posts/LTtNXM9shNM9AC2mp/superintelligence-faq
Has anyone attempted a scientific paper on the role of opinion in the scientific world?
Is there a science of opinion?
I'm not exactly sure what you mean by "opinion" or what counts as a "scientific paper", but in contemporary analytic philosophy of science, there has been a burgeoning research area investigating the role of individual bias towards or against various scientific theories as a way of ensuring that the scientific community maintains sufficient diversity of opinion to properly explore all the angles that should be explored. Some of this picks up on Thomas Kuhn's ideas in The Structure of Scientific Revolutions, which many people naively read as saying that science is irrational, but others read as saying that individual irrationality leads to community rationality.
You might start with this paper by Philip Kitcher, and many of the others that cite it: https://philpapers.org/rec/KITTDO
I downloaded the paper you suggested. I thank you for the suggestion, and I will read it thoroughly. A precursory reading suggested to me that it represents what I was looking for. What I meant by using the term 'opinion' was unproven but seemingly rational formulation--an ugly way of describing an informed opinion.
It is quite possible for rational people to accept theories that are later reversed, sometimes after having been accepted as true for thousands of years; e.g. that the sun revolved around the earth. We know that some rational people in the distant past held to the subsequently held 'truth' that the earth revolves around the sun, which had been rejected by the majority. From my point of view, earth-centric was a widely held opinion of rational thinkers.
https://www.gla.ac.uk/media/Media_829360_smxx.pdf
New preprint suggesting that unlike earlier strains, which enter the cell directly from the surface, omicron viruses sneak inside endosomes (I think they're pockets of cell membrane that are used to recycle all the protiens on the cell surface) and get in from there.
This means that it doesn't do the thing where multple spike proteins on the viral capsid bind to several cells and make them combine into one big infected blob. Which is apparently what wild-type through delta did. I was happier not knowing about syncytia.
This means that its preference for infecting different types of cells is different from what you see in previous strains (infecting more epithelial cells in the nose and throat, fewer pneumocyces in the lung?). This is consistent which higher infectiousness but lower severity. I am not a biologist.
I just wrote about herpesvirus treatment and prevention here: https://denovo.substack.com/p/herpesvirus-treatment-and-prevention
If you are a woman of childbearing age I highly recommend enrolling in Moderna's CMV vaccine trial, which is now recruiting: https://clinicaltrials.gov/ct2/show/NCT05085366
Is there a good Superforecasting/Rationalist-adjacent style book to investing? I'm thinking something relatively accessible to someone with a decent (micro)economic literacy and knowledge?
Investing advice/books seems oddly polarized (not in a big P politics way).
Depending on how you feel about hardcore value investors, I found Seth Klarman's Margin of Safety to be a nuanced take on managing risk and looking for arbitrage opportunities. Physical copies are basically nonexistent but most restructuring bankers or distressed debt investors will have a pdf somewhere.
With that said, I would be super interested in what others suggest.
This was not one I came across or expected, but very much intrigued and looks to this untrained eye as what I was looking for ("This has caused physical copies of the book to be worth to $500-$2,500 a piece. The high price of the book has resulted in piracy of the book.")
Many thanks!
Much more geared towards professional traders but Alpha Trader by Brent Donnelly is quite good. Expected Returns by Antti Ilmanen is ok as well.
I have a clear recollection of a website which had human longevity interventions sorted by level of evidence, so exercise -> eat well -> ... -> metformin etc. Some cursory Googling is not finding such a thing. Anyone know what I'm thinking of?
Do you know what should come back? Illustrations in books. Not talking comics here but those excellent occasional illustrations you would get in Victorian era novels, like Dickens. Maybe on a kindle or ebook they could move a little, like a gif. Scrooge meets Marley and you see him come in.
I’ll write to Bezos.
When you wrote "Maybe on a kindle or ebook they could move a little, like.."
I was expecting you to say ",like Harry Potter!" as if any self- respecting illustration in a book should move.
The Harry Potter gifs are in the newspaper. Regardless of movement illustrations would be nice again. I’ll write to Bezos.
Light novels have entered the chat.
I'm looking for a book about a war between the USA and USSR in an alternate universe where nuclear weapons were never invented. Does anyone have suggestions?
Not exactly what you're looking for, but I really enjoyed Tom Clancy's Red Storm Rising. Published in the mid-'80s, it was set in the then-near future. NATO and the Warsaw Pact nations wage a non-nuclear WWIII. I read it back in the day - should reread it to see if it's held up well.
Christmas is over, but I recently discovered that there is a cartoon where the Grinch gets psychoanalyzed by the Cat In The Hat. No, really- the CITH gets the Grinch on his unlicensed psychiatrist couch and gets him to stop being a jerk by reminding him of his mother’s love. I thought Scott might find this amusing, though I wouldn’t recommend watching it without liquor. Despite being generally terrible, “The Grinch Grinches the Cat In The Hat”. won an Emmy for animation in 1982.
Can anyone recommend a good biography of Thatcher to me? I'm a Canadian so I know very little about British politics, but she is apparently one of the figureheads of neoliberalism and as a dirty neoliberal I keep hearing people complain about her in ways that make me go "She sounds cool, I should learn more about her."
https://en.wikipedia.org/wiki/Bibliography_of_Margaret_Thatcher might be a starting point.
I think that with a modern figure as polarising as Thatcher, you're going to find that basically all biographies are either Pro or Anti, and need to be read with that in mind, and I'd be very wary of deciding that you support or oppose her on the strength of one side of the story.
If you're after a pro-Thatcher biography, Charle's Moore's authorised one is probably a good starting point; if you're after a more critical one I'm less sure not to recommend.
For what it's worth, I think "she was cool" is probably accurate but misleading - she was a women of exceptional ability, force of will and charisma, with a very strong and clear vision of what she thought was right, who bestrode the politics of her era like a colossus and accomplished an immense amount, most of it bad.
@Scott Alexander What do you think of Dr Ken Gillman's writings on his blog? https://psychotropical.com/
For example, here is his overview of depression: https://psychotropical.com/depressive_illness/
I'd be interested in a critique about his stance on MAOIs: https://psychotropical.com/maois/
Mostly in favor, I've learned a lot from him.
Self reflection: The pandemic has made me question my interpersonal relationships. And it is turning me a lot more libertarian. For context, I am--as I suspect many of us here are--a mid 20s Northeasterner earning an advanced degree in STEMM.
Consider: the pediatric population. Many of my fellow parents are up in arms about whether they should home school and restrict their children's social circle, possibly indefinitely. And these actions are accepted, if not encouraged, by media and public health authorities. Just today, I read an NPR interview with a peds ID doctor in which a father asked if they could let their 15 month old daughter "return to normal life" after she had a vaccine eg. play inside with other children or if they should wait until she was boosted. What the fuck? The hospitalization rate for covid in young children is less than it is for influenza, or RSV, which are both extremely common. If, in 2018, you socially isolated your children for years on end because you're afraid of them catching one of these illnesses, people would rightfully consider you a psychotic child abuser. But now it's considered virtuous. There is total disrespect for the baseline level of risk people accept every day (consider the similar prevalence, patient population, and pathophysiology of Kawasaki's disease and MIS-C, vs. the amount of press coverage and fear inspired).
Ignorant people are --often justifiably--mocked when they think they know more than their physicians, because their attempts to perform their own research and analysis lead to worse outcomes. But the liberal response to this behavior is to proclaim a blind allegiance to people with the right credentials, a complete abdication of critical thinking. Many liberals think they need an M.S. in a subject to form any sort of opinion at all. You don't need to know about the fine details of vaccine development or viral entry to think "gee, maybe instituting a public curfew for months on end is ineffective at containing the virus and a fundamental violation of rights and freedoms".
One big problem with "science" based decision making and governance is that it tends to downplay the importance of factors that aren't easily quantifiable--having a normal social life, having friends, etc. Sometimes the right response to a concern, or potential danger is "who gives a shit"? rather than a factual rebuttal. By far, the most likely health outcome my children--or you, if you're young and healthy--are going to experience from the pandemic is anxiety, depression, or both. Public health authorities encouraging extremely risk averse, neurotic behavior and then having the audacity to champion mental health care is absurd. Public health and media fear mongering is the cause of widespread, preventable suffering.
I think it's telling that "muh freedoms" is a slur against Republican protests for mask mandates/ lockdowns/ etc. In the end, having a free society means the right to say 'fuck you, I don't give a shit'--and this is important. "Well, you don't really neeeeeed it, and it could be harmful" is an excuse to take away every single thing that makes life worth living. Even if you believe personal freedoms should be overrun for true emergencies, it is becoming harder and harder to see covid as one--the pandemic is being conducted on an opt-out basis, and those who justify measures to curtail freedoms to people unwilling to either a) suffer a mild illness or b)protect themselves with a vaccine strike me as pathetic and life-hating.
I broadly agree, but not with everything.
Science can tell us things like "How long is a person likely able to transmit COVID after infection", or "this is our estimate of R_0 of this viral strain". Questions like how do we trade off public safety for personal freedoms, or for economic realities are hard. I wish the CDC would be more upfront about that and not say that the "science says" that we can cut COVID isolation from 10 days to 5 days. Science says nothing about what we should value or how we should make these tradeoffs.
I do agree with you that if, as it appears, Omicron will displace Delta and leave us with a more mild virus that is maybe on par with the flu or a bad cold, that a lot of people have to start changing their minds and reassessing risk, and that has to happen now.
There's not much point in mandating vaccines now since most everyone will have had Omicron, or have been exposed to it, by the time such mandates work through the courts and the teeth start biting. It's moot.
I just got an e-mail from my daughter's school that strikes the correct main notes about the focus on keep in-person learning going, but was a bit frustrated to see that this means continued masking and social distancing, etc., etc. People should be asking administrators and people in charge _at what point_ can we remove the restrictions, and I mean tie it to a real number (and I am no longer am convinced case numbers, or case numbers per 100,000 is the right number since Omicron seems so much milder). If policies need to be implemented because, say, the medical system is collapsing in an area, put a sunset clause on the restriction.
People on the left getting conspicuously angry and people who won't or didn't get vaccinated doesn't help, it won't convince them to do it, and anyway it's moot now. I say this as a person who got vaccinated early and boosted over a month ago. Now with Omicron, it make be appropriate to make a flu analogy. No one cares if anyone else gets their yearly flu shot.
I believe that the larger reason for restricting kids is not the risk to the kids themselves, but them transmitting the disease; so the approprate consideration of impact is not the hospitalization rate of young children but the hospitalization rate of the age group that includes you and other parents (and co-living grandparents) of your kids' circle of contacts.
Anecdotally, almost everyone who had Covid in my circle of acquitances had it "brought in the family" through school (probably not because schools are particularly unsafe but because all the other contacts e.g. work/shopping have been significantly reduced and altered); our schools have frequent mandatory testing (but didn't have vaccines for these ages of kids until very recently) so the tracing was quite obvious but not suitable to prevent infecting the family.
On the other hand, I fully agree with your argument that kids experiencing "normal contact" is important and should be considered as something important and relevant whenever considering the risks and costs of mitigating risks.
Scientific experimentation can give you data points. But then the tricky part is to tell a policy story that fits the data points. This story can vary depending on your values and worldview. Science doesn’t make decisions. Anyone who says it does is trying to manipulate you, or may be well-meaning but pushing a specific agenda. “Science-based” decision making is really story-based decision making, where the story contains a few data points narrated with value judgements based on one’s unique worldview perspective (which can appear totally irrational to someone with a different worldview).
Absolutely. Which is why the expression "I believe in science" drives me crazy--science is true whether you believe in it or not! Science can only be used to predict the consequences of certain decisions, it can't tell you which decisions you *should* make, which is a subjective and moral judgement.
Even if my risk tolerance were much lower--the sloppiness and inconsistency of lockdown/quarantine policy makes me hesitant to accept increased restrictions and government powers.
I always assumed "I believe in science" is some kind of (witting or un) weird Möbius strip joke, like saying "My congregation is the First Church of Atheism (Reformed), we meet to reverently not pray every Sunday at 10am -- join us!"
I believe in not believing things? In empirical skepticism? That's....strange. One wonders whether it's possible to fall into heresy. Start disbelieving in disbelief?
For what it's worth, let me make some observations from a few decades further on. First, it would be a mistake to assume parents are uniform on this subject -- or indeed on nearly any subject touching child welfare, alas[1]. I know parents who have kept their children out of school *for going on two full years* because they are freaked out about something happening to their little darlings in that filthy school with the offspring of God Knows Who running around with unwashed hands and snotty noses.
And there are others who breathed a huge sigh of relief when the schools re-opened, and cringe every time the poor kid has to root around in his backpack to see if he's got a mask he has to wear all freaking day, and the psychological trauma it does to them to be (implicitly) told (since they're old enough to know *their own* risk is small) that they better wear the damn mask lest they be guilty of murdering grandma, or at least someone's grandma. Ridiculous, tragicomic, as you observe.
And then you've got the school board, which is in a painful place, caught between the Scylla of acceptable risk (and lowered expense) favored by many, if not most parents, and the Charybdis of the fearful minority and the plaintfiff lawyers[2] who would *love* to have a cherubic promising 8-year-old in a medically-induced coma in the ICU as a client with which to drain $50 million out of the public coffers and into their Next Christmas With Jeff Bezos In St Barts Fund, not to mention the politicians in the capital who, to a man, want to Make An Important Statement About Our Values with the way in which public education is run -- or at the least want to avoid some horrible blowback if some decision or other goes badly wrong (and 8-year-olds in medically-induced comas are features on the nightly news -- and the nightly news would *love* to run with that story if they could, beats the hell out of some tired old mumbo-jumbo about infrastructure or Chinese currency manipulation or what in hell Joe Biden meant[3] to mumble into the mike).
So everybody who has to make decisions for others is in full CYA mode, while everybody who has to endure decisions made for them by others is in full dudgeon, convinced that Authority is 100% inept and/or corrupt fools, and various Other Tribes of their fellow citizens are basically under the sway of Mordor, not mistaken but so deeply evil that only burning them at the stake and throwing the ashes into a volcano would purge the Earth of their malign influence.
Not a great recipe for collective compromise-oriented rational decision-making in mutual good faith.
-------------
[1] In this corner, the challenger Free Range Parenting, weighing in at a trim 180, while in the other at 220 the reigning champeen If It Saves One Child. My Body My Choice is in the audience, seated among her good friends Parental Consent, Trans Rights, and Religious Exemption, all yelling enthusiastically but in support of whom it's hard to tell.
[2] Shakespeare pointed out the only cure for this nigh on 400 years ago already.
[3] Assuming the verb has meaning in this context.
Oh, I agree with all of what you've written, and I don't mean to imply that such parental risk aversion is universal, or even the majority opinion. I DO think that it drives policy, because it influences the public perception of 'acceptable risk', which, as you've said, makes schools open to lawsuits. I also think it has an unfortunate framing effect making 'moderate' opinions seem reasonable that would otherwise be considered beyond the pale.
Yah. School vouchers, eh? Dunno why that slipped off the conservative platform. I was always hoping it would someday make it into law. I guess at least around here it got headed off by the public-charter-school program, the sneaky rat bastards.
Edit: if you don't have yet have kids that age, wait until you ponder sending them to a much better private or parochial high school and paying *twice* for their education -- once to the people actually providing it, and via your taxes to the state for the public HS education you've declined. Bah.
Not school age yet, but our closest public school has a math proficiency rate of 4%....so I see this choice in my future.
At this point a fair number of states have voucher programs, although generally with some restriction in who can use them and with amounts well below the per pupil cost of the public schools.
Really? I had no idea. It was all the rage 20-30 years ago in conservative circles, I always assumed it got eclipsed by social con or neocon bullshit, and not that it was actually achieved to any real degree.
Just my luck that California never had a prayer of being one of them. I can't even count on the benefits of our public university system, Berkeley is not nearly the bargain it was when I was there :(
Incidentally, I think this is the way the rest of the liberal world often views the rationality community. I think that in fact, Effective Altruism is generally getting things better at both the quantified and unquantified effects of charity than the normie opinion, but many people say "One big problem with "science" based decision making and governance is that it tends to downplay the importance of factors that aren't easily quantifiable".
I don't know how to ensure that one is doing this in the right sort of ways and avoiding the bad sorts of ways, but I do think it's helpful to step back from one's frustration with a particular community applying a particular pseudo-scientific lens to a problem, since that is also what we all here are doing in one form or another.
It must be said, that The Last Psychiatrist's "book about porn" has finally been published since first being mentioned in 2013:
_Sadly, Porn_ https://www.amazon.com/dp/1734460822/ref=cm_sw_r_tw_dp_NER175AFT7Q2HX89ZH1R
Thanks for this! I've been very curious about this book, and I've been assuming it was cancelled after the doxxing incident that made TLP stop blogging.
Can anyone give a bit more info about the book, or link to a review that does? The only review at Amazon is by someone who gave it 5 stars, but was not what you'd call articulate about why they did: "The nuggets, big and small. The imperfection. And often, the void I find staring back. . . . the sweat that went into it, finding the time to do it + drinking, the freewheeling ride, the ideas, the brilliant cover, and the Christmas gift release."
There were nuggets. And their size varied, OK, got that. Um . . .
I don't know if there are any reviews of the book, but here's TLP's blog post about what it is (which is still rather vague about everything that comes *after* the porn story): https://thelastpsychiatrist.com/2013/07/still_alive.html
Data Secrets Lox has a thread about it that goes into some detail.
https://www.datasecretslox.com/index.php/topic,5509.0.html
Last week there was much enthusiasm here over the idea of plastic-consuming beetles. I'm ignorant and confused why the existence of plastic is bad. It's non-biodegradable. If you believe man-made climate change is bad, isn't that good? Wouldn't plastic eating beetles speed up the process of releasing carbon absorbed by algae millions of years ago into the atmosphere? Plastic continues to trap the carbon whereas these beetles would release it. Why should I want these beetles?
That's true that there's a pair of competing goals there. I think the main benefit of the plastic-consuming beetles would be to clean up bits of plastic out in the wild in a decentralized massively-parallel way, so that less plastic ends up in the oceans and other waterways as micro plastic pieces that will be very very difficult to clean up
My understanding of the Carboniferous period is that it is a period in Earth's history when plants discovered how to make lignin, but nothing on Earth knew how to digest it. Thus, forests ended up covered in a pile of bark that just didn't break down, and this hugely disrupted ecosystems and led to lots of death and destruction, which promptly changed a few tens of millions of years later when some microorganisms figured out how to break it down. This ended up locking up lots of CO2 in coal formations, that we have recently started breaking down again. (https://en.wikipedia.org/wiki/Carboniferous#Rocks_and_coal)
I don't know whether the amount of plastic we are producing would have significant atmospheric impact if broken down, or whether breaking down the plastic that is accumulating in various parts of the wild would have significant ecological benefits. But the Earth has been through *something* like this before.
Why are so many great contemporary writers Marxists? To name three: our Freddie, Chris Wickham, Justin E.H. Smith.
My best guess is that the fields of Economics and Humanities no longer speak the same language and therefore can't communicate. I believe the above Marxists named are every bit as intelligent and intellectually honest as most top-notch economists, but the language of the respective fields they are in aren't compatible for constructive dialogue.
In agreement with an observation made by Thomas Friedman further below, because it's easier to be engaging and entertaining if you are less strongly moored to objective reality. Marxism has always been a tremendously hopeful and optimistic moralizing story about the destiny of mankind: someday we shall throw off the shackles of our ignorance of the true value of things, eject the parasites, and live anew in Paradise, with plenty for all. Indeed, it's an inevitable consequence of the march of history. Hurray!
By contrast, the traditional economists are a dismal lot, always telling us they can't predict when to buy and when to sell, or even whether the inflation is "transitory" or not -- but by gum they *can* tell us TANSTAAFL and wage-and-price controls won't work, the real minimum wage is $0/hour, the actual always greatly exceeds the projected revenue from some new tax vice dynamic effects, damn them, and so forth. A load of picayune bad news without *any* glorious uplifting narrative at all. Bah.
Somewhat related: I've always thought Paul Krugman could have substantially more impact as a NYT columnist is he had more than a vestigial sense of humor. Jeez, Paul, crank it up!
"Tremendously hopeful and optimistic" are really not the first words that come to my mind to describe Freddie DeBoer...
Hmm, I think of Marxists as more committedly pessimistic and traditional economists as more committedly optimistic.
I'm curious what Scott, and the community, thinks is going on when people "are hypnotized". I seem to be one of the "not hypnotizeable" people, though of course there are plenty of nonmaterial ways to get me to change my mental state: show me a horror movie, tell me an erotic story, etc., so I can believe, to some degree, that a practiced practitioner could get me into a somewhat unusual mental state.
But what's going on when, say, someone is supposedly hypnotized into thinking they're Jimi Hendrix, or a cat, or that their hand is itchy or whatever? I think that social compliance--the pressure to think in a way that will be in accord with group expectations--is unquestionably powerful, so when I hear about a college orientation where a freshman was hypnotized into acting like a cat in front of his peers, I suspect that this is mostly his being game to please the crowd, even if he doesn't know that consciously; if the audience were half-asleep senior citizens he didn't know, he wouldn't be pulled so hard into the role. My hypothesis is that those of us who are not hypnotizable have just as much capacity to get into roles, we're just a bit more aware of those pressures and tend to retreat to first order thinking when we sense those pressures at work, instead of leaning into them.
I have no doubt there exist people who get up and imitate cats entirely out of social compliance or a even just a desire to be the center of attention -- and also that that there are people who are highly socially compliant and therefore lean harder into the hypnotic suggestions, with the result that they actually do experience some alternation of consciousness as a result -- they "get into the role," as you put it.
But there is definitely another relevant variable besides social compliance: ability to experience intense imaginative involvement in something that's not real. Hypnotizability correlates with scores on the Absorption Scale, developed by Tellegen. It has subjects rate how completely they get lost in daydreams, books, movies, etc. All of those are situations where there is no social compliance factor determining how hard one leans into imaginative involvement -- and yet there's a lot of variability in how far in people lean.
Absorption scores decline steeply with age, by the way, as shown both by results on the Absorption scale and most people's life experience. When at age 8 I pretended to be a horse I at least 50% believed I was a horse, and had brief moments when I 100% believed it. In my teens I'd get so absorbed in books that calling my name did not penetrate my attention at all -- somebody had to touch my shoulder or wave their hand in my face. There's much less of that kind of thing in my life now.
Still, I think the situation in which one is being hypnotized can make a big difference in one's experience. I had training in performing hypnosis, but never was able to feel much of anything during the many practice sessions in which members of the class hypnotized each other. What got in the way for me was the social awkwardness of the situation. Everything my partner said when in the hypnotist role pulled for me to do or feel a certain thing, and I knew that if I did not then my partner would be in the position of failing to pull off a successful hypnotic induction. Yet I was very uncomfortable with the option of lying or faking it. My preoccupation with this dilemma blocked my capacity for imaginative involvement in the hypnotist's suggestions. But a few years later I was sitting in on a treatment group run by a colleague, and he did a hypnotic induction of the group. Nobody present, including my colleague, cared how well I responded to the induction, and under those circumstances the guy hypnotized the daylights out of me. I knew I was being hypnotized and going along with suggestions, but in the process of going along with the suggestions I experienced intense imaginative involvement, on the order of what I experienced before I was a teen.
Thanks for this absorption framing. I too remember imagination as a kid that felt almost real; I distinctly remember telling my big sister, with wholehearted belief, that my teddy bear had a headache. And of course, dreams prove that reality is shiftable for me!
I had an experience years ago that's also relevant: at Esalen, I was part of a group that tried doing aura reading. At first only a few people claimed to see something unusual, but slowly everyone got on board until I was the only holdout. Obviously, auras are entirely a psychological phenomenon (aided by the ease of producing vision aftereffects in entirely natural ways), and ever since I have wondered what, precisely, was the internal experience of these friends of mine. I've felt pressure to say I saw them and seriously considered joining in, but I would have had to consciously lie to do so. My best guess is that few other people in the group were consciously lying; I think the distinction between lying and telling the truth just isn't as meaningful for most people as it is for me. A cynical way of looking at this is that most people are in a constant state of lying, and almost never sacrifice the slightest bit of convenience or social comfort for the sake of honesty.
All of which is to say, people who act deeply out of character under hypnosis and report no memory of the experience may have psychologies in which they don't carefully distinguish between reality and fantasy in the first place. If there is no reality you have fidelity to, then you're walking into the hypnosis session already, essentially, hypnotized.
I am also very difficult to hypnotize, so I never had this experience from inside. But when I asked others about their experience, it seemed like a "kinda yes, kinda no" situation.
For example, someone was hypnotized to think they were 10 years old. Afterwards the person told me that in some sense it was a pretense/compliance; that they were quite aware that they were actually just an adult person pretending to be a child. But at the same time, it was a much stronger experience than mere pretending/compliance would achieve -- for example they reported having felt very strong emotions; also acting like a child was very natural for them in the moment.
A person only consciously role-playing a child to comply with the instructions would need to continuously pay attention to "what exactly am I supposed to do now", instead of just acting naturally; and they wouldn't feel strong emotions (unless they consciously decided that this is what they should do, and then somehow manipulated themselves into feeling it). So clearly there is something beyond conscious compliance.
My guess is that part of it is you consciously playing along (or at least not fighting against it), but another part is that some parts of your brain that are not under your direct conscious control are also playing along. That is, hypnosis can recruit some parts of your unconsciousness that mere pretending would not.
When I've acted in plays, I've experienced something similar; when my character was angry, I really was angry, even though I knew I was playing a part. It might be that hypnotism mostly consists of creating a situation where immersion in a role is socially acceptable, as well as expected, for people who aren't used to blurring those lines; and having had them blurred, they might have less of a sophisticated understanding of their own experience than a trained actor who can get legitimately weepy on command, but who knows the sadness isn't "real". I wonder what experiments you might devise to test this theory; you might rate people on hypnotic immersibility, but then put them in a situation where reality and fantasy would call for different responses, say making them think there's a mousetrap in a box IRL, and then under hypnosis telling them there's a cookie in the box instead, and measuring their responses. I would guess that, adjusting for apparent immersibility, actors would hesitate more.
Gratuitous recommendation time, if you're frustrated about not being able to find water filters (due to FDA/labeling issues apparently?!? https://www.reddit.com/r/Frugal/comments/orq45p/does_anyone_know_what_is_going_on_with_pur_water/) I've been quite pleased with the Berkey water system and used this model extensively for the past 4 months: https://www.berkeyfilters.com/products/big-berkey
It's a bit of an investment, ~$350 with shipping, but that's less insane considering what resellers were asking for: https://www.amazon.com/PUR-MAXION-Replacement-Pitcher-Filter/dp/B07CFX1T6D/
I was concerned after a couple months that the flow was slowing but after cleaning off some mineral buildup on the filter elements it's back to several gallons a day, much more than we need for drinking and cooking purposes
Does anyone have actual numbers on how much worse rapid antigen tests are at catching omicron? I've seen a lot of headlines that say "less sensitive", but not a lot of comparative confusion matrices.
Look at Michael Mina's posts on Twitter. I believe TheZvi also covered this recently.
So I think I want to take a shot at the NMN/NR supplements, but the entire field of supplement marketing looks like a howling wilderness populated entirely by predators greased in snake oil.
Scott's "20 bucks a bottle" comment in the lifespan post may be situating my estimate a bit, but the prices I am looking at seem really high (honestly even 20 bucks a bottle seems high if it's 30x300mg NR, given the clinical trials were testing at 1000 mg daily). What I have seen instead is 6 bucks a day for NR alone and 3 ish for NMN, plus tax and shipping. I am not sure $3650 per year for possibly useless pills is wise..
Does anyone have a suggestion on where I can get this stuff cheaply (read: about 1.50 per day for 1000 mg doses). Happy to look at Indian/Chinese sources if someone knows a reliable source.
AllDayChemist (Indian, I think) is reliable, but last time I looked they did not sell supplements.
I'll bet there's a Reddit sub with people knowledgeable about this. Try googling "Reddit online foreign pharmacies" & see what shows up.
I don't understand how internet protocols, packaging, IP addresses, routing, networks, and all of that stuff works. I find it unusually difficult to gain this knowledge from Wikipedia because it feels like every article is both too vague and related to about a dozen others. Is there a good book about this? ( I do have a background in cs and math)
The effortpost series on DSL is a great starting point: How Does the Internet Actually Work (in great detail).
Part 1--LAN Traffic
https://www.datasecretslox.com/index.php/topic,2570.msg74876.html#msg74876
Part 2--ISPs and Routing
https://www.datasecretslox.com/index.php/topic,2708.msg80251.html#msg80251
Part 3--DNS Resolution
https://www.datasecretslox.com/index.php/topic,3005.msg91343.html#msg91343
With 2021 over, we can now assess several Metaculus predictions that were featured in the "Mantic Monday 7/26" post.
https://astralcodexten.substack.com/p/mantic-monday-726
"Will the US have more than 200,000 daily COVID-19 cases (7-day rolling average) before January 1, 2022?
This question resolves positively if at any point between 2021-07-01 to 2022-01-01 the 7-day rolling average of confirmed COVID-19 cases is greater than 200,000. The source will be CDC's official count of Coronavirus cases, unless Metaculus Admins determine there is a significantly superior source of data."
Result: The U.S. surpassed 200,000 new daily cases on December 24 and has stayed above that level.
"Will the US have more than 100,000 new daily COVID-19 cases before January 1, 2022?
This question resolves positively if on any single day between 2021-07-01 to 2022-01-01 there are more than 100,000 COVID-19 cases recorded. This question resolves negatively if there is no single day the United States records more than 100,000 daily COVID-19 cases according to the CDC's official count of Coronavirus cases."
Result: Milestone was passed on August 5.
"Will the US have more than 1000 daily COVID-19 deaths (7-day rolling average) before 1 January 2022?
This question resolves positively if at any point between 20 July 2021 and 1 January 2022 the 7-day rolling average of confirmed COVID-19 deaths is greater than 1000. The source will be CDC's official count of COVID-19 deaths. Make sure the "Daily Deaths" view is selected."
Result: Milestone was passed on August 18.
"Will EA Global London 2021 be cancelled, rescheduled, or moved online again?
The question resolves negatively if the EA Global London 2021 takes place in the originally scheduled physical location (London UK) at the scheduled dates (29-31 October 2021)."
Result: The conference ended up being held on time and in-person.
If anyone with covid is having trouble getting a fluvoxamine prescription, you can sign up for a 50-50 shot at it through the Activ-6 clinical trial (recommended by 'mj robinson' in another post's comment section). That trial has two arms for additional drugs but you can opt out of those. Covid out is another trial with 5 different treatment groups. Activ-6 pays participants $100 and covid out pays $400.
https://activ6study.org/
https://covidout.umn.edu/
I've seen several references from Scott and others to evidence that the actual effects of parenting are pretty minor relative to other factors. Does anyone know a good overview of the evidence on this that they can point me to?
_The Nurture Assumption_ by Judith Harris is the book that originally made a convincing argument along these lines.
The main argument of The Nurture Assumption is that a child’s age peers have a stronger effect on socialization than the parents. She doesn’t say parents have no effect on their children’s life outcomes.
The book has received a lot criticism from Harris’s own peers.
Sorry for asking, but can someone help me with where i can find some study for the Covid Vaccin? if its safe or not?