I trust you have a different but related market to suggest instead. Or a proposed resolution criteria, by which one might create yet another market. It certainly seems clear that this particular market was anointed as the main market by luck rather than some other factor, so the existence of a better market, released later and with more thought, would not be surprising.
I think it's fair to point out a problem with a market's resolution criteria without being able to pose better ones. For one thing, it might be the case that the underlying question is incapable of translation into precise resolution criteria, which would demonstrate a problem with this use of prediction markets.
In this case, the original criterion was "I will resolve it to my best guess of the median informed observers' subjective probability based on all reliable sources". On 18 October 2023, the creator gave some clarification.
I personally don't think it's a problem that the resolution relies on Milford Hammerschmidt's assessment of the evidence. Market participants can form a view on whether Mr Hammerschmidt is a fair judge. More of a problem is that (a) it resolves based on informed observers' "subjective probability", presumably meaning which outcome seems more likely, rather than the ground reality and (b) there is no indication of when this assessment will be made.
The fundamental problem is the same as the Wuhan case: how will we ever know for sure what happened?
We can look at a few other markets on the same topic. This market https://manifold.markets/Joshua/will-the-alahli-baptist-hospital-ex is based on what Wikipedia will say within 30 days, which seems reasonable (provided we trust the creator's interpretation of Wikipedia), but gives the boring answer that Wikipedia will probably be ambiguous.
This market https://manifold.markets/MarcusAbramovitch/will-the-new-york-times-definitivel asks whether the NYT will definitively state Israel was not behind the blast, which according to the market is very unlikely. This is perhaps unsurprising, but it seems worth noting that changing the resolution criterion this way essentially reverses the answer.
Yes, this stuff will always be awkward. And multiple adjacent-to-the-real-question markets is often a good answer to getting a handle on what's "actually" going on. Personally I like a mix of some markets that are best effort at objectivity and clear definitions (but usually aren't exactly the actual question of interest), combined with a couple that are more subjective but closer to asking exactly what people want to know. And then we get a sense of both what's going on, and where the confusion lies, which enables excellent analysis of the sort you're doing in your comment.
I liked the "resolves to Wikipedia" approach as part of that, and used it for a pair of mine on the death toll, as reported by English and Arabic Wikipedia. I think the result is an interesting blend of "objectivity" that highlights the way sources matter.
I also agree that criticizing a market without being able to suggest improvements is fair game. Definitions are hard, and it's entirely possible to make things worse by trying to address more edge cases in advance. (Especially since clear and concise writing is very valuable here!)
But there's a huge range between these sorts of critiques, or pointing out minor issues and edge cases, and calling it "bullshit". If you're calling it "bullshit", I think you should be able to point out _some_ room for improvement.
You can absolutely call bullshit without needing to point to an improvement. Some questions are simply intractable from the position of epistemic certainty. I acknowledge this is not an equivalent issue, but for example if you made *any* sort of resolution criteria for "The existence of a creator God" they would all be bullshit
I think this case is a bit easier than "does god exist". And in fact I think that difference makes a difference; I would not have posted my original reply in that context.
Yeah, I'm extremely disappointed by this post's lack of mention of it. US government statements were used as a resolution source, but whether the US government is honest on the topic is itself a major controversy! Perhaps there should be a blanket rule that resolution source must be mentioned in the market name, to avoid huge discrepancies like this one.
I would argue that any likely resolution criteria on the hospital bombing is likely bullshit.
If you have a prediction market on LK-99 being a super conductor, chances are great that eventually a consensus answer shared by (100%-lizardman constant) of the population will eventually form.
If you have a question on the existence or gender identity of god, chances are good that you won't have a super-majority opinion you can use for resolution.
The hospital question could be restated at "will the author of the question believe that the probability of Israel having caused the explosion is more than 50%?"
So the question does not track the first order probability at all. If an oracle tells me that the at the resolution date, the poser of the question will estimate the chances that the IDF was responsible at precisely 47%, that does not mean I should buy yes up to 47%, but that I should sell yes down to 0%. (This could be prevented if they pre-committed to resolving the question by randomly sampling their best guess probability distribution. The downside is that this would be even more likely to create a feedback loop. Should the yes probability be 10% or 15%? Hard to decide for an individual. Luckily, there is a prediction market out there, let's just take their probability. Oh wait.)
The other problem is that a lot hinges on the market prediction of the judgement of the questioner. If the name of the questioner was clearly Jewish/Arab/Farsi (and they optionally changed their icon to a flag of Israel/Palestine/Iran), this would greatly affect the level on which yes is trading.
On the one hand, such prediction markets are clearly useful to update on emerging information. On the other hand, I don't feel that the absolute magnitude they are trading at is very useful.
What do the holders of yes (7%) bet on, anyhow? That there is a 10% chance that evidence will emerge which pushes the probability of "IDF did it" above 50%? That the creator of the market is giving more weight to Arabian media then other people assume?
I am the author of the market in question. When people on manifold say "resolve to a probability", they mean exactly that - they resolve it (e.g.) "at" 47% and YES shares become worth 0.47c. So such a resolution would avoid this failure mode.
Do you have any evidence about the mutual acceptability of market resolution in controversial cases like the hospital car park explosion? Like, how did the many COVID-19 origins markets resolve?
At some point these markets are really asking "how will the platform/host resolve this market", which may have dramatically different odds from "what is reality". I guess that's banal at this point but how much does it matter after all?
Great post! For those not on Manifold, I think it's also worth directing attention to the biggest market on the site, the main Speaker of The House Market:
"I can’t find any markets on the Middle East topic I’m actually interested in, which is Israel’s medium-term plan. Will they kill some Hamas leaders, then get out? Install a puppet government? Permanently occupy Gaza like they’re occupying the West Bank? These all seem like bad options, but they’re very different bad options, and I haven’t seen much speculation about which is most likely."
Does Israel have ANY non-bad options? I think the answer is unfortunately no...
Oh wow, that IS big news! That was the last one, I think.
Anyway, as regards Israel Palestine, Israel could occuppy Gaza and start to rebuild it. The locals will hate it. The PA can give some rousing speeches about the Iaraeli occupation at the UN, the US president can apply pressure, and 5 years down the line Israel will "cave in to the demands" and let the PA take control. The PA regains prestige, and Nobel peace prizes for everybody.
I like this, but I think the feasibility is complicated by every political leader's desire to appear the victor to their public. Israel "caving in" looks good for PA, bad for the reigning Israeli PM. Of course, "I'm the PM who brought peace" is a win, but only for some sectors in the Israeli public, and not all Israeli politicians could pull it off.
Just spitballing, what would it take to create a new PA-like entity that governs Gaza only? A new government that has nonzero local legitimacy but isn't Hamas and explicitly bans anyone with ties to Hamas, and has some buy-in (even if tacit) from both Israel and the Arab League.
The only reason most people still turn up to work for the PA is because it's so difficult to find alternative decent jobs. Palestinians really do not like the PA and there is significant territory that the PA can't work in for fear of being killed by their own people.
Deport & annex half or a third of Gaza. Less space to live in would probably slow Gaza's demographic growth (which is the big long term problem), would make a bit less israeli territory vulnerable to rockets (by pushing their launching point a couple dozen km south), and make Hamas operate in a smaller urban area, probably making it harder to hide or something.
It would burn a lot of international sympathy & political capital, but these are only as valuable as the strategic goals you aim to achieve. And a step toward the "one israeli jew state" solution is quite valuable.
Any solution involving "destroy/oust Hamas from the strip" begs the question "how?". Removing an urban insurgency -one with at least 20-30% support in the population, from the most conservative estimates i've seen- makes "displacing a million people" looks like the easier solution.
No, except from fringe extremists. But realistically, what other end state is there? Any 2 (or 3) states solution would just keep the conflict going (until they run out of luck/talent and the next arab coalition finally win a war). Two peoples don't cohabitate a land, not without a new culture emerging and fusing the two together (and -correct me if i'm wrong, that's a point on which i'm just running on gut feeling- it's really not the current zeitgeist of either jewish israelis or arabs.
>Wouldn’t it mean deportations of existing Israeli Arabs?
They're, what, 2 millions out of the 9m israelis? That's a sizable minority, but at this point, not too much out of line for a western countrie, but tripling that number by integrating Gaza & the west bank into a "one ecumenical state" would turn Israel into an arab majority democracy, and let's be real, nearly no Israeli jew want that.
I really don't see what else that leaves as the long term state. Either one people go or the other, and in 2023, only one of them is in a position to push the other in that direction. It don't even need to be a willing effort. The fringe elements in Israel will continue slowly pushing on the west bank, and in Gaza, Hamas actions will inevitably force Israel to *do something*, and that *something* will never be "give them more leeway"
I have no doubt that you’re sincere in your stated belief, but I feel like that’s the opinion the crazy fringe elements (on both sides) want you to have.
Consider that Hamas started this war because it was concerned Israel was approaching a rapprochement with Saudi Arabia; an unprecedented step towards a lasting peace in the region.
I'd say the recognition by Egypt was more important, but that was a long time ago. But yes, rapprochement with SA would be a great step for Israel, and for interstate peace in the region. I'm not convinced it would help Palestinians in many ways.
> Less space to live in would probably slow Gaza's demographic growth (which is the big long term problem),
I don't agree. "The big long term problem" is the local culture. Even the Gazans who don't support Hamas still support Hamas's general goal of genocidal antisemitism. They are taught as children that the land belongs to them "from the river to the sea," (aka everything between the Jordan River and the Mediterranean, aka the entirety of Israel,) that Israel has no right to exist and that it is a virtuous act to kill Jews.
The big long term problem is that Israelis and Palestinians *can not coexist peacefully,* because the Palestinians have ruled out the possibility, in very absolute terms, since the very beginning. And that's a pretty big long term problem!
I feel we're saying the same thing. This culture is a long term problem for Israel because their demographic growth is high. If Gaza was slowly losing inhabitants, then it'd be a temporary threat that can be contained until it stops being relevant.
Could we please, please stop openly advocating genocide, or maybe take breaks from it one or two days per week? Historians will not use different rules to study enemies of the west than they do unaligned tribes in Central Africa the way we do.
There are orders of magnitude more Arabs than the population of Gaza; no more Arabs in Gaza - although arguably an ethnic cleansing - is hardly a genocide.
That seems like an over-generalization. I have heard anecdotes of Israeli Palestinians who can live within Israel for decades without killing any Jews.
While genocidal antisemitism is one big driver of the conflict, I feel that Israeli settler activity in the West Bank is also a driver of the conflict.
The West Bank? So... places like Bethlehem, the City of David, which is now claimed for exclusive use by the sworn enemies of the People of David, who are taught to deny that Jews ever had any historical presence in the land, even in the face of all the evidence? ( https://www.youtube.com/watch?v=mkLs-yOi9gQ )
No, by any rational, objective standard, the illegitimate "settlers" and "colonizers" here are and always have been the Palestinians.
It is astounding how Bibi's missiles have blown up the wide safety perimeter established over generations of activity by dedicated organizations like the ADL around subjects like relativization of the holocaust, leading to the reestablishment over a few days of a complex boundary, negotiated individually, that because of overton window dynamics and human stochasticity will be continually crossed over.
I think that the main the Nazi ideology was to a large degree refuted by thee defeat of their "1000 year empire" at the hands of Slavic "Untermenschen". Their propaganda was not big on preparing the population for resistance during occupation. For the leading Nazis defeat was to horrible to imagine. (Werwolf was more of a plan to have partisans in occupied territory while there still was a front rather than a long term plan to establish a fourth Reich after total defeat.) (Also, it is not like many Germans became philo-semitic democrats when they surrendered. That took a few generations.)
By contrast, Palestinians have experienced military defeat multiple times. Any ideology thriving in such circumstances will not be refuted by yet another defeat.
> Q: How long did it take after Germany's defeat for the culture to shift away from genocidal death cult?
More relevant Q: how long had Germany *been* a genocidal death cult before their defeat?
A: just a few short years. Not "throughout the entire lifetime of of every citizen and all of their ancestors for centuries." That kind of changes the calculus.
Look, I am sorry, but the time of David was some time ago. From what I recall, a diaspora happened, and other people settled there.
I generally do not support a policy of allowing descendants of ancient civilizations an automatic claim to the lands they once held. If the Greeks (or Macedonians) were making a claim for Alexandria, or First Nations one for Manhattan, or some tribe which suffered defeat by Israel in biblical times made a claim for their original homeland which now happens to be located within modern Israel, or Austria made claims on Hungary or South Tirol, or Rome made a claim on London, I would be equally opposed to all of these.
I support Israels right to exist not because of any claims in biblical times, but because I consider basically all borders to be historical accidents and generally prefer that arbitrary status quo to attempts to shift borders, which tend to cause a lot of suffering.
From my understanding, Rabin signed the Oslo Accords, which ceded Gaza and the West Bank to the Palestinians. If the Palestinians prefer to claim that there never were ancient Jews in the West Bank, that is unfortunate, but not a reason to take their land.
No. With the notable exception of Jerusalem, which remained significant for religious reasons, most of Israel became barren, inhospitable desert after the diaspora. Very few people wanted to live there, even among the local Arabs who were used to barren, inhospitable deserts. When the Zionist movement of the late 19th century began to draw Jews back to their ancestral homeland, the Ottoman Empire (nominal ruler of those parts) didn't want them coming in, but the Jews had little trouble settling anyway because *there simply wasn't much of an Ottoman presence around to stop them.*
It wasn't until after the Zionists began to reclaim the land and make it an attractive to live again, making "the desert blossom as the rose," that the neighbors decided they'd like a piece of what the Jews had built. And the rest is history. Long, ugly history that led directly to the atrocities of the past month.
I was under the impression that rockets from Hamas can hit Tel Aviv , which is quite a bit (50km?) away from the strip. Making Gaza smaller seems unlikely to help there much.
> probably slow Gaza's demographic growth
That is a very polite way to phrase "conditions so hellish that humans can't thrive".
The solution you advocate for could reasonably be called an ethnic cleansing.
I would prefer for Israel just to occupy Gaza and run it, and make a better job of it than Hamas. Go full surveillance state, kill any insurgents. Try to raise their GDP to a level where the population growth is less of a problem. If in a generation or three, inhabitants should begin to favor peaceful coexistence, give them their own state.
Not an expert on Palestinian rocketry, but I believe by far the most common rockets are semi-homemade and mostly have a max range of <10km. Anything that can hit Tel Aviv is primarily manufactured in Iran and available in much more limited supply.
I also THINK Iron Dome is going to have an easier time dealing with the latter (especially given the more limited numbers).
I don’t think there are any good short term options. Short term they need to not burn too many bridges ie respect international humanitarian treaties and help civilians survive/escape. Medium term help to set up infrastructure. Long term get out and cross their fingers.
Jonathan Anomaly's talk was mostly laying out the basics of polygenic screening. Prediction markets factor in because we want to know which diseases will likely be cured in the coming decades. Genomic Prediction creates an embryo health score to maximize DALYs for the embryo. But a lot of diseases only reduce the quality of life when someone is much later in life. If we think a disease will be cured, we should put less weight on it. I would put stronger weight on traits we likely will not be able to cure such as low mood, poor personality traits, and cognitive ability.
Among the poor personality traits we are not likely to cure are low empathy and a tendency to violence. Often those 2 occur together, because if other people's feelings are not real to you it's not that big a deal to harm them. I think that's the most important target to select for genetically. People here often believe intelligence is more important to select for, and then add that smarter people will be less violent because they will be able to understand that it's morally wrong and often not productive. I don't agree. There are lots of smart people here, and little agreement about what's the right thing to do about this and that. I think that what makes people more reasonable is a sense of connection with other people, and a relatively low level of rage, rather than the ability to reason about ethics and the logic of arguments.
1) Markets resolve "no" when someone presses the "reject" button. So every market is implicitly "Will neither subject press the reject button, and will they date for 6 months?"
2) Since Scott checked the site out, they've added an easier way to see who has bet yes or no on a match. You can also see the full trade history by opening the match's market and looking at the "trades" tab like a normal market.
3) There is currently ongoing debate as to whether the site should focus more on 6 month relationships or 1st/2nd/3rd dates. If you have an opinion, come bet on it here:
While I agree, I think "well enough paired to make it to the 3rd date" is much easier for an external matchmaker to assess, and resolves faster, so might be a much better market criterion.
The markets on whether the two people will date for six months are indeed conditional on whether they reach a decision (either party can reject, but both have to confirm they dated for 6 months to resolve Yes). The markets either don't resolve (depending on loans to get your mana back) or resolve N/A after a long time, like when you get married.
Also, you can now see the positions of who bet on a relationship!
Super excited to see where this goes as we've literally just launched a few days ago.
I don't know anything about the dating or lonely-hearts part, but I appreciate that it gets right something near & dear to my heart... a proper three-state dark-mode toggle which has auto/dark/light (instead of the poorly-thought-out binary toggle most websites/apps have).
I tried out the site. Logged in, went to search for people in my city and... this is not possible!? You can select between San Francisco, London, New York, and All. There is literally no way to search for another city.
Now, I understand and empathize with bootstrapping problems and not wanting the site to look dead. But why would I bother putting up a profile when people relevant to me can't find it?
Yep, this is a problem we should have a fix for soon (today, hopefully). We originally only were thinking of the three regions and decided to allow everyone.
The site appears to be blocked by my ISP (!?!?!?) When I try to access it, I get a little cartoon drawing of a woman in a brown police uniform (!?!?!?!?!?) smiling at me with the words "Access to this website is blocked." and the logo and logotype of a company called Plume, who seem to be some kind of web infrastructure software that presumably my ISP uses.
Wow, that's funny. I'm not sure there's anything we can do.
It's like when early on the San Francisco school system blocked Manifold Markets. They don't want children to have fun learning how to trade markets, apparently.
Or when China blocked us after markets about Nancy Pelosi visiting Taiwan went viral.
So I just tried it out, and I have some suggestions / things that don't seem to be working.
1. There's no way to just enter a free-form profile; everything is in the form of set questions. Having set questions is nice for those who need prompts (also on that note IMO the current ones lean a bit too personal, you may want more of the lighter ones), but I just want to write a free-form profile. Now, there is a field for a link that may be to such a freeform dating profile, but...
2. The link field doesn't work! The Twitter field doesn't either! Neither of these things actually show up on one's profile! Moreover, one free link + one Twitter link seems a bit restrictive. Why not allow more? For instance, I would likely want to link to 1. my personal website, 2. my Dreamwidth blog, 3. my Facebook profile, and possibly 4. my LW user page. Now this could be accomplished with fields for all of these, but it could also be accomplished by letting people just enter multiple links.
3. Editing seems to require you to go through the whole thing every time? There doesn't seem to be a straightforward way to edit just one piece of things. Like, if I want to edit my picture, and I save it, it then takes me to the question page, rather than back to the profile page. Perhaps it's saved regardless of what I do next, but that doesn't matter -- after editing one thing you should be taken back immediately to your profile page.
4. I don't think one should have to answer every one of those slider questions? Again, some of these are fairly personal.
5. Here's a simple website bug -- under "Ethnicity/origin(s)", the label on the "other" checkbox is coded as actually belonging to the "other" checkbox on "relationship style". The same is true for the "other" checkbox under "political beliefs". As such these labels don't work properly.
6. The meaning of the "University" field is unclear. Where I went to undergrad? Where I went to grad school? Or perhaps current academic affiliation for those who have one? (Probably not that last one.)
7. I want to generally express dissatisfaction with some of these checkbox answers (politics, ethnicity). Possibly there should be a way to enter something custom here (although that has obvious downsides so maybe not).
Also, I don't like that the whole thing relies on Google accounts and requires me to be signed in to Google, although I realize that would not be easy to change.
This is really great feedback! Thank you for writing it down here.
1. I agree we should have a free-form spot to say anything.
2. The links missing from your profile is one of the last main bugs that we really need to fix soon. Hopefully will fix that today. It would be nice to allow more links, I agree.
3. Eventually, we want to have more in-place editing so you don't have to go through it all again.
4. You don't have to answer these. Maybe it should be more clearly optional?
5. Thank you, I didn't know about this bug. We will fix it.
6. Agree it's slightly confusing — you get to choose one of those, I guess?
7. Open to suggestions for politics / ethnicity. We were considering maybe removing ethnicity. Otherwise, I think the labels are a little bit off. It should say White instead of Caucasian, for example.
Hm, well as to #4, perhaps you don't have to answer them, but *having* answered one I don't see a way to remove it. Clicking on the same spot again doesn't do anything.
I think that "Moderate" should be relabeled "Centrist" or something like that (although this is a very weak preference). Perhaps it would also be more intuitive if "Liberal" and "Conservative" were relabeled "Left" and "Right" as well.
More granular categories for ethnicity could help as well I suppose.
As an aside, having the "religion" field be multiple choice rather than an arbitrary textbox could also help.
Hi, more comments! This time about the "near me" filter.
1. It seems to be generally acting wonky at the moment, I can't get it to consistently respect the set radius. I'm in New York, and sometimes it seems to work as described, but otherwise it restricts specifically to New York, New York.
This is noticeable because there's a person whose location is *Brooklyn*, New York, and when it's being restrictive this person won't show up even if I set the radius as low as possible.
2. Speaking of which, it ought to be possible to set the radius lower. I'm a New Yorker, I don't have a car. 50 miles is way too high a minimum radius.
3. Perhaps some sort of radius or location flexibility field could be part of the profile, as on https://dateme.directory/ , so that I have some idea of for whom I should look outside my normal radius.
Thanks!
Oh, also, is there any way to affect the order that various elements are displayed on your profile? This would be helpful. (Of course, if one has a freeform text profile that is displayed first, then this is less important.)
1. This is probably more difficult than my other suggestions (well, other than the no-Google one) and also less important, but, while I'm glad you can just straight-up message a person instead of having to go through a "mark them as liked and if they mark you also you'll see" system first, having the latter system is useful in some cases. Of course that's a whole new system to add and not actually the thing that's keeping me from turning on "looking for matches" and starting to mesage people (the main thing there is the lack of a freeform profile tbh, I really want to start messaging people but I just like... need a profile where I can actually put what I want to put), so (at least as regards me) I don't suggest weighting it too heavily, but it would be nice to have eventually.
2. I don't understand what the little green or yellow circles mean, and it's not explained anywhere. (My guess based on general convention was that "green" means "online right now"... but then what's yellow?) Could they have an explanatory tooltip or something?
How does manifold.love prevent the problem of female hypergamy that afflicts every other dating site and app? (Once female users of the app/site have seen the top 20 percent, say, in "attractiveness" of men, the other 80% of males effectively become invisible to all women. Adjust percentage to suit, 50% seems a reasonable upper bound/lower bound.)
I was thinking more adverse action--you're going to downrate the competition to make them seem less attractive, or have a confederate do so, or (rationalists being rationalists) write a bot that does so.
Not to mention you might conspire to downgrade the predicted likelihood of someone you actually want, in order to lower their status and decrease competition for *them*.
Remember that even if you solve physical scarcity, sexual scarcity will persist even if everyone becomes polyamorous (which I find unlikely)--people always want what they want.
The best thought on this is to show predictions for long term relationships.
The theory is that women actually want long term relationships, and they don't realize that by dating only the best men, they are more likely to get a short term relationship.
If we properly inform people how long their relationship will last, then women will choose slightly less attractive men (closer to their own level) to be stable long term partners.
I don't know if it's a "problem" so much as an "inevitable outcome given incentives and preferences of participants". Ultimately monogamy is definitionally 1 to 1 so it will mostly resolve itself, albeit leaving some women and men jaded and embittered (by being discarded and ignored, respectively)
I personally wouldn't waste my time on this dating community, my comparative advantage is awful given that I'm probably the median in many ways
Regarding the hospital bombing, as far as I know it wouldn't be fair to say the markets "got it right" and the evidence that it was fired from Gaza is still weak. We have the aljazerra video which NYT analysts say does not show whatever hit the hospital as Israel is claiming (https://archive.ph/Z3L81). We have an audio recording purported to be two Hamas militants blaming it on PIJ, which apparently does not ring true to Arabic speakers from the region (https://www.channel4.com/news/who-was-behind-the-gaza-hospital-blast-visual-investigation). The lack of a warning shot doesn't mean much since it appears from eye witnesses reports Israel has at least partly discontinued that policy after Oct. 7. We have vague assertions from Israeli and American intelligence with no info about how they came to that conclusion, historically that type of evidence has been.. unreliable. I'm not aware of the photos you refer to, but if they are the nail in the coffin I'd be surprised. Perhaps I've missed something, if anyone has more please let me know
This CNN article has a pretty good collection of the evidence in favor of it being a Palestinian rocket instead of an Israeli artillery strike or air bomb. There's a lot in there, but I found their analysis of photographs and video of the explosion site, particularly of the crater, to be worth quoting:
"Eight weapons and explosive experts who reviewed CNN’s footage of the scene agreed that the small crater size and widespread surface damage were inconsistent with an aircraft bomb, which would have destroyed most things at the point of impact. Many said that the evidence pointed to the possibility that a rocket was responsible for the explosion.
Marc Garlasco, a former defense intelligence analyst and UN war crimes investigator with decades of experience assessing bomb damage, said that whatever hit the hospital in Gaza was not an airstrike. “Even the smallest JDAM [joint direct attack munition] leaves a 3m crater,” he told CNN, referring to a guided air-to-ground system that is part of the Israeli weapons stockpile provided by the US.
Chris Cobb-Smith, a British weapons expert who was part of an Amnesty International team investigating weapons used by Israel during the Gaza War in 2009, told CNN the size of the crater led him to rule out a heavy, air-dropped bomb. “The type of crater that I’ve seen on the imagery so far, isn’t large enough to be the type of bomb that we’ve that we’ve seen dropped in, in the region on many occasions,” he said.
An arms investigator said the impact was “more characteristic of a rocket strike with burn marks from leftover rocket fuel or propellant,” and not something you would see from “a typical artillery projectile.”
Cobb-Smith said that the conflagration following the blast was inconsistent with an artillery strike, but that it could not be entirely ruled out.
Others said the damage seen at the site – specifically to the burned-out cars – did not seem to suggest that the explosion was the result of an airburst fuze, which is when a shell explodes in the air before hitting the ground, or artillery fire. Patrick Senft, a research coordinator at Armament Research Services (ARES), said that he would have expected the roofs of the cars to show significant fragmentation damage and the impact site to be deeper, in that case.
“For a 152 / 155 mm artillery projectile with a point detonation fuz (one that initiates the explosion upon hitting the ground) I would expect a crater of about 1.5m deep and 5m wide. The crater here seems substantially smaller,” Senft said."
I'm not an expert here, but aren't there Forecasting sites and Prediction Markets and aren't they different things? Should they be compared? The mechanics of them are different.
Manifold markets having fairly flat DAUs is a pretty negative signal to me on finding a real niche (you'd expect it to be growing a lot). Curious what people think about this.
The thing about prediction markets is that it's generally a bad idea to play them unless either you are better at them than the average player or you're hedging something. Manifold being play-money-only removes the latter, which in turn makes the former really, really hard (you cannot, after all, have 100% of players be better than average).
It has flat DAUs punctuated by events that spike DAUs, and leave them seemingly permanently higher afterwards. There's no growth day-to-day, but big events bring users in, and it looks like growth in the medium term to me.
Yes, this is true. On a longer scale, we've clearly grown a lot. Last August we were around ~100 daily active users. Now we're 1800-1900. I still think it can grow much more.
I always respected Scott's choice not to do podcasts. Still do. (Though he has an invitation from Tyler Cowen.) Now, doing an "interview/pretend talk" on youtube seems in all aspects like a (video)podcast, just worse. But who can say "No" to persons in skirts/cilts. ;) - Whatever, we all get to do stuff once - I was Santa.
The implication that Israel so far hasn't been occupying Gaza is misleading at best. They simply chose the somewhat unorthodox "open-air prison that prisoners run" model of occupation.
Not actually *that* unorthodox; many Latin American prisons are more or less run that way, they just don't happen to be mainly populated by women and children and have over two million people.
I'm not sure? AFAIK no other country would take them in, I think if some country did say they'd accept meaningful numbers of Gazans Israel might be quite happy to deport its problem?
Yes. Occupation is occupation. You have to physically be there to occupy something, whether that's a job or a country.
Blockades are a different thing that can also be bad. So is infrastructure dependence. Both of those things can be happening and can be bad, without an occupation occurring. And since 2005, Israel has not occupied Gaza.
In contrast, Israel does occupy the West Bank. Their soldiers are there, they determine the laws and control the justice system, they have bases and settlements all over West Bank. That's what you call an occupation.
And yet, per Wikipedia, "the UN, International Committee of the Red Cross, and many human-rights organization continued to consider it occupied as Israel military controls Gaza's borders, airspace, and sea access".
Do they get to define for the world what an occupation is? You can't occupy a place without being there. You can put a place under siege, blockade, embargo, exile, etc, but unless you're actually there it's not an occupation.
EDIT: If we're going to go beyond the typical definition of occupy to appeal to what organizations thing, the Regulations of the Hauge (an international agreement laying out the laws of war) defines an occupation as follows:
"Art. 42. Territory is considered occupied when it is actually placed under the authority of the hostile army.
The occupation applies only to the territory where such authority is established, and in a position to assert itself."
So with this definition we can see that the West Bank is definitely occupied: it is literally under the authority of the IDF, IDF determines and enforces the laws, and IDF is in a position to assert itself (military personal, bases, etc).
Gaza on the other hand, since 2005, has no Israeli military forces within it and Israel officially abandoned authority over the Gaza Strip. They do not make its laws, police it, or otherwise take a position of authority.
Now ever since Hamas took authority in the Gaza Strip in 2006 they have put significant restrictions on the ability of Gazan's to enter Israel, and they have blockaded the sea ports and airspace of Gaza. That may be a very bad thing! But it's not an occupation.
Now as Israel sends soldiers into Gaza and takes control of it in the days to come, *that* will be an occupation. If Gaza is occupied already, then that would imply that when the IDF storms in and takes control nothing has changed, which is ridiculous.
I don't understand how that's an occupation. It's a blockade. Israel tried occupying Gaza a while back and it was a very different regime (also, compare their occupation of the West Bank, which looks very different from "open-air prison that prisoners run")
There are reasons to think manifold.love is a terrible idea, e.g. Scott could bet on being in relationship with Aella not because he thinks this is at all likely to happen, but just to boost his status as measured by the prediction market. People being willing to loose money to influence the market makes it not work.
But, misgivings about the whole idea aside, it seems to be have been a big hit with the LGBT segment of the effective altruist community. It's looking like a Pride march or a gay bar in San Francisco. So it could still work, and the lack of cis-gendered women might not necessarily be a problem for that demographic...
This! So much this! There's room for innovation and progress in a lot of areas, but there are plenty of other areas that we got right long ago and should just be left as-is.
Totally agree. These dating sites are one step away from putting people on Amazon as consumer goods. Soon people on fuckingTinder will have star ratings from previous daters, and also comments: "Looks dumb but that just turns out to be his facial conformation. Was decent to talk to and quite fun to fuck, so long as you're not bothered by a lush growth of back hair." "She's lively, but after a while you realize that the energy is emanating from a merely average level of intelligence. She's pretty, but smells too funky." I met the people I fell for through friends, or work, or school or sports. Many of them I knew for months or even years before we felt a romantic spark. Meeting for drinks with a stranger to audition each other, each of you knowing you need to come across as caring and responsible, but each of you also knowing that the chance one of you will eventually ghost the other is high . . . OMG, so brutal. Fuck that.
I disagree with "got right long ago" - pre-app dating was awful and that's why there have been so many attempts to replace it. But I agree (with, thankfully, no personal experience) that most of the attempts seem to have made things worse.
I don't agree. Seems to me the primary reason why there have been so many attempts to replace it is because there have been attempts to replace (or augment) *everything* with "on the Internet" over the last 30 years, whether there was any need to or not, and dating is just another aspect of life that got caught up in that cultural wave. The Internet is our culture's shiny new toy, and we want to play with it everywhere.
Dating isn't awful as long as you go into it with reasonable expectations. The real problem is that we've spent the last 60-ish years dismantling the societal expectations that kept it manageable, so now it's difficult to find reasonable dating prospects unless you confine your search to some very specific and narrow pockets of people who reject those changes to our culture.
My most likely guess is the lack of women. But...if it makes rationalist gay guys happy, that's a net win!
Also, I'm not signing up (too old, too far, and sworn off relationships), but I love the guy whose profile is Asuka from Neon Genesis Evangelion (who's female, for the uninitiated). Way to show what you want. Doubt it's going to work, but who knows?
This is exactly what prediction markets are great at!
If the smart money knows that they have to match certain hiring practices to not run afoul of the law, then they will actively trade so as to make it match closely enough, no chance required.
I'm not going to bet that my company won't regret hiring someone if it's going to get them sued or whatever.
I can picture it now - one set of traders naively assessing candidates on their merits, and another running roughshod through the resulting prices to get them to add up to the demographic ratios they know are the minimum acceptable. End result: both types of information incorporated. Though the first group of traders will lose money if they don't wisen up and realise they have to hedge every merit-based bet they make with other bets that preserve the demographic ratios. Else they're just giving free money to the demographic traders.
It seems worth recalling the problems which actually did occur resolving the 2020 presidential election. I personally made a reasonable amount of money betting that Biden would win after the results had been announced. I can't recall the exact details now, but there was some date at which I thought the market ought to have resolved according to the published criteria, and there was some later date at which it actually did resolve. People were unhappy about both parts: the Biden bettors thought their payments were unjustly delayed (depriving them of the use value of the money) while a significant number of Trump bettors continue to believe that their man really won. People threatened to sue (https://www.insider.com/bettors-plan-to-sue-betfair-after-they-called-election-biden-2020-12), although I haven't heard anything about that since, so I guess it didn't go anywhere.
> People seem to trust the new Speaker to avert a government shutdown.
That's not exactly what's going on here. The market you linked shows the chance of a shutdown by the end of 2023 decreasing from ~55% to ~35% comparing before and after the speaker election.
But this market shows the chance of a shutdown by the end of 2024 dropping from ~70% to ~55% (still above 50%). And presumably Mike Johnson will remain speaker until the end of 2024!
Some missing context here is that after election, Mike Johnson immediately announced his plan to *temporarily* avert a shutdown by passing a stopgap (continuing resolution), pushing the potential shutdown date in the new year (January or April).
So, it's definitely a possibility that the market isn't showing so much "averting a shutdown" as "delaying a shutdown" (possibly only a delay of two months from November to January).
With that said, it is still notable that even the 2024 shutdown odds dropped after Johnson's election. And it would be nice to have a market for shutdown before April 30, 2024, to separate out a shutdown over the FY2024 budget vs a shutdown over the FY2025 budget.
"For real steam engines, that first niche was pumping water out of coal mines."
It helped a lot for that niche that the fuel was more or less just lying around, because the first ones installed were very inefficient. They quickly became more efficient as the owners could see that every shovelful going into the engine wasn't getting sold. But the first ones would not have made sense if the fuel had to be brought in from any distance.
So where's the "free fuel" and "quick feedback loop" for prediction markets?
It's more than that. Newcomen steam engines also had the shortcoming that they needed an ongoing supply of cold water to cool off the cylinder and condense the steam, as it was the vacuum pulled by the steam condensing that drove the power stroke of the engine. And water-flooded coal mines had plenty of that lying around, too.
While we're nitpicking on the examples I want to point out there's a lot more separating an ancient aeolipile from a borderline-practical steam engine than just a useful application. There's about 1600 years of advances in materials science, precision manufacturing and basic thermodynamics separating them.
I'm not an expert in the matter but it always upsets me when people call aeropyles "steam engines" even though the core innovations behind actual steam engines seem related to pistons with tight tolerances
This video about one of the first precision lathes is excellent:
> Hanson is less sure about this answer than the overall story, but he suggests hiring. You could create some kind of product that companies could buy and give their hiring managers at the beginning of a hiring round, asking them to predict which candidates would get good employee evaluation results or promotions at the end of X amount of time.
Honestly, that sounds horrifying, as it opens up the door to all manner of abuse and conflict-of-interest questions. Predicting who will win an election that millions of people participate in is one thing; "predicting" what will happen to people when you have some degree of control over their fate is quite another! (And yes, the "hiring manager" is not necessarily the *manager* manager who will be making these decisions on evaluation or promotion, but I bet they know each other and talk about work stuff on a pretty regular basis...)
That sounds like it will just be a fancier way of "what university did you go to? what major? your work history? are you going to fit in here with us or not?" that most hiring managers use anyway, it will just be packaged as New Better 'Cos AI.
Most hiring managers are likely to 'predict' that the candidate who went to Hahvaad is going to get good evaluations/promotions, whereas the guy from Clodkickers U... not so much.
The New York Times has a history of getting things wrong-- that hospital in Gaza, weapons of mass destruction, Kitty Genovese, crack babies.... Is there money to be made in betting against the NYT?
Maybe they could have an internal market about whether particular stories will have to be edited or retracted or apologized for?
I think they mostly get individual facts right. They've got systemic bias in their selection and presentation, but I don't see a clean way to run a market about that. The big failures seem to me to be when their sense of righteousness completely overwhelms their commitment to factual accuracy. Those feel too rare to be the source of much income, and also they're the sort of thing where most observers are also polarized at the same time.
"Normally this is a problem for Congress, election regulators, the courts, the media, and the American people. But if there are election prediction markets, then election fraud would indirectly become a type of financial fraud, and now it is also a problem for the CFTC."
Yeah, I imagine that if prediction markets set themselves up to be "we're like the stock market" rather than "we're like bookies", then financial regulation bodies, instead of Revenue or gambling regulatory bodies, will be the ones overseeing them. And if they have to decide on the outcome of elections, no wonder they don't want to be involved.
You're thinking "surely reasonable people won't need big decisions made", but you're forgetting there's a lot of unreasonable people out there. Who would have thought that the courts would have to decide if a piece of paper that was nearly, but not quite, all the way punched through counted as a vote or not? I don't blame the CFTC for not wanting to get involved in the financial equivalent of hanging chads.
For the love market, I would ordinarily advise user Nathan to ditch the left-side black and white photo, as "My hobbies include dressing up as Slenderman" wouldn't seem to appeal to most women, but if he's looking for love among the rationalists, who knows? Slendy could be just the guy they're looking for!
I tried to phrase two mid-term solutions for Israel as prediction markets: withdraw and occupy. No idea how phrase "puppet-government" in a way that I could reasonably resolve it.
I think Manifold.love should *embrace* people betting up their own candidacy, rather than trying to restrict it. The basic problem with Tinder or Hinge is that men swipe right on ~everyone and women are overloaded with potential matches. But if you start charging people money (indirectly via mana purchases) to make bets, men would be incentivized to think carefully before "swiping right" and the cost of swiping will go up if the person in question is more attractive.
So, eventually, if Manifold.love gets enough traction, it will cost (say) $100 to get to the top of Aella's match queue, $50 for someone attractive but not well known or $5 for the average person. Thus no one will be overwhelmed with potential matches.
I'd make two changes to the process:
1. Let each person immediately resolve any market to No if they're confident they're not a good match with one of the candidates in the pool
2. Introduce a fee for men to create a profile, in order to encourage a better gender ratio. Perhaps something on the order of 1000 mana?
"The basic problem with Tinder or Hinge is that men swipe right on ~everyone and women are overloaded with potential matches."
The system already disincentivizes this. You only have a finite number of swipes, and presumably the algorithm penalizes you for swiping indiscriminately.
You have a finite number of swipes but also a finite number of hours in the day - if you're not using all your 'yes' swipes each day you're wasting them, and how long do you want to spend on the app rather than doing fun things in your precious free time?
So, personally I'm picky and absolutely spent time considering profiles before swiping, but I also really hated dating apps and didn't have much success with any except OKCupid. I think a lot of men out there have fairly modest standards for women they'll have casual sex with (as the joke goes, "1. Female, 2. Willing, 3. Breathing")
Re algorithmic penalties, most people don't know how those work, even if nerds like us are aware of them
TL;DR: Here’s how I’d use prediction markets if I was a dictator.
Prediction markets are about providing information. The steam engine use case for prediction markets is throttling the information ecosystem. Right now the way that something like twitter (“X”) works is that you make a divisive tweet, then other people fight about it, and you get visibility and more followers. You do it again and this time it goes out to even more people. A feedback loop of rage. You do this enough times and end up with a power law distribution of users-to-followers, and twitter has turned into a dumpster fire and everyone is screaming at each other and we can’t agree on who won the election and is supposed to have control of the nuclear football.
So instead, you have a site where you start with 1,000 points, like manifold, and this is your max number of followers. When you hit your limit, no one else can follow you. You’re capped. If you want the ability to have > 1,000 followers then you need to win some bets.
The rule doesn't mean that “You must follow superforecasters,” because superforecasters might be super boring and no one wants to follow them. Instead, the rule means, “Only super forecasters are allowed to have a disproportionate influence on the information ecosystem.” You’re only allowed to have a hundred thousand followers if you have a basic grasp of reality. Lots of people don’t want to do forecasting and that’s fine. You’d still have your default 1,000 limit and most social media users never get above that anyway.
Of course, this only improves the information ecosystem if people actually use it. Lots of people want rage. But if I was a dictator, I might just ban everything else.
Here in the US, maybe Elon Musk had an opportunity, but since the system was already in place, he couldn’t have done anything without ripping it apart. You’d have to start from scratch. Or maybe substack had an opportunity when they rolled out Notes.
Huh, I actually really like this idea. You would probably have to ban everything else to stop a competitor that panders to toxoplasmosis from taking over, but you did start the hypothetical with "if I were a dictator"
A few years back, I listened to an interview with the founders of airbnb where they were talking about growing pains in their early days. They complained to their investors that people weren't getting into the spirit of how to set up their pages so people would be enticed to rent from them. They wanted to create a culture of high-quality pages, but their guidelines weren't being followed.
"Where are your users?"
"About 80% of them are in New York City."
"You should go to New York and help them set up their pages, then."
"That doesn't make any sense. We want this to be a global system. Helping every user set up their page will never scale."
"Right, but when you're small you do things that don't scale."
So they went to NYC with a DSLR, telling users they'd been selected to have a professional photographer and a consultant help them set up their page. It worked, and the rest is history.
I think about this idea a lot when it comes to early growth models for a company or an idea. Whether it's prediction markets or impact markets, I think you have to be willing to do things that don't scale when you're small. And once you're willing to do things that don't scale, this opens up a lot of ideas for WHAT you could do to grow the idea that hadn't been options before.
Better said than I could have. The question is: how to implement these ideas in prediction and impact markets? Or have prediction markets already crossed the scalar threshold?
I've mentioned it in one of the open threads, but Fox News has this America Predicts [0] thing which seems like it deserves an honorable mention. I'm not aware of any other major mainstream news organization doing something similar.
I'd be interested in seeing prediction markets used to decide whom to invite to political debates. Rather than the current system of coming up with a bunch of ad hoc criteria, just have a prediction market for each candidate's odds of winning the primary or the general election contingent on them being invited to the next debate. Everyone over a certain threshold that increases the closer you get to the election (say, start at 10% and ramp it up to 30%) gets a spot on the stage.
It'll also be an excellent test case for arguments over how robust prediction markets are against manipulation by people with deep pockets and large stakes in the market's prediction.
That actually sounds like a sensible use for a prediction market! Let people estimate the chances of potential debaters and that should clear out the no-hopers who only got a spot for unknown reasons (why on earth, or how on earth, did Julián Castro get his five minutes in the spotlight?)
Plasma's survey of manifold users found a lot of furries.
Bit of an omission for manifold.social here ... manifold,social asks you for your preferences re. M/f/nb/trans m/trans f, but no questions on how you feel about furries.
This is on the Internet .. of course there will be furries.
A lot of people who believe the American government lies about everything are sure that Israel didn't bomb the hospital because the American government says so
My reaction to the legalization talk was "well, what did they expect?" Trying to pass off gambling on elections as "hedging" just sounds disingenuous on top of the existing problems.
Next Mantic Monday maybe you should look into markets on the Tesla Strike in Sweden. I think there is probably alpha in those. Remarkable situation really, historically the unionizers usually win by getting their demands met, but if the CEO is outspoken anti-unionizer Elon Musk will it truly go down that way?
Here goes my careful filtering of the specifics of October 7th...
I'm sorry, I should have thought of that and not put this here.
The resolution criteria on the hospital bombing question are bullshit.
I trust you have a different but related market to suggest instead. Or a proposed resolution criteria, by which one might create yet another market. It certainly seems clear that this particular market was anointed as the main market by luck rather than some other factor, so the existence of a better market, released later and with more thought, would not be surprising.
In any case, please share.
I think it's fair to point out a problem with a market's resolution criteria without being able to pose better ones. For one thing, it might be the case that the underlying question is incapable of translation into precise resolution criteria, which would demonstrate a problem with this use of prediction markets.
In this case, the original criterion was "I will resolve it to my best guess of the median informed observers' subjective probability based on all reliable sources". On 18 October 2023, the creator gave some clarification.
I personally don't think it's a problem that the resolution relies on Milford Hammerschmidt's assessment of the evidence. Market participants can form a view on whether Mr Hammerschmidt is a fair judge. More of a problem is that (a) it resolves based on informed observers' "subjective probability", presumably meaning which outcome seems more likely, rather than the ground reality and (b) there is no indication of when this assessment will be made.
The fundamental problem is the same as the Wuhan case: how will we ever know for sure what happened?
We can look at a few other markets on the same topic. This market https://manifold.markets/Joshua/will-the-alahli-baptist-hospital-ex is based on what Wikipedia will say within 30 days, which seems reasonable (provided we trust the creator's interpretation of Wikipedia), but gives the boring answer that Wikipedia will probably be ambiguous.
This market https://manifold.markets/MarcusAbramovitch/will-the-new-york-times-definitivel asks whether the NYT will definitively state Israel was not behind the blast, which according to the market is very unlikely. This is perhaps unsurprising, but it seems worth noting that changing the resolution criterion this way essentially reverses the answer.
Yes, this stuff will always be awkward. And multiple adjacent-to-the-real-question markets is often a good answer to getting a handle on what's "actually" going on. Personally I like a mix of some markets that are best effort at objectivity and clear definitions (but usually aren't exactly the actual question of interest), combined with a couple that are more subjective but closer to asking exactly what people want to know. And then we get a sense of both what's going on, and where the confusion lies, which enables excellent analysis of the sort you're doing in your comment.
I liked the "resolves to Wikipedia" approach as part of that, and used it for a pair of mine on the death toll, as reported by English and Arabic Wikipedia. I think the result is an interesting blend of "objectivity" that highlights the way sources matter.
https://manifold.markets/EvanDaniel/how-many-people-died-in-the-alahli
https://manifold.markets/EvanDaniel/how-many-people-died-in-the-alahli-98d0ef8cbcf6
I also agree that criticizing a market without being able to suggest improvements is fair game. Definitions are hard, and it's entirely possible to make things worse by trying to address more edge cases in advance. (Especially since clear and concise writing is very valuable here!)
But there's a huge range between these sorts of critiques, or pointing out minor issues and edge cases, and calling it "bullshit". If you're calling it "bullshit", I think you should be able to point out _some_ room for improvement.
You can absolutely call bullshit without needing to point to an improvement. Some questions are simply intractable from the position of epistemic certainty. I acknowledge this is not an equivalent issue, but for example if you made *any* sort of resolution criteria for "The existence of a creator God" they would all be bullshit
I think this case is a bit easier than "does god exist". And in fact I think that difference makes a difference; I would not have posted my original reply in that context.
Yeah, I'm extremely disappointed by this post's lack of mention of it. US government statements were used as a resolution source, but whether the US government is honest on the topic is itself a major controversy! Perhaps there should be a blanket rule that resolution source must be mentioned in the market name, to avoid huge discrepancies like this one.
That was true for the Polymarket market but not my manifold one.
I would argue that any likely resolution criteria on the hospital bombing is likely bullshit.
If you have a prediction market on LK-99 being a super conductor, chances are great that eventually a consensus answer shared by (100%-lizardman constant) of the population will eventually form.
If you have a question on the existence or gender identity of god, chances are good that you won't have a super-majority opinion you can use for resolution.
The hospital question could be restated at "will the author of the question believe that the probability of Israel having caused the explosion is more than 50%?"
So the question does not track the first order probability at all. If an oracle tells me that the at the resolution date, the poser of the question will estimate the chances that the IDF was responsible at precisely 47%, that does not mean I should buy yes up to 47%, but that I should sell yes down to 0%. (This could be prevented if they pre-committed to resolving the question by randomly sampling their best guess probability distribution. The downside is that this would be even more likely to create a feedback loop. Should the yes probability be 10% or 15%? Hard to decide for an individual. Luckily, there is a prediction market out there, let's just take their probability. Oh wait.)
The other problem is that a lot hinges on the market prediction of the judgement of the questioner. If the name of the questioner was clearly Jewish/Arab/Farsi (and they optionally changed their icon to a flag of Israel/Palestine/Iran), this would greatly affect the level on which yes is trading.
On the one hand, such prediction markets are clearly useful to update on emerging information. On the other hand, I don't feel that the absolute magnitude they are trading at is very useful.
What do the holders of yes (7%) bet on, anyhow? That there is a 10% chance that evidence will emerge which pushes the probability of "IDF did it" above 50%? That the creator of the market is giving more weight to Arabian media then other people assume?
I am the author of the market in question. When people on manifold say "resolve to a probability", they mean exactly that - they resolve it (e.g.) "at" 47% and YES shares become worth 0.47c. So such a resolution would avoid this failure mode.
Do you have any evidence about the mutual acceptability of market resolution in controversial cases like the hospital car park explosion? Like, how did the many COVID-19 origins markets resolve?
At some point these markets are really asking "how will the platform/host resolve this market", which may have dramatically different odds from "what is reality". I guess that's banal at this point but how much does it matter after all?
This is why I advocate for the same thing Scott suggests for election markets: Resolve to a single source: https://manifold.markets/Joshua/will-the-alahli-baptist-hospital-ex
Great post! For those not on Manifold, I think it's also worth directing attention to the biggest market on the site, the main Speaker of The House Market:
https://manifold.markets/SimonGrayson/who-will-be-the-next-speaker-of-the-0b49bf53ad12?r=Sm9zaHVh
It was an absolutely insane three week rollercoaster, as the graph suggests. I also recommend this post by Marketwise summarizing winning trading strategies: https://marketwise.substack.com/p/speaker-election-interviewing-the?r=28sbv1&utm_campaign=post&utm_medium=web
"I can’t find any markets on the Middle East topic I’m actually interested in, which is Israel’s medium-term plan. Will they kill some Hamas leaders, then get out? Install a puppet government? Permanently occupy Gaza like they’re occupying the West Bank? These all seem like bad options, but they’re very different bad options, and I haven’t seen much speculation about which is most likely."
Does Israel have ANY non-bad options? I think the answer is unfortunately no...
I guess oust Hamas, and then hand the keys over to the PA.
Yeah, that would be one of my first ideas too, but see https://www.economist.com/briefing/2023/10/23/can-the-palestinian-authority-control-gaza-if-hamas-is-ousted (or https://archive.ph/BeZFz for the unpaywall version). The PA is corrupt and ineffectual and really needs the IDF to prop it up.
P.S. https://arxiv.org/abs/2310.19779
Oh wow, that IS big news! That was the last one, I think.
Anyway, as regards Israel Palestine, Israel could occuppy Gaza and start to rebuild it. The locals will hate it. The PA can give some rousing speeches about the Iaraeli occupation at the UN, the US president can apply pressure, and 5 years down the line Israel will "cave in to the demands" and let the PA take control. The PA regains prestige, and Nobel peace prizes for everybody.
I like this, but I think the feasibility is complicated by every political leader's desire to appear the victor to their public. Israel "caving in" looks good for PA, bad for the reigning Israeli PM. Of course, "I'm the PM who brought peace" is a win, but only for some sectors in the Israeli public, and not all Israeli politicians could pull it off.
I am not very optimistic, unfortunately.
Just spitballing, what would it take to create a new PA-like entity that governs Gaza only? A new government that has nonzero local legitimacy but isn't Hamas and explicitly bans anyone with ties to Hamas, and has some buy-in (even if tacit) from both Israel and the Arab League.
Definitely difficult, but impossible?
The only reason most people still turn up to work for the PA is because it's so difficult to find alternative decent jobs. Palestinians really do not like the PA and there is significant territory that the PA can't work in for fear of being killed by their own people.
It's a bad option.
Not options that they tell their public, at least. Here is a somewhat related market:
https://manifold.markets/Shump/who-will-control-gaza-6-months-afte?r=WW90YW1GZWRlcm1hbg
Deport & annex half or a third of Gaza. Less space to live in would probably slow Gaza's demographic growth (which is the big long term problem), would make a bit less israeli territory vulnerable to rockets (by pushing their launching point a couple dozen km south), and make Hamas operate in a smaller urban area, probably making it harder to hide or something.
It would burn a lot of international sympathy & political capital, but these are only as valuable as the strategic goals you aim to achieve. And a step toward the "one israeli jew state" solution is quite valuable.
Any solution involving "destroy/oust Hamas from the strip" begs the question "how?". Removing an urban insurgency -one with at least 20-30% support in the population, from the most conservative estimates i've seen- makes "displacing a million people" looks like the easier solution.
“ And a step toward the "one israeli jew state" solution is quite valuable.”
Is that stated policy? Wouldn’t it mean deportations of existing Israeli Arabs?
>Is that stated policy?
No, except from fringe extremists. But realistically, what other end state is there? Any 2 (or 3) states solution would just keep the conflict going (until they run out of luck/talent and the next arab coalition finally win a war). Two peoples don't cohabitate a land, not without a new culture emerging and fusing the two together (and -correct me if i'm wrong, that's a point on which i'm just running on gut feeling- it's really not the current zeitgeist of either jewish israelis or arabs.
>Wouldn’t it mean deportations of existing Israeli Arabs?
They're, what, 2 millions out of the 9m israelis? That's a sizable minority, but at this point, not too much out of line for a western countrie, but tripling that number by integrating Gaza & the west bank into a "one ecumenical state" would turn Israel into an arab majority democracy, and let's be real, nearly no Israeli jew want that.
I really don't see what else that leaves as the long term state. Either one people go or the other, and in 2023, only one of them is in a position to push the other in that direction. It don't even need to be a willing effort. The fringe elements in Israel will continue slowly pushing on the west bank, and in Gaza, Hamas actions will inevitably force Israel to *do something*, and that *something* will never be "give them more leeway"
I have no doubt that you’re sincere in your stated belief, but I feel like that’s the opinion the crazy fringe elements (on both sides) want you to have.
Consider that Hamas started this war because it was concerned Israel was approaching a rapprochement with Saudi Arabia; an unprecedented step towards a lasting peace in the region.
I'd say the recognition by Egypt was more important, but that was a long time ago. But yes, rapprochement with SA would be a great step for Israel, and for interstate peace in the region. I'm not convinced it would help Palestinians in many ways.
> Less space to live in would probably slow Gaza's demographic growth (which is the big long term problem),
I don't agree. "The big long term problem" is the local culture. Even the Gazans who don't support Hamas still support Hamas's general goal of genocidal antisemitism. They are taught as children that the land belongs to them "from the river to the sea," (aka everything between the Jordan River and the Mediterranean, aka the entirety of Israel,) that Israel has no right to exist and that it is a virtuous act to kill Jews.
The big long term problem is that Israelis and Palestinians *can not coexist peacefully,* because the Palestinians have ruled out the possibility, in very absolute terms, since the very beginning. And that's a pretty big long term problem!
I feel we're saying the same thing. This culture is a long term problem for Israel because their demographic growth is high. If Gaza was slowly losing inhabitants, then it'd be a temporary threat that can be contained until it stops being relevant.
Ah, that makes more sense, thanks for the clarification.
Could we please, please stop openly advocating genocide, or maybe take breaks from it one or two days per week? Historians will not use different rules to study enemies of the west than they do unaligned tribes in Central Africa the way we do.
I second this and think your name is very appropriate in this context
There are orders of magnitude more Arabs than the population of Gaza; no more Arabs in Gaza - although arguably an ethnic cleansing - is hardly a genocide.
That seems like an over-generalization. I have heard anecdotes of Israeli Palestinians who can live within Israel for decades without killing any Jews.
While genocidal antisemitism is one big driver of the conflict, I feel that Israeli settler activity in the West Bank is also a driver of the conflict.
The West Bank? So... places like Bethlehem, the City of David, which is now claimed for exclusive use by the sworn enemies of the People of David, who are taught to deny that Jews ever had any historical presence in the land, even in the face of all the evidence? ( https://www.youtube.com/watch?v=mkLs-yOi9gQ )
No, by any rational, objective standard, the illegitimate "settlers" and "colonizers" here are and always have been the Palestinians.
You dodged his main point. Palestinians are capable of living in peace with Jews in many contexts.
Q: How long did it take after Germany's defeat for the culture to shift away from genocidal death cult?
A: A year or two I think.
It is astounding how Bibi's missiles have blown up the wide safety perimeter established over generations of activity by dedicated organizations like the ADL around subjects like relativization of the holocaust, leading to the reestablishment over a few days of a complex boundary, negotiated individually, that because of overton window dynamics and human stochasticity will be continually crossed over.
I think that the main the Nazi ideology was to a large degree refuted by thee defeat of their "1000 year empire" at the hands of Slavic "Untermenschen". Their propaganda was not big on preparing the population for resistance during occupation. For the leading Nazis defeat was to horrible to imagine. (Werwolf was more of a plan to have partisans in occupied territory while there still was a front rather than a long term plan to establish a fourth Reich after total defeat.) (Also, it is not like many Germans became philo-semitic democrats when they surrendered. That took a few generations.)
By contrast, Palestinians have experienced military defeat multiple times. Any ideology thriving in such circumstances will not be refuted by yet another defeat.
> Q: How long did it take after Germany's defeat for the culture to shift away from genocidal death cult?
More relevant Q: how long had Germany *been* a genocidal death cult before their defeat?
A: just a few short years. Not "throughout the entire lifetime of of every citizen and all of their ancestors for centuries." That kind of changes the calculus.
Look, I am sorry, but the time of David was some time ago. From what I recall, a diaspora happened, and other people settled there.
I generally do not support a policy of allowing descendants of ancient civilizations an automatic claim to the lands they once held. If the Greeks (or Macedonians) were making a claim for Alexandria, or First Nations one for Manhattan, or some tribe which suffered defeat by Israel in biblical times made a claim for their original homeland which now happens to be located within modern Israel, or Austria made claims on Hungary or South Tirol, or Rome made a claim on London, I would be equally opposed to all of these.
I support Israels right to exist not because of any claims in biblical times, but because I consider basically all borders to be historical accidents and generally prefer that arbitrary status quo to attempts to shift borders, which tend to cause a lot of suffering.
From my understanding, Rabin signed the Oslo Accords, which ceded Gaza and the West Bank to the Palestinians. If the Palestinians prefer to claim that there never were ancient Jews in the West Bank, that is unfortunate, but not a reason to take their land.
> From what I recall, a diaspora happened,
Yes.
> and other people settled there.
No. With the notable exception of Jerusalem, which remained significant for religious reasons, most of Israel became barren, inhospitable desert after the diaspora. Very few people wanted to live there, even among the local Arabs who were used to barren, inhospitable deserts. When the Zionist movement of the late 19th century began to draw Jews back to their ancestral homeland, the Ottoman Empire (nominal ruler of those parts) didn't want them coming in, but the Jews had little trouble settling anyway because *there simply wasn't much of an Ottoman presence around to stop them.*
It wasn't until after the Zionists began to reclaim the land and make it an attractive to live again, making "the desert blossom as the rose," that the neighbors decided they'd like a piece of what the Jews had built. And the rest is history. Long, ugly history that led directly to the atrocities of the past month.
I was under the impression that rockets from Hamas can hit Tel Aviv , which is quite a bit (50km?) away from the strip. Making Gaza smaller seems unlikely to help there much.
> probably slow Gaza's demographic growth
That is a very polite way to phrase "conditions so hellish that humans can't thrive".
The solution you advocate for could reasonably be called an ethnic cleansing.
I would prefer for Israel just to occupy Gaza and run it, and make a better job of it than Hamas. Go full surveillance state, kill any insurgents. Try to raise their GDP to a level where the population growth is less of a problem. If in a generation or three, inhabitants should begin to favor peaceful coexistence, give them their own state.
Not an expert on Palestinian rocketry, but I believe by far the most common rockets are semi-homemade and mostly have a max range of <10km. Anything that can hit Tel Aviv is primarily manufactured in Iran and available in much more limited supply.
I also THINK Iron Dome is going to have an easier time dealing with the latter (especially given the more limited numbers).
Hey, at least he didn't say "economic degrowth" or "create fifteen-minute cities."
Yeah I think this is the best (?least bad) of the possible solutions I've seen so far
Basically the "solution" is Xinjiang
I don’t think there are any good short term options. Short term they need to not burn too many bridges ie respect international humanitarian treaties and help civilians survive/escape. Medium term help to set up infrastructure. Long term get out and cross their fingers.
It seems like the government is strongly considering expelling all Gaza residents to Egypt as the least-bad option: https://www.972mag.com/intelligence-ministry-gaza-population-transfer/
No that's just some ministry with no power; see e.g. Richard Hanania's twitter.
The key problem with that is that Egypt doesn't want them either and do you want to go to war with Egypt to try and force the issue?
Jonathan Anomaly's talk was mostly laying out the basics of polygenic screening. Prediction markets factor in because we want to know which diseases will likely be cured in the coming decades. Genomic Prediction creates an embryo health score to maximize DALYs for the embryo. But a lot of diseases only reduce the quality of life when someone is much later in life. If we think a disease will be cured, we should put less weight on it. I would put stronger weight on traits we likely will not be able to cure such as low mood, poor personality traits, and cognitive ability.
Manifold is way too optimistic on PES (I wish):
https://manifold.markets/Gigacasting/embryo-selection2030
https://manifold.markets/Gigacasting/embryo-selection2050
Among the poor personality traits we are not likely to cure are low empathy and a tendency to violence. Often those 2 occur together, because if other people's feelings are not real to you it's not that big a deal to harm them. I think that's the most important target to select for genetically. People here often believe intelligence is more important to select for, and then add that smarter people will be less violent because they will be able to understand that it's morally wrong and often not productive. I don't agree. There are lots of smart people here, and little agreement about what's the right thing to do about this and that. I think that what makes people more reasonable is a sense of connection with other people, and a relatively low level of rage, rather than the ability to reason about ethics and the logic of arguments.
Regarding manifold.love:
1) Markets resolve "no" when someone presses the "reject" button. So every market is implicitly "Will neither subject press the reject button, and will they date for 6 months?"
2) Since Scott checked the site out, they've added an easier way to see who has bet yes or no on a match. You can also see the full trade history by opening the match's market and looking at the "trades" tab like a normal market.
3) There is currently ongoing debate as to whether the site should focus more on 6 month relationships or 1st/2nd/3rd dates. If you have an opinion, come bet on it here:
https://manifold.markets/SG/will-manifoldlove-hit-1000-signed-i?r=Sm9zaHVh
I've gone on plenty of third dates that still didn't go anywhere, so I hope that doesn't end up being taken as indicative of relationship success.
While I agree, I think "well enough paired to make it to the 3rd date" is much easier for an external matchmaker to assess, and resolves faster, so might be a much better market criterion.
Awesome!
Happy to answer any questions about manifold.love.
The markets on whether the two people will date for six months are indeed conditional on whether they reach a decision (either party can reject, but both have to confirm they dated for 6 months to resolve Yes). The markets either don't resolve (depending on loans to get your mana back) or resolve N/A after a long time, like when you get married.
Also, you can now see the positions of who bet on a relationship!
Super excited to see where this goes as we've literally just launched a few days ago.
I don't know anything about the dating or lonely-hearts part, but I appreciate that it gets right something near & dear to my heart... a proper three-state dark-mode toggle which has auto/dark/light (instead of the poorly-thought-out binary toggle most websites/apps have).
Thank you!
I tried out the site. Logged in, went to search for people in my city and... this is not possible!? You can select between San Francisco, London, New York, and All. There is literally no way to search for another city.
Now, I understand and empathize with bootstrapping problems and not wanting the site to look dead. But why would I bother putting up a profile when people relevant to me can't find it?
Yep, this is a problem we should have a fix for soon (today, hopefully). We originally only were thinking of the three regions and decided to allow everyone.
The site appears to be blocked by my ISP (!?!?!?) When I try to access it, I get a little cartoon drawing of a woman in a brown police uniform (!?!?!?!?!?) smiling at me with the words "Access to this website is blocked." and the logo and logotype of a company called Plume, who seem to be some kind of web infrastructure software that presumably my ISP uses.
Wow, that's funny. I'm not sure there's anything we can do.
It's like when early on the San Francisco school system blocked Manifold Markets. They don't want children to have fun learning how to trade markets, apparently.
Or when China blocked us after markets about Nancy Pelosi visiting Taiwan went viral.
So I just tried it out, and I have some suggestions / things that don't seem to be working.
1. There's no way to just enter a free-form profile; everything is in the form of set questions. Having set questions is nice for those who need prompts (also on that note IMO the current ones lean a bit too personal, you may want more of the lighter ones), but I just want to write a free-form profile. Now, there is a field for a link that may be to such a freeform dating profile, but...
2. The link field doesn't work! The Twitter field doesn't either! Neither of these things actually show up on one's profile! Moreover, one free link + one Twitter link seems a bit restrictive. Why not allow more? For instance, I would likely want to link to 1. my personal website, 2. my Dreamwidth blog, 3. my Facebook profile, and possibly 4. my LW user page. Now this could be accomplished with fields for all of these, but it could also be accomplished by letting people just enter multiple links.
3. Editing seems to require you to go through the whole thing every time? There doesn't seem to be a straightforward way to edit just one piece of things. Like, if I want to edit my picture, and I save it, it then takes me to the question page, rather than back to the profile page. Perhaps it's saved regardless of what I do next, but that doesn't matter -- after editing one thing you should be taken back immediately to your profile page.
4. I don't think one should have to answer every one of those slider questions? Again, some of these are fairly personal.
5. Here's a simple website bug -- under "Ethnicity/origin(s)", the label on the "other" checkbox is coded as actually belonging to the "other" checkbox on "relationship style". The same is true for the "other" checkbox under "political beliefs". As such these labels don't work properly.
6. The meaning of the "University" field is unclear. Where I went to undergrad? Where I went to grad school? Or perhaps current academic affiliation for those who have one? (Probably not that last one.)
7. I want to generally express dissatisfaction with some of these checkbox answers (politics, ethnicity). Possibly there should be a way to enter something custom here (although that has obvious downsides so maybe not).
Also, I don't like that the whole thing relies on Google accounts and requires me to be signed in to Google, although I realize that would not be easy to change.
This is really great feedback! Thank you for writing it down here.
1. I agree we should have a free-form spot to say anything.
2. The links missing from your profile is one of the last main bugs that we really need to fix soon. Hopefully will fix that today. It would be nice to allow more links, I agree.
3. Eventually, we want to have more in-place editing so you don't have to go through it all again.
4. You don't have to answer these. Maybe it should be more clearly optional?
5. Thank you, I didn't know about this bug. We will fix it.
6. Agree it's slightly confusing — you get to choose one of those, I guess?
7. Open to suggestions for politics / ethnicity. We were considering maybe removing ethnicity. Otherwise, I think the labels are a little bit off. It should say White instead of Caucasian, for example.
Hm, well as to #4, perhaps you don't have to answer them, but *having* answered one I don't see a way to remove it. Clicking on the same spot again doesn't do anything.
Glad this is helpful!
Thanks, that's another thing we want to fix.
> Open to suggestions for politics/ethnicity.
I think that "Moderate" should be relabeled "Centrist" or something like that (although this is a very weak preference). Perhaps it would also be more intuitive if "Liberal" and "Conservative" were relabeled "Left" and "Right" as well.
More granular categories for ethnicity could help as well I suppose.
As an aside, having the "religion" field be multiple choice rather than an arbitrary textbox could also help.
Hi, more comments! This time about the "near me" filter.
1. It seems to be generally acting wonky at the moment, I can't get it to consistently respect the set radius. I'm in New York, and sometimes it seems to work as described, but otherwise it restricts specifically to New York, New York.
This is noticeable because there's a person whose location is *Brooklyn*, New York, and when it's being restrictive this person won't show up even if I set the radius as low as possible.
2. Speaking of which, it ought to be possible to set the radius lower. I'm a New Yorker, I don't have a car. 50 miles is way too high a minimum radius.
3. Perhaps some sort of radius or location flexibility field could be part of the profile, as on https://dateme.directory/ , so that I have some idea of for whom I should look outside my normal radius.
Thanks!
Oh, also, is there any way to affect the order that various elements are displayed on your profile? This would be helpful. (Of course, if one has a freeform text profile that is displayed first, then this is less important.)
Thank you, shared this with the team!
Thank you again! Two more notes:
1. This is probably more difficult than my other suggestions (well, other than the no-Google one) and also less important, but, while I'm glad you can just straight-up message a person instead of having to go through a "mark them as liked and if they mark you also you'll see" system first, having the latter system is useful in some cases. Of course that's a whole new system to add and not actually the thing that's keeping me from turning on "looking for matches" and starting to mesage people (the main thing there is the lack of a freeform profile tbh, I really want to start messaging people but I just like... need a profile where I can actually put what I want to put), so (at least as regards me) I don't suggest weighting it too heavily, but it would be nice to have eventually.
2. I don't understand what the little green or yellow circles mean, and it's not explained anywhere. (My guess based on general convention was that "green" means "online right now"... but then what's yellow?) Could they have an explanatory tooltip or something?
Thanks again!
How does manifold.love prevent the problem of female hypergamy that afflicts every other dating site and app? (Once female users of the app/site have seen the top 20 percent, say, in "attractiveness" of men, the other 80% of males effectively become invisible to all women. Adjust percentage to suit, 50% seems a reasonable upper bound/lower bound.)
I was thinking more adverse action--you're going to downrate the competition to make them seem less attractive, or have a confederate do so, or (rationalists being rationalists) write a bot that does so.
Not to mention you might conspire to downgrade the predicted likelihood of someone you actually want, in order to lower their status and decrease competition for *them*.
Remember that even if you solve physical scarcity, sexual scarcity will persist even if everyone becomes polyamorous (which I find unlikely)--people always want what they want.
The best thought on this is to show predictions for long term relationships.
The theory is that women actually want long term relationships, and they don't realize that by dating only the best men, they are more likely to get a short term relationship.
If we properly inform people how long their relationship will last, then women will choose slightly less attractive men (closer to their own level) to be stable long term partners.
I don't know if it's a "problem" so much as an "inevitable outcome given incentives and preferences of participants". Ultimately monogamy is definitionally 1 to 1 so it will mostly resolve itself, albeit leaving some women and men jaded and embittered (by being discarded and ignored, respectively)
I personally wouldn't waste my time on this dating community, my comparative advantage is awful given that I'm probably the median in many ways
Regarding the hospital bombing, as far as I know it wouldn't be fair to say the markets "got it right" and the evidence that it was fired from Gaza is still weak. We have the aljazerra video which NYT analysts say does not show whatever hit the hospital as Israel is claiming (https://archive.ph/Z3L81). We have an audio recording purported to be two Hamas militants blaming it on PIJ, which apparently does not ring true to Arabic speakers from the region (https://www.channel4.com/news/who-was-behind-the-gaza-hospital-blast-visual-investigation). The lack of a warning shot doesn't mean much since it appears from eye witnesses reports Israel has at least partly discontinued that policy after Oct. 7. We have vague assertions from Israeli and American intelligence with no info about how they came to that conclusion, historically that type of evidence has been.. unreliable. I'm not aware of the photos you refer to, but if they are the nail in the coffin I'd be surprised. Perhaps I've missed something, if anyone has more please let me know
This CNN article has a pretty good collection of the evidence in favor of it being a Palestinian rocket instead of an Israeli artillery strike or air bomb. There's a lot in there, but I found their analysis of photographs and video of the explosion site, particularly of the crater, to be worth quoting:
"Eight weapons and explosive experts who reviewed CNN’s footage of the scene agreed that the small crater size and widespread surface damage were inconsistent with an aircraft bomb, which would have destroyed most things at the point of impact. Many said that the evidence pointed to the possibility that a rocket was responsible for the explosion.
Marc Garlasco, a former defense intelligence analyst and UN war crimes investigator with decades of experience assessing bomb damage, said that whatever hit the hospital in Gaza was not an airstrike. “Even the smallest JDAM [joint direct attack munition] leaves a 3m crater,” he told CNN, referring to a guided air-to-ground system that is part of the Israeli weapons stockpile provided by the US.
Chris Cobb-Smith, a British weapons expert who was part of an Amnesty International team investigating weapons used by Israel during the Gaza War in 2009, told CNN the size of the crater led him to rule out a heavy, air-dropped bomb. “The type of crater that I’ve seen on the imagery so far, isn’t large enough to be the type of bomb that we’ve that we’ve seen dropped in, in the region on many occasions,” he said.
An arms investigator said the impact was “more characteristic of a rocket strike with burn marks from leftover rocket fuel or propellant,” and not something you would see from “a typical artillery projectile.”
Cobb-Smith said that the conflagration following the blast was inconsistent with an artillery strike, but that it could not be entirely ruled out.
Others said the damage seen at the site – specifically to the burned-out cars – did not seem to suggest that the explosion was the result of an airburst fuze, which is when a shell explodes in the air before hitting the ground, or artillery fire. Patrick Senft, a research coordinator at Armament Research Services (ARES), said that he would have expected the roofs of the cars to show significant fragmentation damage and the impact site to be deeper, in that case.
“For a 152 / 155 mm artillery projectile with a point detonation fuz (one that initiates the explosion upon hitting the ground) I would expect a crater of about 1.5m deep and 5m wide. The crater here seems substantially smaller,” Senft said."
https://www.cnn.com/2023/10/21/middleeast/cnn-investigates-forensic-analysis-gaza-hospital-blast/index.html
I'm not an expert here, but aren't there Forecasting sites and Prediction Markets and aren't they different things? Should they be compared? The mechanics of them are different.
Manifold markets having fairly flat DAUs is a pretty negative signal to me on finding a real niche (you'd expect it to be growing a lot). Curious what people think about this.
The thing about prediction markets is that it's generally a bad idea to play them unless either you are better at them than the average player or you're hedging something. Manifold being play-money-only removes the latter, which in turn makes the former really, really hard (you cannot, after all, have 100% of players be better than average).
I agree that hedging isn't a factor, but the problem with Manifold definitely isn't the efficiency of its markets.
It has flat DAUs punctuated by events that spike DAUs, and leave them seemingly permanently higher afterwards. There's no growth day-to-day, but big events bring users in, and it looks like growth in the medium term to me.
Yes, this is true. On a longer scale, we've clearly grown a lot. Last August we were around ~100 daily active users. Now we're 1800-1900. I still think it can grow much more.
I always respected Scott's choice not to do podcasts. Still do. (Though he has an invitation from Tyler Cowen.) Now, doing an "interview/pretend talk" on youtube seems in all aspects like a (video)podcast, just worse. But who can say "No" to persons in skirts/cilts. ;) - Whatever, we all get to do stuff once - I was Santa.
The implication that Israel so far hasn't been occupying Gaza is misleading at best. They simply chose the somewhat unorthodox "open-air prison that prisoners run" model of occupation.
Not actually *that* unorthodox; many Latin American prisons are more or less run that way, they just don't happen to be mainly populated by women and children and have over two million people.
Do they prevent people from leaving Gaza to go to other countries?
Yes, with help from Egypt.
I'm not sure? AFAIK no other country would take them in, I think if some country did say they'd accept meaningful numbers of Gazans Israel might be quite happy to deport its problem?
That's not an occupation. Withdrawal is not occupation.
Is infrastructure dependence and a total decades-long blockade not occupation as well?
Yes. Occupation is occupation. You have to physically be there to occupy something, whether that's a job or a country.
Blockades are a different thing that can also be bad. So is infrastructure dependence. Both of those things can be happening and can be bad, without an occupation occurring. And since 2005, Israel has not occupied Gaza.
In contrast, Israel does occupy the West Bank. Their soldiers are there, they determine the laws and control the justice system, they have bases and settlements all over West Bank. That's what you call an occupation.
And yet, per Wikipedia, "the UN, International Committee of the Red Cross, and many human-rights organization continued to consider it occupied as Israel military controls Gaza's borders, airspace, and sea access".
Do they get to define for the world what an occupation is? You can't occupy a place without being there. You can put a place under siege, blockade, embargo, exile, etc, but unless you're actually there it's not an occupation.
EDIT: If we're going to go beyond the typical definition of occupy to appeal to what organizations thing, the Regulations of the Hauge (an international agreement laying out the laws of war) defines an occupation as follows:
"Art. 42. Territory is considered occupied when it is actually placed under the authority of the hostile army.
The occupation applies only to the territory where such authority is established, and in a position to assert itself."
So with this definition we can see that the West Bank is definitely occupied: it is literally under the authority of the IDF, IDF determines and enforces the laws, and IDF is in a position to assert itself (military personal, bases, etc).
Gaza on the other hand, since 2005, has no Israeli military forces within it and Israel officially abandoned authority over the Gaza Strip. They do not make its laws, police it, or otherwise take a position of authority.
Now ever since Hamas took authority in the Gaza Strip in 2006 they have put significant restrictions on the ability of Gazan's to enter Israel, and they have blockaded the sea ports and airspace of Gaza. That may be a very bad thing! But it's not an occupation.
Now as Israel sends soldiers into Gaza and takes control of it in the days to come, *that* will be an occupation. If Gaza is occupied already, then that would imply that when the IDF storms in and takes control nothing has changed, which is ridiculous.
I suppose in the end somebody does get to define what it is. The alternatives seem to be either them or Israel doing it.
I don't understand how that's an occupation. It's a blockade. Israel tried occupying Gaza a while back and it was a very different regime (also, compare their occupation of the West Bank, which looks very different from "open-air prison that prisoners run")
There are reasons to think manifold.love is a terrible idea, e.g. Scott could bet on being in relationship with Aella not because he thinks this is at all likely to happen, but just to boost his status as measured by the prediction market. People being willing to loose money to influence the market makes it not work.
But, misgivings about the whole idea aside, it seems to be have been a big hit with the LGBT segment of the effective altruist community. It's looking like a Pride march or a gay bar in San Francisco. So it could still work, and the lack of cis-gendered women might not necessarily be a problem for that demographic...
Every innovation in dating I see makes me more and more grateful to be already married.
This! So much this! There's room for innovation and progress in a lot of areas, but there are plenty of other areas that we got right long ago and should just be left as-is.
Totally agree. These dating sites are one step away from putting people on Amazon as consumer goods. Soon people on fuckingTinder will have star ratings from previous daters, and also comments: "Looks dumb but that just turns out to be his facial conformation. Was decent to talk to and quite fun to fuck, so long as you're not bothered by a lush growth of back hair." "She's lively, but after a while you realize that the energy is emanating from a merely average level of intelligence. She's pretty, but smells too funky." I met the people I fell for through friends, or work, or school or sports. Many of them I knew for months or even years before we felt a romantic spark. Meeting for drinks with a stranger to audition each other, each of you knowing you need to come across as caring and responsible, but each of you also knowing that the chance one of you will eventually ghost the other is high . . . OMG, so brutal. Fuck that.
Nah the chances of ghosting specifically should be pretty low, how dare you not to resolve a market when the answer is clear!
I disagree with "got right long ago" - pre-app dating was awful and that's why there have been so many attempts to replace it. But I agree (with, thankfully, no personal experience) that most of the attempts seem to have made things worse.
I don't agree. Seems to me the primary reason why there have been so many attempts to replace it is because there have been attempts to replace (or augment) *everything* with "on the Internet" over the last 30 years, whether there was any need to or not, and dating is just another aspect of life that got caught up in that cultural wave. The Internet is our culture's shiny new toy, and we want to play with it everywhere.
Dating isn't awful as long as you go into it with reasonable expectations. The real problem is that we've spent the last 60-ish years dismantling the societal expectations that kept it manageable, so now it's difficult to find reasonable dating prospects unless you confine your search to some very specific and narrow pockets of people who reject those changes to our culture.
Well, I'm glad you had a better time than me.
I'm mostly curious to see how it blows up.
My most likely guess is the lack of women. But...if it makes rationalist gay guys happy, that's a net win!
Also, I'm not signing up (too old, too far, and sworn off relationships), but I love the guy whose profile is Asuka from Neon Genesis Evangelion (who's female, for the uninitiated). Way to show what you want. Doubt it's going to work, but who knows?
Hiring prediction markets will immediately run into disparate impact doctrine.
Unless they somehow match the demographic distribution of the population exactly by chance.
This is exactly what prediction markets are great at!
If the smart money knows that they have to match certain hiring practices to not run afoul of the law, then they will actively trade so as to make it match closely enough, no chance required.
I'm not going to bet that my company won't regret hiring someone if it's going to get them sued or whatever.
I can picture it now - one set of traders naively assessing candidates on their merits, and another running roughshod through the resulting prices to get them to add up to the demographic ratios they know are the minimum acceptable. End result: both types of information incorporated. Though the first group of traders will lose money if they don't wisen up and realise they have to hedge every merit-based bet they make with other bets that preserve the demographic ratios. Else they're just giving free money to the demographic traders.
It seems worth recalling the problems which actually did occur resolving the 2020 presidential election. I personally made a reasonable amount of money betting that Biden would win after the results had been announced. I can't recall the exact details now, but there was some date at which I thought the market ought to have resolved according to the published criteria, and there was some later date at which it actually did resolve. People were unhappy about both parts: the Biden bettors thought their payments were unjustly delayed (depriving them of the use value of the money) while a significant number of Trump bettors continue to believe that their man really won. People threatened to sue (https://www.insider.com/bettors-plan-to-sue-betfair-after-they-called-election-biden-2020-12), although I haven't heard anything about that since, so I guess it didn't go anywhere.
> People seem to trust the new Speaker to avert a government shutdown.
That's not exactly what's going on here. The market you linked shows the chance of a shutdown by the end of 2023 decreasing from ~55% to ~35% comparing before and after the speaker election.
But this market shows the chance of a shutdown by the end of 2024 dropping from ~70% to ~55% (still above 50%). And presumably Mike Johnson will remain speaker until the end of 2024!
https://manifold.markets/MarcusAbramovitch/will-there-be-a-us-government-shutd-9878106a2b1c
Some missing context here is that after election, Mike Johnson immediately announced his plan to *temporarily* avert a shutdown by passing a stopgap (continuing resolution), pushing the potential shutdown date in the new year (January or April).
https://www.msn.com/en-us/news/politics/mike-johnson-lays-out-appropriations-timeline-to-avoid-november-shutdown/ar-AA1iQ3iu
So, it's definitely a possibility that the market isn't showing so much "averting a shutdown" as "delaying a shutdown" (possibly only a delay of two months from November to January).
With that said, it is still notable that even the 2024 shutdown odds dropped after Johnson's election. And it would be nice to have a market for shutdown before April 30, 2024, to separate out a shutdown over the FY2024 budget vs a shutdown over the FY2025 budget.
"For real steam engines, that first niche was pumping water out of coal mines."
It helped a lot for that niche that the fuel was more or less just lying around, because the first ones installed were very inefficient. They quickly became more efficient as the owners could see that every shovelful going into the engine wasn't getting sold. But the first ones would not have made sense if the fuel had to be brought in from any distance.
So where's the "free fuel" and "quick feedback loop" for prediction markets?
It's more than that. Newcomen steam engines also had the shortcoming that they needed an ongoing supply of cold water to cool off the cylinder and condense the steam, as it was the vacuum pulled by the steam condensing that drove the power stroke of the engine. And water-flooded coal mines had plenty of that lying around, too.
Your point for prediction markets stands.
While we're nitpicking on the examples I want to point out there's a lot more separating an ancient aeolipile from a borderline-practical steam engine than just a useful application. There's about 1600 years of advances in materials science, precision manufacturing and basic thermodynamics separating them.
I was looking for this comment
I'm not an expert in the matter but it always upsets me when people call aeropyles "steam engines" even though the core innovations behind actual steam engines seem related to pistons with tight tolerances
This video about one of the first precision lathes is excellent:
https://www.youtube.com/watch?v=djB9oK6pkbA
> Hanson is less sure about this answer than the overall story, but he suggests hiring. You could create some kind of product that companies could buy and give their hiring managers at the beginning of a hiring round, asking them to predict which candidates would get good employee evaluation results or promotions at the end of X amount of time.
Honestly, that sounds horrifying, as it opens up the door to all manner of abuse and conflict-of-interest questions. Predicting who will win an election that millions of people participate in is one thing; "predicting" what will happen to people when you have some degree of control over their fate is quite another! (And yes, the "hiring manager" is not necessarily the *manager* manager who will be making these decisions on evaluation or promotion, but I bet they know each other and talk about work stuff on a pretty regular basis...)
That sounds like it will just be a fancier way of "what university did you go to? what major? your work history? are you going to fit in here with us or not?" that most hiring managers use anyway, it will just be packaged as New Better 'Cos AI.
Most hiring managers are likely to 'predict' that the candidate who went to Hahvaad is going to get good evaluations/promotions, whereas the guy from Clodkickers U... not so much.
He mentions that every better could start out with a positive stake, and have a floor at 0.
Wait, every better what?
Perhaps that should be spelled "bettor".
Maybe the first niche for prediction markets can be betting on the probability of celebrity divorces.
The New York Times has a history of getting things wrong-- that hospital in Gaza, weapons of mass destruction, Kitty Genovese, crack babies.... Is there money to be made in betting against the NYT?
Maybe they could have an internal market about whether particular stories will have to be edited or retracted or apologized for?
I think they mostly get individual facts right. They've got systemic bias in their selection and presentation, but I don't see a clean way to run a market about that. The big failures seem to me to be when their sense of righteousness completely overwhelms their commitment to factual accuracy. Those feel too rare to be the source of much income, and also they're the sort of thing where most observers are also polarized at the same time.
+1
It's rather ironic that you would say this in the very post where Scott praised their coverage of ManiFest.
https://manifold.markets/BenjaminIkuta/what-is-israels-mediumterm-plan-wha?r=QmVuamFtaW5Ja3V0YQ
I made a market as you were wondering.
"Normally this is a problem for Congress, election regulators, the courts, the media, and the American people. But if there are election prediction markets, then election fraud would indirectly become a type of financial fraud, and now it is also a problem for the CFTC."
Yeah, I imagine that if prediction markets set themselves up to be "we're like the stock market" rather than "we're like bookies", then financial regulation bodies, instead of Revenue or gambling regulatory bodies, will be the ones overseeing them. And if they have to decide on the outcome of elections, no wonder they don't want to be involved.
You're thinking "surely reasonable people won't need big decisions made", but you're forgetting there's a lot of unreasonable people out there. Who would have thought that the courts would have to decide if a piece of paper that was nearly, but not quite, all the way punched through counted as a vote or not? I don't blame the CFTC for not wanting to get involved in the financial equivalent of hanging chads.
For the love market, I would ordinarily advise user Nathan to ditch the left-side black and white photo, as "My hobbies include dressing up as Slenderman" wouldn't seem to appeal to most women, but if he's looking for love among the rationalists, who knows? Slendy could be just the guy they're looking for!
I thought this article on the manifold speaker market had some good takes. https://marketwise.substack.com/p/speaker-election-interviewing-the?r=28sbv1&utm_campaign=post&utm_medium=web&nthPub=1291
I tried to phrase two mid-term solutions for Israel as prediction markets: withdraw and occupy. No idea how phrase "puppet-government" in a way that I could reasonably resolve it.
https://manifold.markets/marketwise/will-israel-occupy-gaza-in-april?r=bWFya2V0d2lzZQ
https://manifold.markets/marketwise/will-israel-withdraw-again-from-gaz?r=bWFya2V0d2lzZQ
I think Manifold.love should *embrace* people betting up their own candidacy, rather than trying to restrict it. The basic problem with Tinder or Hinge is that men swipe right on ~everyone and women are overloaded with potential matches. But if you start charging people money (indirectly via mana purchases) to make bets, men would be incentivized to think carefully before "swiping right" and the cost of swiping will go up if the person in question is more attractive.
So, eventually, if Manifold.love gets enough traction, it will cost (say) $100 to get to the top of Aella's match queue, $50 for someone attractive but not well known or $5 for the average person. Thus no one will be overwhelmed with potential matches.
I'd make two changes to the process:
1. Let each person immediately resolve any market to No if they're confident they're not a good match with one of the candidates in the pool
2. Introduce a fee for men to create a profile, in order to encourage a better gender ratio. Perhaps something on the order of 1000 mana?
"The basic problem with Tinder or Hinge is that men swipe right on ~everyone and women are overloaded with potential matches."
The system already disincentivizes this. You only have a finite number of swipes, and presumably the algorithm penalizes you for swiping indiscriminately.
Isn't 1 already the case?
You have a finite number of swipes but also a finite number of hours in the day - if you're not using all your 'yes' swipes each day you're wasting them, and how long do you want to spend on the app rather than doing fun things in your precious free time?
It takes, what, a few seconds to look at a profile just to see if they look attractive?
So, personally I'm picky and absolutely spent time considering profiles before swiping, but I also really hated dating apps and didn't have much success with any except OKCupid. I think a lot of men out there have fairly modest standards for women they'll have casual sex with (as the joke goes, "1. Female, 2. Willing, 3. Breathing")
Re algorithmic penalties, most people don't know how those work, even if nerds like us are aware of them
Also, that's assuming there's no penalty for swiping like that.
TL;DR: Here’s how I’d use prediction markets if I was a dictator.
Prediction markets are about providing information. The steam engine use case for prediction markets is throttling the information ecosystem. Right now the way that something like twitter (“X”) works is that you make a divisive tweet, then other people fight about it, and you get visibility and more followers. You do it again and this time it goes out to even more people. A feedback loop of rage. You do this enough times and end up with a power law distribution of users-to-followers, and twitter has turned into a dumpster fire and everyone is screaming at each other and we can’t agree on who won the election and is supposed to have control of the nuclear football.
So instead, you have a site where you start with 1,000 points, like manifold, and this is your max number of followers. When you hit your limit, no one else can follow you. You’re capped. If you want the ability to have > 1,000 followers then you need to win some bets.
The rule doesn't mean that “You must follow superforecasters,” because superforecasters might be super boring and no one wants to follow them. Instead, the rule means, “Only super forecasters are allowed to have a disproportionate influence on the information ecosystem.” You’re only allowed to have a hundred thousand followers if you have a basic grasp of reality. Lots of people don’t want to do forecasting and that’s fine. You’d still have your default 1,000 limit and most social media users never get above that anyway.
Of course, this only improves the information ecosystem if people actually use it. Lots of people want rage. But if I was a dictator, I might just ban everything else.
Here in the US, maybe Elon Musk had an opportunity, but since the system was already in place, he couldn’t have done anything without ripping it apart. You’d have to start from scratch. Or maybe substack had an opportunity when they rolled out Notes.
What happens if you bet your initial mana and lose it? Do you lose your followers?
Yeah, they randomly lose followers.
Huh, I actually really like this idea. You would probably have to ban everything else to stop a competitor that panders to toxoplasmosis from taking over, but you did start the hypothetical with "if I were a dictator"
Yeah, a lot of flawed business ideas get more viable when you start with "if I were a dictator."
A few years back, I listened to an interview with the founders of airbnb where they were talking about growing pains in their early days. They complained to their investors that people weren't getting into the spirit of how to set up their pages so people would be enticed to rent from them. They wanted to create a culture of high-quality pages, but their guidelines weren't being followed.
"Where are your users?"
"About 80% of them are in New York City."
"You should go to New York and help them set up their pages, then."
"That doesn't make any sense. We want this to be a global system. Helping every user set up their page will never scale."
"Right, but when you're small you do things that don't scale."
So they went to NYC with a DSLR, telling users they'd been selected to have a professional photographer and a consultant help them set up their page. It worked, and the rest is history.
I think about this idea a lot when it comes to early growth models for a company or an idea. Whether it's prediction markets or impact markets, I think you have to be willing to do things that don't scale when you're small. And once you're willing to do things that don't scale, this opens up a lot of ideas for WHAT you could do to grow the idea that hadn't been options before.
Paul Graham of YCombinator (one of the investors concerned) has written an essay on this: http://paulgraham.com/ds.html
Better said than I could have. The question is: how to implement these ideas in prediction and impact markets? Or have prediction markets already crossed the scalar threshold?
Paul Graham's a great writer, and it always amazes me how far haters will go to misunderstand his very clear writing in order to dunk on him.
I've mentioned it in one of the open threads, but Fox News has this America Predicts [0] thing which seems like it deserves an honorable mention. I'm not aware of any other major mainstream news organization doing something similar.
[0] https://www.foxnews.com/americapredicts
I'd be interested in seeing prediction markets used to decide whom to invite to political debates. Rather than the current system of coming up with a bunch of ad hoc criteria, just have a prediction market for each candidate's odds of winning the primary or the general election contingent on them being invited to the next debate. Everyone over a certain threshold that increases the closer you get to the election (say, start at 10% and ramp it up to 30%) gets a spot on the stage.
It'll also be an excellent test case for arguments over how robust prediction markets are against manipulation by people with deep pockets and large stakes in the market's prediction.
That actually sounds like a sensible use for a prediction market! Let people estimate the chances of potential debaters and that should clear out the no-hopers who only got a spot for unknown reasons (why on earth, or how on earth, did Julián Castro get his five minutes in the spotlight?)
It also lets the punters make free money off of the people with big egos deep pockets and no hope - those politicians subsidise the market!
"Prediction markets" have reached stage 6 in Scott's "every cause wants to be a tribe" cycle (https://slatestarcodex.com/2016/04/04/the-ideology-is-not-the-movement/)
Plasma's survey of manifold users found a lot of furries.
Bit of an omission for manifold.social here ... manifold,social asks you for your preferences re. M/f/nb/trans m/trans f, but no questions on how you feel about furries.
This is on the Internet .. of course there will be furries.
A lot of people who believe the American government lies about everything are sure that Israel didn't bomb the hospital because the American government says so
My reaction to the legalization talk was "well, what did they expect?" Trying to pass off gambling on elections as "hedging" just sounds disingenuous on top of the existing problems.
Next Mantic Monday maybe you should look into markets on the Tesla Strike in Sweden. I think there is probably alpha in those. Remarkable situation really, historically the unionizers usually win by getting their demands met, but if the CEO is outspoken anti-unionizer Elon Musk will it truly go down that way?