GISS says 2020 was hotter. "Effectively tied" does not mean the same thing as tied.

I disagree. If "within the margin of error of the analysis" is not a tie, then what would be? This is them saying that they literally can't tell by their analysis which year was hotter.


White wins more at higher levels of human play, and also now when AI plays with each other.

No. White wins more than black at high levels of play, but the most common result is a draw. This is even more pronounced with computer-vs-computer games.

If we accept going after the limit here, it should result in White winning always.

Top level play, even among computers, is not perfect play, so I don't think this is a completely valid line of reasoning. But if anything, it should count towards the game being a draw for the above reason.

I think it is a tad bit confusing that the title here is the complement of the resolution criteria.

This is laughably overoptimistic. For one thing, the technology for ISRU isn't there yet. You'd need a transfer window to test fuel generation on Mars, another one to demonstrate the ability to return to earth and yet another for the actual human mission. That means a technology that doesn't exist yet should be developed *and sent to Mars* by 2024 at the very latest to get even a single attempt at demonstrating the possibility of a return journey. Fat chance. And that's but one of many hurdles that has to be overcome. Y'all put too much faith in Elon. ...

Except for point 7, I'm pretty sure this can easily be achieved by a program that is not "intelligent" at all. For example, tracking the community prediction with a calibration-curve correction should do the trick. Even tracking the community prediction with a small random fudge will have a reasonable chance of success.

Half of GDP!? I'm not sure that's even possible. Only slightly above 1% to hedge against the chance that I seriously misunderstood the question.

Tunnel digging is not exactly a new technology, and the Boring Company aren't really adding anything revolutionary to it. Why then should one believe that they can magically be several times more efficient than anyone else?

Routine poisson calculation based on the data in the description. About 0.216 qualifying events per 8 month period, that's exp(-0.216)=0.806 chance of no qualifying event.

@(kokotajlod) I presume you are referring to their [faq](https://www.boringcompany.com/faq): > * Increase TBM power. The machine’s power output can be tripled (while coupled with the appropriate upgrades in cooling systems). > * Continuously tunnel. When building a tunnel, current soft-soil machines tunnel for 50% of the time and erect tunnel support structures the other 50%. This is inefficient. Existing technology can be modified to support continuous tunneling activity. > * Automate the TBM. While smaller diameter tunneling machines are automated,...

I'm putting 99% here, even though I don't actually believe white wins. The game is probably objectively a draw from the starting position - actually I'd say I believe that with >99% certainty. However, it would be much easier to (weakly) solve the game if there was a winning strategy; you can reduce the search space by up to a square root. Chess will almost certainly not be solved before 2080 - or probably ever - but if it is, it would almost certainly be a forced win. Black having a winning strategy is not even worth considering.

@(nhuvelle) "Market cap" (a term unfairly borrowed from finance) isn't really a good measure of anything for something as illiquid as Bitcoin. It is simply the current price times the number of Bitcoin in circulation, which isn't really a number that means anything. For something like a stock, that would be a meaningful number (it is proportional to the amount of money you would need to buy the company in question), but it doesn't represent anything analogous for bitcoin. If, for example, everyone sold off all their bitcoin on the same day, nothing near ...


There's at least some indication that the number of present cases is much larger than the number of confirmed cases, maybe as much as 10x (or more?), due to most people not getting very sick (ie. not sick enough to see a doctor and get tested). That would also mean a 10x lower CFR.

— edited by sbares

@Jgalt I'd hardly say 25% is low. Heck, I was at 15%, and I still don't think that was a wrong judgement.

It will resolve ambiguously [...] if the Prize is awarded for a proof that the Riemann hypothesis is undecidable in ZFC set theory.

I don't necessarily disagree with this, but it is worth noting that RH is a equivalent to a sentence (for example: "for all integers greater than , the sum of the positive divisors of is less than "), so if it is unprovable then it is true.

@zxi Yes. This is yet another example of a mismatch between question title and criteria. The title is "Will the US-China trade war still be ongoing by November 2nd, 2020?", but the criteria are more like "Will the US-China trade war be further escalated between August 2nd and November 2nd, 2020?"

Disclaimer: I have nothing to gain or lose by the resolution of this question.

@Jgalt That scenario is even less likely. Real currencies with real, non-speculative valuations don't see sudden order-of-magnitude jumps in value like that, especially not the USD.

This concept fails the smell test for me. It's probably not a coincidence that this idea has not been implemented in any useful way for several decades. But since the resolution criteria only requires someone to throw money after it - not that they actually get anything for their money - there is still a significant chance of positive resolution.

I think a large underestimation is more likely than a large overestimation, but a median of zero is probably a sensible guess, so I tried to create a distribution like that with quartiles -3, 0, 2 and a long tail to the left.