Resolves YES if by the end of 2024, Manifold officially announces their intention to shut down. Otherwise resolves NO.
The date of the planned shutdown does not matter for this question. For example, if in November 2024 Manifold announces they will shutdown in June 2025, that resolves YES.
@MartinRandall Hmm, interesting question. I think this question probably should resolve only on an unconditional announcement. As an example, PredictIt basically announced that they would have to shut down unless they could get a reversal or injunction against the CFTC, and then they did get a (temporary) injunction, so they aren't shutting down yet, and I don't think it would be accurate to say that PredictIt announced their intent to shut down.
Thoughts?
@jack It's awkward because Manifold founders know that probabilities are "never zero". There's always a chance!
https://www.predictit.org/platform-announcements
This doesn't read to me as "we intend to shut down unless...". But I'm not sure how to operationalize that.
The original announcement text can be found here: https://manifold.markets/jack/will-predictit-still-be-operating-m. I agree that it's not super clear, but it seems to say that they intend to operate markets until the CFTC's deadline (and stop after) unless something changes.
@jack I don't read the original PredictIt announcement as expressing an intention on what they would do after 2023-02-15, and I imagine that's deliberate.
Whereas I expect an announcement like this rando crypto one I found would resolve NO: https://blog.ethereum.org/2022/06/21/testnet-deprecation
(in this case the shutdown appears to be happening as announced)
Some grey area between those examples.
@Gen it pays on announcement so YES bettors can quickly spend their mana on donations, assuming there is a window to do so.
Also the site could shut but reach a deal to transfer mana into Metaculus points, for example.
@mqp How much of the money burning is doling out free mana and shipping new features vs. keeping the website online?
@mqp (crossposting from Discord) Not true. I think we'll be able to raise, but even if we don't we can transition to a lower burn rate, grant-supported model if we absolutely have to. The chance of Manifold truly dying is actually very low now.
@MichaelWheatley Right now we burn thousands of dollars a month just to run the website. I expect that later this year after doing low-hanging-fruit cost savings, this might be "only" about $2k/mo. It could go lower with actual effort -- obviously there's no inherent reason that Manifold needs to be expensive. But I don't think anyone is going to do that effort. Anyone would be better advised to rewrite from scratch, if they really wanted one that was cheap to run.
But on reflection I agree with SG in that we can probably manage to run it on grants at this point, since a lot of people like it a lot.
@mqp how much are the server / firebase / services costs now? $2/month doesn't seem high at all for around 2k weekly active users.
@mqp If there were any paid staff in low-budget mode, that would presumably be the major budget item, right?
@firstuserhere A hack wouldn't count, because that wouldn't meet the "officially announces" wording.
@jack It should still trade way below it's actual probability. I think "Anthropic Bias/Reasoning" comes closest, but not sure what to call this issue.