Alright, lesson learned. Perfection is not an attainable goal.

I'm glad Manifold lets us re-resolve markets now.

I believe this should resolve NO due to this market. Any concerns?

https://manifold.markets/IsaacKing/will-i-block-markingraham-by-the-en

@IsaacKing none at all

@IsaacKing resolve this?

Here's one that could potentially count:

https://manifold.markets/IsaacKing/this-market-resolves-to-whatever-pe

A user checked the API and told me what percentage it should resolve to. They seemed trustworthy, so I said I'd accept their percentage, and gave people a day to express disagreement. No one did, so I resolved to that percentage. Shortly afterwards, they told me their percentage had been incorrect. (I don't know whether this was an intentional ploy on their part or an honest mistake, though either way I won't be trusting them as much for market resolutions in the future.)

I'm not sure whether this should count as incorrect. I gave all market participants about 21 hours to notify me of any errors with my suggested resolution, and none of them did so. But apparently the resolution did end up being incorrect. (At least, according to that user. I haven't checked their math myself, which I would recommend doing before you resolve this market to YES based on this.)

I lean towards saying "not incorrect", since there's an implied "speak now or forever hold your peace" when I make a "this market will resolve to X unless somebody says something" declaration. But I don't think it would be unreasonable to count it.

@IsaacKing That particular resolution is extremely confusing right now and I'm going to withhold judgement.

But in general, resolving to what you and the entire rest of the world thinks is correct, and then discovering later that everyone was wrong, is still an incorrect resolution. This is why we need /jack/will-it-be-possible-to-undochange-t .

Just like on the previous markets, I'm predicting a high chance of at least one mistake happening, given that you make and resolve a lot of markets.

One clarifying question: suppose /jack/will-it-be-possible-to-undochange-t is implemented, which the markets think is likely. If you make a mistake and later correct it, does that count for this market or not?

My opinion: for the question we care most about, correcting a mistake in a timely fashion after the mistake is discovered shouldn't count. But a question on whether a mistake will be made and later corrected is also interesting, although carrying a very different interpretation.

Either way, it makes predicting on this market complicated because a lot depends on when fixing resolutions is implemented.

@jack For simplicity, I'll exclude that. If I resolve a market incorrectly and then later correct it, this resolves YES.

(Of course now you have to factor in the chance that an "edit resolution" button would make me start caring less about getting it right the first time. But that still seems simpler than the alternative.)

@IsaacKing Did you mean YES or NO? The way you phrased your answer makes me think you meant NO.

@jack I meant YES. The first click is incorrect, regardless of whether I'm able to fix it later.

I don't think it would be fair to traders in this market to give the other answer, since that would just turn this market into a market on whether Manifold will give us the ability to reverse our resolutions.