Skip to main content
MANIFOLD
Yes versus No: Experimental resolution by proof of work
17
Ṁ350Ṁ1k
resolved Apr 14
Resolved
NO

To vote yes or no, post a comment consisting solely of alphanumeric character (A-Z, a-z, 0-9) which starts with either Y (for yes) or N (for no) and ends with POW. Please do not include any other text in your comment. For example, a valid comment voting for YES could be: "Ysd3kjfh2jasdfPOW" (without the quotes). There is no limit to the number of votes each user can submit.

Out of all the comments submitted by close, whichever one has the lowest SHA256 sum will determine which direction the market gets resolved.

Example of checking the SHA256 sum of a given comment:

$ echo -n Yff2d799f213f3670bf88081e3c095a5b647fab297ebc95c6e2d29712f63934a9POW | sha256sum
000001afa5c4bd8a20d1ccb3399e979758425b24745dd15c81ec750d99e9b57b  -

And here's a reference implementation in Python of the proof of work procedure. Obviously this example is extremely inefficient compared to a fullly optimized approach.

from hashlib import sha256

side = "Y" # change this to "N" if you want to vote NO instead
seed = "whatever" # change this to any value you want so everyone doesn't search the same space
suffix = "POW" # leave this as-is
best = None

while True:
    hashed = sha256((side+seed+suffix).encode('ascii')).hexdigest()
    if best is None or int(hashed, 16) < best:
        print(side+seed+suffix, hashed)
        best = int(hashed, 16)
    seed = hashed

I promise that I haven't put any significant effort into "pre-mining" this question (just 30 minutes or so to test whether the procedure works).

Market context
Get
Ṁ1,000
to start trading!

🏅 Top traders

#TraderTotal profit
1Ṁ147
2Ṁ38
3Ṁ26
4Ṁ18
5Ṁ15
Sort by:

Closed, seems like we have no last minute submissions :(

I guess this resolves NO unless anybody thinks I overlooked something?

predictedYES

@A how disappointing. I was expecting some sort of last minute shenanigans

predictedNO

It wasn't for the lack of trying on my part

NciaoPOW

@pizzzzzza d78d90cbacf2d9e6820a0836f079b0a7bd8269dec79c5f1cab47896658764072 (not in the lead)

predictedYES

I believe there's an advantage in this market to commenting as late as possible, so let me clarify the exact deadline. The market closes April 12th 11:59 PM CDT, which corresponds to Unix timestamp 1681361940. If there's ambiguity about whether a comment is before the deadline, I'll check the API for the comment timestamp (at whatever precision manifold provides) and compare whether it is strictly less than that cutoff. Any comment with a timestamp >= 1681361940 will be ignored.

predictedNO

Nfd71d46ccda488f169cab2fb3a1e22a3817597e7fab73006137aa863ae9bcc0
0POW

predictedNO

@AlexbGoode Not great, is it?

predictedNO

Ignore the newline please :(

@AlexbGoode 00000001a00d16b27c1b54ed7af909ebd431b9df169318e64bfd2a57ac3bd3db (doesn't take the lead)

Please be careful in future to make sure your exact comment text is a valid submission. I want these to be parsable by bots and in future I'm only going to use the exact text to determine the final winner, not any external instructions like "remove whitespace/newlines"

Doesn't seem like a big issue, Alex could have just reposted it correctly.

predictedYES

@IsaacKing Right, he is welcome to do that. But I want to be clear on the official criteria in case there's a dispute where someone posts something in the last few seconds of the market that is intentionally invalid to confuse bots and then tries to "clarify" it later. Only things that are posted with the correct format prior to market close will be included in the final resolution criteria.

Yffee22ce8af3ccf7282cfc86be0c75aa409034b05b974a275760436d01756321POW

predictedYES

@firstuserhere (just testing, not the lowest)

@firstuserhere Yep: 00000002f85871792dbe0771eb96232a88182d06040e7b7a0de516caa03edb8c

I used a couple of hundred core hours without finding a better solution. Without some effort it will be hard to beat Train Jumpers answer.

NLevinaFinkelsteinIsNotPretty1925703430POW

predictedYES

@IsaacKing 🙉 🙈

@IsaacKing 00000000c7ff8a757fc82b83f9c5411a628068fad6cb214441df58f4527bd0b7 (does not take the lead)

@A Yeah I let it run for a few days, but I don't think I have enough computing power to overtake the current winner before the market closes. (My implementation was probably not very optimized either.)

@IsaacKing Yeah, I'm curious to see if anyone ends up jury-rigging a Bitcoin miner or something to mine this market...

YuwuAnyCuteNerdyBoysOutThereAsterisksDroolsAsterisksUwUPOW

@levifinkelstein cb4224b908204c8cde74b66f2713018761f6025ce54ec98919e6aa0fe3818025

Nc1b28e7850bf7a0a548e271ad619510b7c3f1f3c1255db3a4949acadc1793877POW

@TrainJumper 00000000000137f9938c2c8e980f45ebcf749a93decf9b66648ac39bff30dc4a

@A I believe this one is in the lead as of April 8

@TrainJumper After the market resolves, please share how you found this one. :)

@TrainJumper Seems like this is the winner, congrats!

@A @IsaacKing Yay! You might be disappointed to hear that I didn't do anything too special to find this, though..

I threw together a multi-threaded version of the naive Python implementation from the market description, used b"Nthe secret pc seed string {N}POW" as the search string (replace {N} for thread ID in 0..16), left it running on my PC for I think around 3 hours?

I noticed the result seemed exceptionally lucky so I stopped searching beyond that point.

@TrainJumper specifically N=7, and 4 core hours @3.9GHz.