I'm considering buying some shares on the Craig-Wright=Satoshi market. I'm not sure I understand how the question will be resolved though. What Oracle determines, and how, whether it is yes, no, or still unknown? Short of Wright admitting it was a fraud or another Satoshi with undeniable evidence stepping forward, I just can't imagine how it would be anything other than "unknown" at the end of the period? How would this be resolved if the outcome is still in dispute? Does it default to "unknown?"
I had a similar thought the problem with the "unknown" outcome i think this is heuristically similar to the problem of disproving a negative claim. Russell's teapot claims there is a teapot between earth and mars.
So based on this train of reasoning I can see it resolving to either yes or no:
1) It only resolves to yes if there is evidence, so between now and the resolution date it would have to be proven that Craig-Wright is satoshi/or used the name...this would probably require something like validation and use of his bitcoin forum login, a private key that isn't backdated or maybe someone like Nick Sazbo coming out and saying he is. (it would be all over reddit at this point and a myriad of other tech cites)
2) As currently all the pieces of "evidence" listed in the three sources provided in the market have been proven faulty or faked it seems that there no longer exists any evidence for a link at all.
These along with a half dozen posts on reddit basically ripped all the evidence to shreds and actually gave proof that some of the "evidence" was manufactured after the fact. This points to a clear intention to create a hoax...the most damming of which, are in my opinion, the public keys created with an encryption that did not exist at the time those keys are dated.
Basically as there is now no evidence that links Craig-Wright to Satoshi so in my mind it actually resolves to "no" not to "unknown." For example I could create a key at this point and claim I used the pseudonym in 2009 but that would not resolve to unknown it would resolve to no as well.
So basically as I see it as right now there is no evidence for yes it actually resolves to no, unless between now and resolution date new evidence emerges.
It could also resolve to unknown still but that would, i think, require a new piece of evidence to emerge. However, since the existence of evidence that was manufactured points to a hoax, this is actually evidence for no.
It could also resolve to unknown if between now and resolution it was somehow shown that the fake evidence was created by a 3rd party for some other reason and not by Craig Wright himself. For example, he actually did use the pseudonym and someone else manufactured the evidence to force out him as Satoshi; this would create some believe that someone wanted to out him implying he may be Satoshi. However, no such evidence as this exists as of today.
Just my 2 cents; and this is more fun than writing my thesis Hit me back with your thoughts!
Yeah, I agree with everything youv'e said. It seems like a clear hoax. But the difficulty, as you pointed out, is proving the negative. Does the fact that no one has found evidence of him using the name prior to 2010 mean that he never did? Basically, I think it is a pretty clear hoax, and you give a strong overview of the evidence for hoax. But my real question is who decides that this is a settled matter? I just signed up for Gnosis today, so maybe I'm just missing that. Does the question ever get "settled," or is the final answer just the state of the market on closing day?
In other words, is my betting challenge here trying to guess if more people will start betting that Wright=Satoshi before closing day or whether an oracle of some sort will give a final decision. If the former, no way can I see anything other than "no" carrying the day. If the latter, "unknown" seems undervalued to me.
Your betting is trying to guess what you think is the answer. If you bet no and there is no proof to lead to either conclusion, (which at the current moment is all speculation) it will be resolved unknown.
So the market resolves as the situation is seen on the resolution date. I believe, unless new evidence emerges that it will resolve to no.(and again this is just my opinion on how it should play) (and maybe i will put another 5 either on the wager having convinced myself of this at this point)
The existence of fabricated evidence, is in fact evidence of a hoax. So i guess I would disagree with your point that it is "all speculation" . http://motherboard.vice.com/read/satoshis-pgp-keys-are-probably-backdated-and-point-to-a-hoax
points out the keys used by craig wright were back dated to a time before the technology existed that could have produced said keys, that is not speculation, that is proof of impropriety.
The absence of any evidence pointing to yes, plus the existence of evidence pointing to a hoax is actually a resolution to no not unknown. This is because no information exists that points to yes and some evidence exists that points to no.If evidence of a hoax, and some evidence which has not been disproved existed then it would resolve to unknown. However since no evidence exists that has not been disproven it resolves to no. Conversely, if we existed in a universe where those articles had come out and none of the evidence had been challenged, but at the same time no confirmation to yes had been given then it would resolve to unknown.
Consider the situation on an infinite time continuum, if between now and the end of time no more information ever emerges then at time infinity one would logically resolve to no. So basically since the resolution date is the end of time with regard to the market, unless there exists a piece of evidence which disputes a resolution to no emerges, then the market should resolve to no.
This is a clear example of the challenges associated with an oracle, cool stuff
So who makes this decision and the methodologies make a big difference here. Can someone from Gnosis explain this?
I know this is still in testing, but the oracle information really should be visible in the event.
Does this count as evidence towards no?
Now that the market has closed, I would be interested in the thinking of the oracle as to the difference between the no and unknown potential for resolution. Clearly this was a good test of the system but as noted above by another user maybe the rational of the oracle should be included in future markets such as this one.
Thanks and have a great day,
I agree. I eventually bet on unknown because in my view neither proposition had been proved. I believe Craig Wright is not Satoshi. But the bet did not ask that question. It asked a much more specific qiuestion of whether or not he had used the pseudonym Satoshi Nakamoto prior to 2010. Since conclusive evidence of scam was not produced (though some pretty strong indicators did emerge), and nothing emerged showing specifically that he had used or not used the pseudonym prior to the mentioned date, it apparently seemed to the market that unknown was the best answer since that is the outcome that won the most bets.
That is fine, and I agree with that outcome. But I do wish there had been more clarity about the terms of resolution when the market was created. It should have been more clearly stated what evidentiary thresholds would have to be met for either the yes or no outcomes.