4 ms·

That's mostly true, with a caveat. By default the oracle is not decentralized, but what is decentralized is the "backstop". If I create a market, I can also be

That's mostly true, with a caveat. By default the oracle is not decentralized, but what is decentralized is the "backstop". If I create a market, I can also be the one to report on the outcome of the market (versus it being decentralized in the original plan). What they introduced is the concept of a "dispute". So if I report on a market you can dispute it and say "Hey you are lying" and if enough people agree that I lied, you get a portion of the market fees. So you are correct, it's no longer decentralized, but it's also not centralized. Not perfect, but still pretty good.

JumpCrisscross 8y ago
> if enough people agree that I lied, you get a portion of the market fees This creates a host of post-closing political incentives. You are no longer betting on an event outcome, but on a consensus process.
- pjc50 8y ago
  It's consensus all the way down: ultimately a cryptosystem is what the majority of the voting/staking/hashing nodes say it is. Remember the DAO reversal?
- DennisP 8y ago
  This is true, so it comes down to how well the consensus process reflect reality. It's a pretty interesting process: there are multiple layers of dispute escalation, and the ultimate backstop is that Augur splits in two. You keep REP only on the side of the split you voted for. The theory is that the version of REP that reports the truth will be more valuable on the open market, and that the split is a credible threat which keeps everyone honest, so it will seldom actually happen.
- sigstoat 8y ago
  > > if enough people agree that I lied, you get a portion of the market fees > This creates a host of post-closing political incentives. You are no longer betting on an event outcome, but on a consensus process. even on scicast, which was running (largely) apolitical questions and had impartial academics resolving the questions, you needed to be aware that you weren't betting on the outcome, you were betting on how somebody would decide the outcome. it tended to be more of an issue the fuzzier the question, or the the punchier and more news-headline-y the short version of the question was. (which also made it more likely to attract the newer, less sophisticated users.) seeing the problems that even scicast had with poorly worded questions, i don't have much hope for augur.
- zodiac 8y ago
  In what real-world prediction market / sports gambling / poker tournament etc are you betting on anything other than the output if some consensus process?

JumpCrisscross 8y ago

> if enough people agree that I lied, you get a portion of the market fees This creates a host of post-closing political incentives. You are no longer betting on an event outcome, but on a consensus process.

pjc50 8y ago
It's consensus all the way down: ultimately a cryptosystem is what the majority of the voting/staking/hashing nodes say it is. Remember the DAO reversal?
DennisP 8y ago
This is true, so it comes down to how well the consensus process reflect reality. It's a pretty interesting process: there are multiple layers of dispute escalation, and the ultimate backstop is that Augur splits in two. You keep REP only on the side of the split you voted for. The theory is that the version of REP that reports the truth will be more valuable on the open market, and that the split is a credible threat which keeps everyone honest, so it will seldom actually happen.
sigstoat 8y ago
> > if enough people agree that I lied, you get a portion of the market fees > This creates a host of post-closing political incentives. You are no longer betting on an event outcome, but on a consensus process. even on scicast, which was running (largely) apolitical questions and had impartial academics resolving the questions, you needed to be aware that you weren't betting on the outcome, you were betting on how somebody would decide the outcome. it tended to be more of an issue the fuzzier the question, or the the punchier and more news-headline-y the short version of the question was. (which also made it more likely to attract the newer, less sophisticated users.) seeing the problems that even scicast had with poorly worded questions, i don't have much hope for augur.
zodiac 8y ago
In what real-world prediction market / sports gambling / poker tournament etc are you betting on anything other than the output if some consensus process?