Re:Truthcoin: Trustless, Decentralized Bitcoin Prediction Ma...
by Humjob on Saturday May 03, 2014 @01:13PM
(#7691513)
From what I gather, in a weighted vote PM system the 'weighting' of the vote (AKA, the expected likelihood of the event occurring) is based on how much money you invest for or against the event occurring. Hence, 999 'for' votes of $1 are worth less than a single $1000 vote 'against.'
Re:Truthcoin: Trustless, Decentralized Bitcoin Prediction Ma...
by gwern on Saturday May 03, 2014 @03:32PM
(#7692080)
It's not quite that simple: it's not just a weight based on voting, since that's obviously gameable by anyone who can accumulate enough coins to guarantee a win on a single contract, but it's a weight based partially on how each voter reflects the majority in all the other contracts. So anyone trying to corrupt a particular contract has to deal with the combined opposition of everyone who wants weight for every other contract.
Re:Truthcoin: Trustless, Decentralized Bitcoin Prediction Ma...
by maaku on Saturday May 03, 2014 @08:28PM
(#7692862)
So this relies on an identity system?
Re:Truthcoin: Trustless, Decentralized Bitcoin Prediction Ma...
by gwern on Saturday May 03, 2014 @11:12PM
(#7693172)
No more than Bitcoin does.
Re:Truthcoin: Trustless, Decentralized Bitcoin Prediction Ma...
by vbuterin on Sunday May 04, 2014 @06:58AM
(#7693876)
Bitcoin does rely on an identity system; it's called proof-of-work. It counts hashing cycles rather than humans, but it's close enough since the final goal is decentralization and not perfect equality. Here proof-of-work or proof-of-stake are also equally necessary for bootstrap purposes; otherwise you can use whatever mechanism nodes use to get into the system in the first place ten thousand times instead of one time and thereby massively magnify your reward.
TLDR: proof of voting is a good consensus algorithm, it's not a good distribution model.
> through a weighted vote based on present and past consensus
How can Truthcoin protect against votestuffing?