Suppose it’s true that mathematically speaking, a trustworthy reputation system is exactly as good as genetic kinship in promoting prosocial behavior by group members.
Boiling that down into an online model, reputation systems are critical to social software, which depend on their users behaving in positive ways.
The motivation of A to add value to an online commons depends critically on A’s belief that good users are labeled as good and bad users as bad.*
It also follows that any social software can benefit simply by publicizing its reputation system.
The fact that eBay’s reputation system is digital and public while wikipedia’s reputation system is complex and unevenly enforced may account for the more frequent trollish disruptions at wikipedia.* *
A portable online reputation system could be somewhat like the “online citizenship” model proposed by
Ken Jordan, Jan Hauser, and Steven Foster, and expanded at today’s Symposium on Social Architecture by Kaliya Hamlin. I look forward to re-seeing Kaliya tonight, to ask her about it.
* Your public reputation score doesn’t have to jeopardize your right to privacy for personal data like real names, etc.
* * The reputation systems used by eBay, wikipedia, Amazon, SlashDot, etc. are very different. (So are the kinds of behavior each hopes to promote.) The type of reputation system doesn’t matter, however–just its accuracy, as perceived by users. The only detail of desired behavior that matters is its cost-to-benefit ratio.
Technorati Tags: corantessa