News for hackers. Stuff that matters.
Re:Karma-rank (Pagerank for social sites) by wehriam on Monday June 30, 2008 @11:35PM (#232674)
This would create a feedback loop. Users that exhibited a certain type of behavior would quickly amass a large amount of karma.

Unfortunately Pagerank is not inherently resistant to spam. To extend your metaphor, this would be good if we had a lot of "Ms. Wikipedia's" but bad if we had more than a few "Mr. Linkspam's."

It would discourage new users, as they would likely never catch up to early adopters. The A-List blogger phenomenon is a good example of this. Big, popular blogs stay big and popular. While possible, it is extremely difficult to break in to the upper echelons of blogging.

It would also reward those who post and vote on topics that are mainstream, at the expense of diversity in the community.

The normalizing effect might be particularly well suited for some specific communities - financial advisers for example, or those dealing specifically with popular culture.

Re:Karma-rank (Pagerank for social sites) by ntoshev on Tuesday July 01, 2008 @02:50AM (#232774)
The social sites are much more dynamic than the web. This means, for example, that people high in the ranking like pg would upvote newcomers and they would instantly get a karma boost.

Pagerank is usually log-scaled because of these feedback loop effects.

Re:Karma-rank (Pagerank for social sites) by Malcx on Tuesday July 01, 2008 @08:12AM (#232867)
>>It would discourage new users, as they would likely never catch up to early adopters.

Adding a time based decay to the rank could address that. I.e. my vote for you 1 month ago is worth less then the vote 1 min ago...