The everyday blog of Richard Bartle.

RSS feeds: v0.91; v1.0 (RDF); v2.0; Atom.

Previous entry. Next entry.

5:16pm on Thursday, 28th July, 2011:

Reputation Systems Revisted


Facebook has Like buttons and Google+ has +1 buttons. OK, fair enough, they're pretty much the same thing, right? Well, they're almost the same. The difference is that Google uses its +1s as data for its search engine. A site with a lot of +1s will get higher up the listings than would the same site with fewer +1s. This means that getting a +1 is actually worth something to a site. This means that sites in need of +1s might be willing to pay for them. Therefore, an industry has sprung up to sell +1s.

This is not the way to go about things. What we have here is a basic reputation system, which has been shown many, many times to be easily be gamed and therefore practically useless. A change of the use of the data could, however, translate it into something far more useful.

Some five years ago I described a reputation system that could be used for virtual worlds. It actually has a wide variety of uses, and this is one of them. Here's how it works...

Users rate things +1 just like they would anyway. They're not used as a general pool, however, they're used specific to individuals. If I give a site a +1 and you give the same site a +1, that doesn't mean the site is good in an absolute sense: what it means is that you and I have the same tastes. If you and I have +1ed many of the same sites, then the chances are that if I come across a site you've +1ed, I'll like it too. So, the search algorithm doesn't add points for sites that have more +1s, it adds them for sites that have more +1s from people who have +1ed other sites that the individual doing the search has also +1ed. It works better if you can -1 a site, too, but even without that it's still virtually impossible to game.

This is not difficult to do. Maybe Google are planning on introducing it once they have a big enough stash of +1s to be able to do the matching?

Latest entries.

Archived entries.

About this blog.

Copyright © 2011 Richard Bartle (richard@mud.co.uk).