Idea for cpanratings

djberg96 on 2003-10-21T15:39:39

This journal entry is in response to Ethan's journal about CPAN authors rating their own modules (which I agree is lame).

My solution (which I mentioned in a reply to Ethan's entry but wanted mentioned here) would be to require a minimum of 3 (or whatever) ratings before a rating actually becomes "visible" for that module. This prevents that silliness and helps guarantee a more balanced review.

Just an idea.


still...

spur on 2003-10-21T17:23:07

Suppose the author enters the site from various machines and rates the module...

Such problems are inherent to all rating systems on the net, unfortunately...

Re:still...

djberg96 on 2003-10-21T18:01:50

Well, if someone is really willing to go out of their way to do something like that, it ought to tell you a lot about the author.

Generally speaking, I think this falls into the "more trouble than it's worth" category for all but the most determined (not to mention kooky) authors.

What's the problem?

petdance on 2003-10-21T20:45:01

Where's the problem? Surely you're not suggesting that self-ratings will allow distortions of the otherwise accurate rating system.