Online Reputation Systems: how to design one that does what you need
sloanreview.mit.edu
sloanreview.mit.edu
That said, it seems to me that sites like HN vastly over-reward early posters to popular threads. On the other hand, this is only a bug in so far that it leads to problematic behavior. I don't see people taking advantage of it (yet). I'm more concerned with the problem of under-rewarding late, thoughtful replies. I think this is a critical problem with the Stack Exchange (StackOverflow) platform.
My thinking on late (and hence low-exposure) contributions is more along the lines of weighting. If 10 people read a comment and 5 upvote it, it should be rewarded similarly to a comment with 100 reads and 50 upvotes.
Just yesterday I noticed that OReilly recently published Building Web Reputation Systems (Randy Farmer, Bryce Glass)
http://oreilly.com/catalog/9780596159801
http://buildingreputation.com/
Randy was consulting with a company I was at a couple of years ago and while the two of us certainly didn't see eye-to-eye on a lot of things, it encouraged a lot of thought and debate.
I'll be ordering the book today.
There can be no 'reputation' without an answer to the problem of pseudonymity.
People then focus on the content of the comment, not the posters aura/name/backing.
BTW, everyone keeps assuming that reputation is all about karma. It's not. We cover that in chapter 1 (and over and over again in the rest...)
Cheers,
Randy Farmer, co-author Building Web Reputation Systems
Also, as much as i hate blippy and swipely - if companies know who actually purchased a good - then their feedback becomes hugely valuable.