As always, check out www.dilbert.com for the original and all the other mash ups – now on to my comment on voting.

I have a group of people that I know who regular read my Dilbert cartoons, and they are extremely blunt in their assessments – whether they’re great, terrible, or somewhere in between. Interestingly enough, whenever they say that I’ve done a “good or better” job lately, I tend to find I have a relatively large number of ratings on the Dilbert site – but the ratings tend to be quite low. One possibility is that everyone I know has equally bad senses of humor, but I’m going to ignore that for now to look at what could be a fundamental flaw in the voting system.

There are no financial incentives for doing a Dilbert Mash up, just like there aren’t financial incentives to participate in many other collaborative platforms (like Wikipedia, etc.). In turn, those of us crazy enough to participate regularly in such things are likely, whether we admit it or not, to be interested in building a “reputation”. In the Dilbert mash up case, such reputations come through the voting mechanism – do people find you funnier than others or not?

Now unless it’s changed recently, the only way to vote on the comics is to be a registered user on the site. I would think the only people interested in registering are the people that want to do the mash ups. In turn, if they are interested in relative reputations, they’re incentive is to vote low scores for everyone else – particularly anything that looks half decent. It would be like a class full of students, knowing they were being graded on a bell curve, voting on each other’s scores.

I’m in no way saying that most people do this – in fact, it looks like the vast majority are like me and never vote at all, which leads to a low overall vote count that is easy to sway. In turn, when you see vote tallies in the single digits, do these scores really mean anything? And if so, what? Could a low score actually indicate that a comic was funnier than one with a high score (or no score at all)?

While I’m focused on the Dilbert example here, it’s more of a broader question in terms of mass collaboration / wisdom of the crowds platforms – are the incentives of the crowd aligned with the outcome the platform is trying to achieve? In a future post I’ll tackle some examples of situations this has (or might) arise that have far greater implications than how my Dilbert cartoon was rated…