I came across this interesting op-ed entitled Identifying Social News Clustering which talks about the inevitable phenomenon of user-clustering on Digg. It turns out that heavy users of Digg have been essentially gaming the system to achieve top Digg user ranking and influence over top stories by clustering together. Although in a social network, clustering is a natural result, in some ways, it is similar to click fraud by undermining the integrity of the system's metrics.
As a result, Digg has changed its story promotion algorithm to make it more democratic, rewarding those who digg stories submitted by a diverse amount of people, rather than the same few. In this way, it is trying to prevent a phenomenon similar to the Pareto principle where 80% of articles would be digged by 20% of users (these are not the real numbers). There has been some backlash from top users to be sure, but in the long run, this looks like a good move. Only time will tell...
the link to Adotas
No comments:
Post a Comment