The DIGG “Friends” System: Does It Work?
Time and again, we seen blog posts alleging that DIGG’s algorithm is bent against catching the “gamers” or those who try to manipulate the system by DIGGing (or voting) stories en masse. Here comes the question, should DIGG count DIGGs for a friend’s story less than DIGGs by strangers? Or alternatively, if a group is always DIGGing the same stories, should their votes count less?
I would say that this goes against the very concept of having “friends” on DIGG. I would say this even goes against the concept of social news itself. Social news, after all, relies on people deciding what they think is important enough to merit being a headline, or on front-page. And social news is also about communities and friends influencing other friends—whether explicitly or not—on what they think is important.
So does the inherent “social”-ness of a system also mean that it is easy to corrupt? I’ve always thought that one of the problems of social media is that it’s social. So whatever problems society has would also reflect in social media, such as social news sites. In real life, friends have preferences and collude (a.k.a. “cooperate” or “conspire”) to achieve whatever goals they have. Same with social news. Friends can work together to arrive at their own headlines.
Of course, no one except the powers that be knows exactly how the DIGG algorithm works, and how it treats votes by friends, or even non-friends who constantly DIGG the same items. For one, even if people aren’t explicitly connected as friends, they can have similar interests and inevitably get to vote on the same things one way or another.
My concern here would be if DIGG is indeed actively downgrading the value of DIGGs by friends or non-friends who vote on the same stories. Same goes wtih other social news sites. Granted, there’s always the issue of collusion or vote-buying like this hilarious antic a Wired editor recently did. But if such controls are in place, then that dilutes the social value of such a supposedly social web application.
What do you think?