Stephen Ou, a 15 year old web developer, has identified what he claims to be “multiple” problems with Twifficiency, a program which calculates a percentage figure for Twitter users' tweet “efficiency”.
Ou notified The Drum of three specific flaws with the service after putting it through its paces for a day. This is what he found:
Reacting to the “spamming” reports of users Ou said: “Technically, it doesn't require OAuth at all (a kind of ‘key’ that gives limited access to your Twitter account), the only reason it does is because they want users to tweet out their scores.” This Ou suspects was a conscious effort by Cunningham to guarantee the “terribly viral growth” of the service.
Another “problem” Ou identified is a deficiency in Twifficiency’s algorithm which calculates its scores based upon how many people you follow, who follows you, how often you tweet and how many tweets you read. It is this last parameter which raises a red flag in Ou’s mind: “Through Twitter API, it's (nearly) impossible to get the number of ‘how many tweets you read’ on the Twifficiency website”, he says.
This discrepancy, Ou observes, leads to: “scores which are different each time”, even amongst users with the same following, followers and status counts.