[repsilat]

If most papers have dozens of citations, shouldn't each paper on average be cited dozens of times? Obviously with massive skew.