Cites are supposed to be more analogous to how many times a programmers' library has been used. Unless gamed, it _is_ a useful measure. But if your programmin

Cites are supposed to be more analogous to how many times a programmers' library has been used. Unless gamed, it _is_ a useful measure. But if your programming ability was measured by how much are your libraries used (eg. in hiring, determining salary, seniority...), there would be every incentive to aim for your library to be used as much as possible even when it is superflouous. That is what has happened in academia, basically.