3 ms·
> The paper assumes that popular repositories are popular on their own merit. I read the paper and don't see this assumption. The authors seem well aware of ma
by mlinksva 10y ago
> The paper assumes that popular repositories are popular on their own merit.
I read the paper and don't see this assumption. The authors seem well aware of marketing, calling the fastest growth cluster "viral", and conducting a tiny survey of developers finding ... that the developers posted to HN and social media.
Added: I agree only looking at most popular repos is a big problem. The authors (inadequately in my view) acknowledge this in treats to validity.
- biot 10y ago> The authors seem well aware of marketing, calling the > fastest growth cluster "viral" Viral is a description of rate of growth. It makes no statement about the reason for the growth, so much of the viral growth may have been due to people searching GitHub for keywords that would describe a solution to the problem they're having. I know that's what I do.