The problem isn’t that stars are imperfect, but they have become gamified. The emphasis on GitHub stars as a form of authority makes me trust this project even less. You’d expect VCs to do due diligence on the authenticity and quality of GitHub reviews, but I’m dubious about that, whether they are doing so out of willful ignorance or not is a separate question.
All of the contributors on this project are from the country that feeds to Infosys. I’ve seen enough inflated resumes from candidates who couldn’t write a loop despite having a masters degree. Ironically, the one candidate from this country I gave raving recommendations for was the one who had bad school grades. This country is also a hotbed for clickfarms and selling fake social media engagement/reviews.
I agree on the gamification, but I think they are still useful.
I mean this project has 18k stars that just tells me it’s popular. It’s possible they bought fake stars but at 18k that would be strange.
I think stars are useless for comparing a 17k project to an 18k project and anyone comparing that way would be flagged by me as not so smart.
But it is useful to distinguish something with just a few stars from thousands.
The challenge is that there’s not really a good way to tell what a successful project is short of looking at it closely and using it. I’d like a better measure, but this is what we have.
So it was like a developer saying “I know stars are a dumb and imperfect way to judge a project, but here’s what it is.”
If they hadn’t written this I would think they put too much weight on what’s really not a very important measure.