Skip to content

Comment on I algorithmically donated $5000 to Open Source

Comments

Oh what a coincidence. I wrote pypular (https://codeberg.org/ltworf/pypular) to artificially increase download counts on PyPI (talked about it at a MiniDebConf).

I was claiming that there are important incentives to play this sort of games.

Seems OP proved me right!

In your opinion, which metrics could be better than # downloads to evaluate PyPI packages at scale?

# of downloads from unique IP would be better for sure.

Also if you have a local cache and just check the version, that won't count.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.