Hacker News Reader

This feels slightly malicious, but I can’t help but admire the curiosity that takes someone to actually see what happens if. That said, now we know, so nobody else needs to bother GitHub engineers by doing this again, hopefully.

View on HN

Not like it would take a lot of code to check on push if the repository has more than 10k commits per day since its creation date or something, to stop such abuse. Doesn't thwart existing repositories with millions of commits (Linux is at ~2M) and gives time to formulate a long-term plan for what's allowed and what's paid or just disallowed.

So even if people were to try, I don't see that being a big bother. Not that it's not malicious to do this now

I don't think you can limit pure commit counts though, because you can push many commits/massive history changes in one go.

Monorepo's in particular could be impacted

Good point, perhaps (age_in_years+1)×1M would be a better limit. Anyone wanting to import more than 1M commits could get a paid tier or beg support. At any rate, not that hard to implement is what I would expect

Git commit timestamps are 100% fudgeable. You could implement this based on GitHub repository age, but the assumptions would break for imported repos.

(Understanding we’re waaaay off in edge case territory here and this is all basically academic.)

Reply on news.ycombinator.com