alt Hacker NewsI have the same pattern and same observation. No one wants a PR any more as far as I can see. In some sense, this has always been the case. People are happy to do their own thing but not so keen on integration with other people's code. Now though it is on a different level.
I can see maintainers are being overwhelmed by AI driven PRs but if we filter our new features and concentrate just on bug fixes, does it really matter where a PR comes from if it fixes a bug?
I do work on a very complicated agent based simulation. The data shaping and loading is all open source python. There are dozens of long standing bugs that prevent the simulation from loading some of the data correctly. I used to send PRs but they were always ignored so I gave up. Now, when there is a new release I need to spend a day reviewing the new code to see what patches I need to re-apply.
> fixes a bug
Few changes are as cut and dried as fixing a bug with zero side effects or change in behaviour or need to consider future support or architectural plans.
As a maintainer, I'm always happy to quickly merge something that's like "I hit this corner case, here's a tightly-scoped change that catches and fixes it" but in reality not a lot of changes actually looked like that.