I hope someone can explain the sentiment on HN to me. I don't get it, why is this popular?
I want to know how many PRs a project is getting, but more than that how receptive the maintainers are. Issues don't tell the whole picture, because work gets backlogged, and you can't expect people doing this for free to have an SLA or something. but PRs.. the work is ideally at least mostly done.
There is the one project for example, very popular in the industry it's used in. There is a specific use-case that I run into repeatedly, that it fails at. The project has lots of open issues (understandably), and there are multiple PRs to address that, but the maintainers give no good reason for not accepting it. I've been using some random guy's branch (who isn't even keeping up with the latest releases and backporting) for many years now, waiting for the maintainers to either reject it or accept the PR. Lots of people upvote, comment, and beg.
I want to see how maintainers handle that. This is really bad. I'd prefer if they stopped reporting of issues instead of PRs. Issues is providing support, PRs let other people who fixed something or added a feature attempt to contribute.
You can't just "fork it", that means you have to be the maintainer now. And how will people even find your "fork" which may have fixed things? I'd like to be able to at least find open and unmerged forks with a fix in place I could apply, even if the maintainer never got around to it.
Turning PRs off is the software equivalent of hardware makers turning off support for aftermarket parts.
Honestly, if you don't like PRs, ignore them like many already do. Does it look bad when you do that? Yes. As it should! Don't hide away from your preferences, own it. Let other people get access to fixes you either have no time to get to, or unwilling to implement.
Just the discussions alone on security related issues (or PRs as in this case) is telling sometimes.
FeistySkink 50 seconds ago [-]
An another thing I hope is added is some kind of internal karma system. E.g. if a user is spamming multiple PR to multiple repos, or is otherwise being disruptive and reported, their contributions should be flagged for review, or optionally not accepted at all.
tlhunter 4 hours ago [-]
About time. It's absolutely ridiculous that this hasn't existed for the past 10 years.
wavemode 3 hours ago [-]
yeah, I thought they were going to provide some sort of rationale as to why they've never implemented this. instead this post just basically goes "yeah, you guys have been asking for this feature for 10 years, and... it's a good idea! let's do it."
drob518 3 hours ago [-]
Exactly. Yes, please.
gscho 3 hours ago [-]
Just make the repo private?
Carrok 2 hours ago [-]
"I am fine with my code being public, but I am not fine being badgered by people about changes I have no interest in." is a perfectly valid stance.
ethin 3 hours ago [-]
That doesn't work. What if your repo is a mirror of another repo?
Tldw: a popular YouTube video on “how to open a PR on GitHub” by an Indian channel (targeting Indian audiences) showed how to add their name to a PR step by step. The rest is just the scale of the Indian population in action. I hope the maintainers of expressjs can rest easy
snigsnog 9 minutes ago [-]
Indians
x3n0ph3n3 2 hours ago [-]
Advise from low-quality bootcamp-like training programs that encourage open-source contribution, providing low-quality examples of such contribution, in order to improve one's resume and career chances.
deckar01 30 minutes ago [-]
I have always been an advocate of forking, despite the overhead of maintaining patches, but porting patches should be trivial to automate now. There needs to be an easy way to publish, discover, and require community patches even if they don’t have the maintainer’s blessing.
etoxin 1 hours ago [-]
I think we should still allow open contribution to OSS.
Maybe, a "Contributor Requests".
It would be a gate for new contributors. For maintainers, they would see what they have contributed to and see their new PR. It would show "open contributor requests"
Once approved, The PR will then appear under PRs.
And obviously this is opt in.
FeistySkink 49 minutes ago [-]
I like the idea. Right now it's only possible to require approval for actions for new contributors. But once they get a PR in, they're free to spam new PRs that can clog resource-expensive pipelines. Would be nice to have something like 5+ PRs merged and be a contributor for 1 month before a PR is auto created and actions are allowed to run.
aaronbrethorst 1 hours ago [-]
I've started aggressively blocking low-quality contributions that have that AI-generated je ne sais quoi.
snigsnog 8 minutes ago [-]
Jeet ne saar quoi
csmantle 2 hours ago [-]
It's a founded move. GitHub is code hosting platform, so there are both grounds and needs for read-only repos without PRs.
zekenie 2 hours ago [-]
They need to talk about how the pr itself should change. The text diff just is not the right thing to center. We should be using ai to chunk changes into reviewable bytes and to align on semantics and contracts.
Supermancho 52 minutes ago [-]
> They need to talk about how the pr itself should change.
When PRs are spammed, it's impractical to discuss each submitted change. The existence of the PRs interferes with the ability of maintainers to continue making directed changes.
> We should be using ai to chunk changes into reviewable bytes and to align on semantics and contracts.
That statement is a convoluted version of the narcissist's entitlement. ie "other people should realize my vision".
Rendered at 05:29:08 GMT+0000 (Coordinated Universal Time) with Vercel.
I want to know how many PRs a project is getting, but more than that how receptive the maintainers are. Issues don't tell the whole picture, because work gets backlogged, and you can't expect people doing this for free to have an SLA or something. but PRs.. the work is ideally at least mostly done.
There is the one project for example, very popular in the industry it's used in. There is a specific use-case that I run into repeatedly, that it fails at. The project has lots of open issues (understandably), and there are multiple PRs to address that, but the maintainers give no good reason for not accepting it. I've been using some random guy's branch (who isn't even keeping up with the latest releases and backporting) for many years now, waiting for the maintainers to either reject it or accept the PR. Lots of people upvote, comment, and beg.
I want to see how maintainers handle that. This is really bad. I'd prefer if they stopped reporting of issues instead of PRs. Issues is providing support, PRs let other people who fixed something or added a feature attempt to contribute.
You can't just "fork it", that means you have to be the maintainer now. And how will people even find your "fork" which may have fixed things? I'd like to be able to at least find open and unmerged forks with a fix in place I could apply, even if the maintainer never got around to it.
Turning PRs off is the software equivalent of hardware makers turning off support for aftermarket parts.
Honestly, if you don't like PRs, ignore them like many already do. Does it look bad when you do that? Yes. As it should! Don't hide away from your preferences, own it. Let other people get access to fixes you either have no time to get to, or unwilling to implement.
Just the discussions alone on security related issues (or PRs as in this case) is telling sometimes.
Maybe, a "Contributor Requests".
It would be a gate for new contributors. For maintainers, they would see what they have contributed to and see their new PR. It would show "open contributor requests"
Once approved, The PR will then appear under PRs.
And obviously this is opt in.
When PRs are spammed, it's impractical to discuss each submitted change. The existence of the PRs interferes with the ability of maintainers to continue making directed changes.
> We should be using ai to chunk changes into reviewable bytes and to align on semantics and contracts.
That statement is a convoluted version of the narcissist's entitlement. ie "other people should realize my vision".