Figuring out how to contribute to open source | Hacker News Reader

Figuring out how to contribute to open source

jvns.ca

62 pointseaguyhn10 comments

The only part I disagree with is "close the PR if nobody replies". Open PRs without a clear blocker are important indicators of project health. If a PR stays open for a long time, it might mean that the maintainers have trouble reviewing outside contributions, or aren't that interested in incorporating them. If I see that in a project repo, I think twice about doing the work to contribute when it might end up ignored.

Had a quick look at the article and this immediately caught my eye. Leave the PR open! The maintainer might need a few more days to respond or might eventually reconsider. Also do not forget about other developers interested in the project. I've more than once merged PRs locally to profit from them before they were merged. Last but not least, open PRs are a good health indicator, I have to agree there.

What about stepping up to help? Many projects suffer from maintainer overload and it's hugely duplicative to start another project which is likely to repeat the cycle.

Good post. For me what helped most is definitely to get into the habit of reading more code.

It is not something I did all the time previously. But coming in with a goal and then just reading code tends to help a lot, whereas I normally lean more on my coworkers when getting familiar with a new codebase (where do we do X?) simply getting into the habit of reading more code helped.

I think an important realization is that there is (hardly ever) a deadline. You can take your time to get as comfortable as you wish with the code before you start.

As a PS: I have seen this blog a few times, I gotta say I love the orange bars around the post.

> But a lot of open source projects have a mailing list / Slack / gitter / IRC channel for discussion. I haven’t really figured this out yet because the social norms are kind of unclear to me (there are often hundreds or thousands of people in the Kubernetes Slack channels and I don’t know almost any of them), but it seems like something I should figure out.

Glad to know I'm not the only person who is intimidated by talking in a large open source chat room.

In my experience, usually the worst that will happen is that you'll get ignored. In that case, you can try making your question simpler or just go back and try and figure it out on your own (possibly coming back with another question later on). Usually, you're not really any worse off than if you don't ask at all.

I think people worry that they'll hit one of the rare but bad outlier responses. A good sign of project health are things like codes of conduct and other social cues that this isn't tolerated in that community.

I think the thing that really intimidates me is not seeing other people asking beginner level questions. If I see a channel for newbie questions or I see that people are asking newbie questions I feel welcome. If I can't easily see something like that I feel like I'm intruding in a chat for "real" contributors.

Maybe it's because I'm familiar with IRC culture, this never was an issue for me, but when you are writing to an IRC channel you are not bothering anyone. IRC clients don't usually have the habit of notifying you when someone writes in a channel. So when one checks irc, one makes the conscious decision to wanting to talk to someone or see what's going on. Usually not much, so any constructive contribution is welcome.

When there is an ongoing discussion, it's of course better to wait for it to come to a conclusion before introducing a new idea, but that's not unique to irc.

"code-reading is a muscle that maybe I don’t always exercise as much as I could and so this was a good reminder of how far I can get without asking any questions at all"

Always read the code before asking is good advice for people starting out. Because when you are finished you'll have other questions.

Reply on news.ycombinator.com