search

isaacs / github

Just a place to track issues and feature requests that I have for github
2.21k stars 129 forks source link

"Contributing" to a repo should require more than just a PR #332

Closed stuartpb closed 5 years ago

stuartpb commented 9 years ago

See https://github.com/torvalds/linux/pull/148#issuecomment-69055722 - spammers / fraudsters can make it look like they're a "contributor" to any major project by just filing some junk pull request against it.

(This has been the first crossposted issue I've performed using https://github.com/ghes/isaacs-github-contact-xposter. Hooray!)

awilfox commented 9 years ago

I believe that a PR can be considered a contribution only after it is merged and not while it is still open. There are other reasons besides spam/fraud that PRs may be closed (perhaps the code quality just isn't there) and allowing it to be considered a contribution before being reviewed by the project's actual team members seems a bit insulting.

jruderman commented 9 years ago

You can also "contribute" to a repository by submitting a bug report – and I think that's a good thing!

Rather than putting repo owners in the position of policing what counts as a "contribution", how about renaming the column from "contributes to..." to "participates in..."?

cirosantilli commented 9 years ago

issue +1 https://github.com/isaacs/github/issues/9 and doing metrics on it is a possible solution... but in the end, merged pull requests under graphs/contributors, e.g. https://github.com/gitlabhq/gitlabhq/graphs/contributors, will always be a much stronger metric.

quasipedia commented 9 years ago

The question seems to me quite philosophical. Are bug report contributions? Are translations contributions? Are testing contributions? Are feature requests contributions? Are rejected pull requests contributions?

I think that the current model of considering participation to a discussion as "contributing" goes well with what GitHub stands for: social coding (emphasis on social, in this case, in which any interaction with others count).

I have no problem with it, and when I evaluate GitHub profiles prior to interviews, I always check what kind of contribution the candidate did to various projects, so it never misled my judgment so far.

That's not to say that I don't see the point of the OP, though. I just wanted to point out why I don't see the present model as "broken" but just as one of the possible alternatives.

TPS commented 5 years ago

@stuartpb I'm leaning toward closing this "invalid," as the premise of the issue seems wrong — a PR does need to be merged (as @awilfox already noted) before "contributor" is assigned. What do you think? Or do you mean to say something different?

awilfox commented 5 years ago

At the time this issue was created, you could become a "contributor" by simply opening a PR, even if it wasn't merged. That seems to be fixed, so rather than 'invalid', perhaps it should be considered fixed.

TPS commented 5 years ago

Ok! Just as a clarification, though, currently opening a first-time (in a repo) pull request does show First-time contributor only in that PR, but won't be converted to Contributor in whole repo until it's merged.