Someone submits a 1-line PR changing the gender used in a code comment
PR rejected on the grounds that the change is “politically motivated”
Submitter got mad, and proposed removing the rule against “politically motivated” changes, calling it “white supremacist,” which is closed
Someone wrote a blog post about it
Here’s my analysis:
Stupid change - don’t make PRs that simply correct an irrelevant typo in a comment somewhere; some people do this to put stuff on a resume (look at how much FOSS work I do!), and it just wastes everyone’s time
Stupid response - it should’ve been rejected because it’s a useless change, not because it’s “politically motivated”
Stupid proposal - do you really want to waste a bunch of time fighting over wording in a comment? Because that’s the kind of crap you get without a rule like this.
This is all about an irrelevant change to a comment? Why is this getting so much attention?
I should be an idiot. I dont see a direct relationship between race and sexual orientation. Even if the PR was rejected because a pronounce how the hell this is white supremacist?
Fixing that makes sense because it’s wrong and misleading (it’s actually Manhattan distance), and a quick glace is insufficient to tell the difference.
But fixing a typo or something that wouldn’t be confusing is just noise and should only be fixed with other changes. For example, I intentionally misspelled Pythagorean in my comment above, fixing that to be the right spelling would be a useless change, even if the distance formula used the hypotenuse. It wouldn’t be an unreasonable policy to reject PRs that only fix spelling or similar to reduce noise for the maintainers.
Yep, I understand but disagree. Maybe it comes from working with so many ESL coders, but I’ll happily accept typo corrections because it’s not always obvious what words should be if you’re not steeped in the culture.
If you’re a larger project, you can see a ton of these from people hoping to land a commit to put “contributor to X” on a resume somewhere. Those add up and are really distracting and possibly automated. They waste everyone’s time, especially if they spawn a bunch of conversion like this did.
If you’re a smaller project, it doesn’t matter as much. I work with ESL coders too, so I get it (1/4 of my office is ESL immigrants, and ~2/3 of the broader team is ESL). I fix comments all the time, I just include them with other changes.
So it depends. But in general, a high profile project should reject this noise to discourage this behavior.
Did he? I only saw him point to the rule against politics.
He should have said it’s because the PR isn’t worth the time, but it also seems motivated by something that’s against the rules (i.e. why make a PR that only fixes gender in one comment? There was a later PR that was accepted that fixed it in several places).
So without more evidence, I cannot say what the dev’s motivations for rejecting the PR were, aside from the apparent rule breakage mentioned. They didn’t say they disagreed with the change (i.e. that the change was wrong), just the proposal of the change (i.e. seems more motivated by virtue signaling instead of improving the dev experience). And you can look at the comments and see justification for that position, since it quickly devolved into actual politics with people accusing the dev of being a Nazi.
Maybe if you showed a pattern across more than just this incident (i.e. over months or years), but this sounds more like people being stubborn than tolerant.
I’m surprised this got any kind of attention.
Here’s the turn of events from my perspective:
Here’s my analysis:
I should be an idiot. I dont see a direct relationship between race and sexual orientation. Even if the PR was rejected because a pronounce how the hell this is white supremacist?
Well, didn’t the Nazis also discriminate against gay people?
That said, it’s a massive leap to go from “rejects 1 line PR that only changes gender in a comment” to literal Nazi…
“comments must be accurate,” is not a rule you should bend. Bending it even a little leads to last programming and shit code.
True, but that only applies if it’s misleading. For example:
// pythagoran theorem distance = abs(p2.x - p1.x) + abs(p2.y - p1.y);Fixing that makes sense because it’s wrong and misleading (it’s actually Manhattan distance), and a quick glace is insufficient to tell the difference.
But fixing a typo or something that wouldn’t be confusing is just noise and should only be fixed with other changes. For example, I intentionally misspelled Pythagorean in my comment above, fixing that to be the right spelling would be a useless change, even if the distance formula used the hypotenuse. It wouldn’t be an unreasonable policy to reject PRs that only fix spelling or similar to reduce noise for the maintainers.
Yep, I understand but disagree. Maybe it comes from working with so many ESL coders, but I’ll happily accept typo corrections because it’s not always obvious what words should be if you’re not steeped in the culture.
It really depends on the project.
If you’re a larger project, you can see a ton of these from people hoping to land a commit to put “contributor to X” on a resume somewhere. Those add up and are really distracting and possibly automated. They waste everyone’s time, especially if they spawn a bunch of conversion like this did.
If you’re a smaller project, it doesn’t matter as much. I work with ESL coders too, so I get it (1/4 of my office is ESL immigrants, and ~2/3 of the broader team is ESL). I fix comments all the time, I just include them with other changes.
So it depends. But in general, a high profile project should reject this noise to discourage this behavior.
In theory that’s fair reasoning. Unfortunately the dev made it clear that his reasoning was based on politics
Did he? I only saw him point to the rule against politics.
He should have said it’s because the PR isn’t worth the time, but it also seems motivated by something that’s against the rules (i.e. why make a PR that only fixes gender in one comment? There was a later PR that was accepted that fixed it in several places).
So without more evidence, I cannot say what the dev’s motivations for rejecting the PR were, aside from the apparent rule breakage mentioned. They didn’t say they disagreed with the change (i.e. that the change was wrong), just the proposal of the change (i.e. seems more motivated by virtue signaling instead of improving the dev experience). And you can look at the comments and see justification for that position, since it quickly devolved into actual politics with people accusing the dev of being a Nazi.
Maybe if you showed a pattern across more than just this incident (i.e. over months or years), but this sounds more like people being stubborn than tolerant.