I think we do need to have empathy in our dealings with people online, and in general it's in our own best interests to do so. Many open source projects' lifeblood are their communities, and other things being equal, you'll get more contributions if you're not a complete jerk.
The flip-side is that high quality maintainable code is the product of top-notch commits, and rejecting commits is sometimes necessary to keep the standard of quality high. A good maintainer shouldn't cave to pressure of accepting a flawed commit just to avoid hurting someone's feelings.
This article in fact had what looks like a prime example of that. The comment mentioning a PR might "break a limit" but "we'll cross that bridge when we get to it" was touted as an example of how to give guidance. I'd argue that code quality slipped right there as a direct result of social pressure to accept a subpar commit.
It's not easy by any measure, but I think it pays to be not only clever and kind, but also consistent and firm when it comes to reviewing people's work.
> This article in fact had what looks like a prime example of that. The comment mentioning a PR might "break a limit" but "we'll cross that bridge when we get to it" was touted as an example of how to give guidance. I'd argue that code quality slipped right there as a direct result of social pressure to accept a subpar commit.
In the absence of the broader context, I would tend to guess that that comment was trying to avoid overengineering. We've certainly had more than enough discussions on HN about "move fast and break things". This is the kind of review comment that goes along with doing that intentionally rather than accidentally.
That technical decision might be right, or might be wrong, but either way it can be presented with empathy; empathy doesn't need to change the message. You can say "I think this might cause problem X, but that won't be an issue for a while, so let's deal with it later", or you can say "I think this might cause problem X, you need to fix that before this can go in", and either way you can be professional and kind while getting your point across.
The flip-side is that high quality maintainable code is the product of top-notch commits, and rejecting commits is sometimes necessary to keep the standard of quality high. A good maintainer shouldn't cave to pressure of accepting a flawed commit just to avoid hurting someone's feelings.
This article in fact had what looks like a prime example of that. The comment mentioning a PR might "break a limit" but "we'll cross that bridge when we get to it" was touted as an example of how to give guidance. I'd argue that code quality slipped right there as a direct result of social pressure to accept a subpar commit.
It's not easy by any measure, but I think it pays to be not only clever and kind, but also consistent and firm when it comes to reviewing people's work.