Blunt, hyperbolic statements like "massively overcomplicated" can feel like a shot across the bow when someone is talking about code you wrote. Kindness and gentleness go a long way in OSS.
> If it was your code they were talking about, you may feel differently.
I'm not the commenter you replied to, but no, I'm not my code, so I'm not going to get offended by someone criticising it. If someone does, it's on them.
There were a number of times in my career when someone called something I wrote overcomplicated, and I've never been even close to feeling any emotion about that fact at all. That remark always prompted me to hear carefully what they have to say about improving it. That's not a brag, just a certain level of emotional maturity.
People shouldn't be held responsible for every emotional reaction people have to their actions/words, because part of the responsibility for that rests on the person experiencing those emotions. In that case I don't think Casey said anything offensive, so if that made someone angry, they have themselves to blame for that.
I'd much prefer to-the-point feedback like that, even if it's somewhat vague. But maybe that's just my ability to swallow my pride/ego.
> Blunt, hyperbolic statements like "massively overcomplicated" can feel like a shot across the bow when someone is talking about code you wrote. Kindness and gentleness go a long way in OSS.
Casey's statements don't seem hyperbolic to me at all - they accurately reflect the performance and simplicity that the Windows Terminal developers were leaving on the table. We know this because Casey literally created refterm to prove his claims.
If anything, the developer's response in the initial GitHub issue, that optimizing the performance of a terminal requires "an entire doctoral research project in performant terminal emulation", is more hyperbolic than anything Casey has said.
If you do anything out in public, you either develop a bit thicker skin or have a bad time. Frankly, if this bothers you at all that is indicative of a very fragile ego and not anyone's problem but yours. The rest of the world doesn't exist to coddle you and this cultural shift folks sometimes try to propose into that exact behaviour is unhealthy. We're adults, not children.
At the time of this writing their repo has over 1400 open issues.
At the time of the "kerfuffle" their code couldn't output more than 10 (or was it 40?) characters per second to the screen on what is essentially a supercomputer. And they claimed it required "a doctoral research into performance" to make it better.
What other trivial improvements to their code also "require a doctoral research"?
Yes. Their code is definitely massively overcomplicated and I doubt they have any control over it any more.
That kind of response is pretty much never an outright lie and will always make you look good.
Or... you could get defensive and turn it into a months-long saga of bile, half-assed apologies, and bad PR.