3 ms·
People just don't work that way; at least, I don't. Who reads code without thinking of the author? When I'm derailed by a bug in some library I'm depending on
by jtheory 14y ago
People just don't work that way; at least, I don't.
Who reads code without thinking of the author? When I'm derailed by a bug in some library I'm depending on, or see code that works in a way I find unintuitive, clunky, etc., I absolutely think judgmentally about the person who wrote it.
When I'm reading someone else's code and find a solution that's significantly more elegant than I was imagining, I'm impressed by the person's accomplishment, and they basically get points in my head. If I meet them in person I'll quite possibly shake their hand a bit more warmly.
This is normal and natural, and I have to consciously counter-balance that with the learned (but not instinctive...) knowledge that I've written bad code myself, released buggy or clunky code, taken questionable shortcuts, etc..
When I'm in a position to criticize someone else's code (and it's part of my job to do it sometimes!) I usually rewrite my feedback a few times (and sometimes decide it's not needed at all) before writing or speaking to the developer, because it takes a lot of conscious thought to modify my instinctive response into something more fair and effective based on what I've learned over the years.
That's my experience.
Does everyone else actually think that differently?