I'd like to believe that one of the reasons Linus has that kind of reaction is because of the scope and impact his decisions and this kind of patches can have on people all around the world. This is a very specific situation, in which the advices in Sandya's article may not apply.

However, I fail to see how Sandya's observations are BS. A code review is a place to improve code and oneself, and there's a pretty large distinction between constructive criticism and sarcastic/aggressive comments. Being respectful/mindful of your colleague in that situation is in my opinion the minimal requirement.

In many cases it's not a patch that is a problem, it's the submitter and their approach/mindlessness/laziness/ethics that led to the patch being of an unacceptable quality. If you are trying to fix a watch and someone comes and suggest to hit it with a hammer, there's really not much place for being "respectful" with an answer.

> If you are trying to fix a watch and someone comes and suggest to hit it with a hammer, there's really not much place for being "respectful" with an answer.

That's actually the opposite of what I see in Linus. He seems to reserve these responses for people who should have known better.

So if I suggest fixing the watch by smashing it with a hammer, it probably wouldn't even merit a sideways glance. I'd be ignored like background noise. (or at most, invited to look for something more in tune with my skillset, perhaps hanging wallclocks).

When Intel show up with a hammer, the hammer is only half of the offence.

People are quick to downvote calling such things BS, because it sounds just nice and well-intentioned when presented like that. The problem is that such "rules" can be and are used to harass and shame people who are focused on the code and just want to get things done, without either sugarcoating their thoughts or being intentional assholes about it. It also just plain ignores people who actually prefer getting plain, straight to-the-point reviews, and makes their preferences less valid.

In an ideal case, the reviewer "should" perhaps consult with the reviewee's therapist and craft the reviews to most benefit the reviewee based on their emotional state - but we don't have an ideal world of infinite resources, and some of these rules go beyond a reasonable "don't be an asshole" line and make code review into an even more mentally taxing task: one where you can't treat the other person as an adult peer, instead having to talk to them as if they were insecure teenagers who could misinterpret anything as rudeness or attacks on them.

Actually, if what you are looking for is "to the point" reviews, then you should agree with the article's advices. To name a few:

- Pointing to a common reference instead of simply stating your opinion in a review will get the improvement you want done faster

- Writing a single comment for a repeated error is faster to do for the reviewer

- Explicit communication instead of emojis and sarcastic comments will bypass the need for further explanation, which will make the whole process faster

Finally, you seem to take issue with only a few of the author's points ('some of these rules go beyond a reasonable "don't be an asshole" line'), could you explain which ones and why? I am interested to know what exactly makes the process too taxing for you.

Don’t be silly. Those are just the baseline of efficient professionalism. Harnessing anger is out of scope of those, just like knowing mouth-to-mouth, how to use the fire escapes, knowing what your stock options are worth, when to blow a whistle to corruption, when to break the speed limit to save a life, or whatever the hell other important facets of life exist.

Linus Torvalds stands guard over the the most import bit of human infrastructure ever in the history of the world. I mean that literally. There is no one thing that is so universally important.

If I were Lord God of the cosmos, I'd give Linus Torvalds my magic and go farm Yabbies and sleep like a retired God at night, knowing that the world will be even better in the morning.

Sorry, but being "nice" is totally inappropriate and unhelpful.

We all get X +/- N points to assign to various traits and skills. Linus maxed the intelligence and technical ability sliders and somewhat ignored the charisma and tact ones.

I agree with you, and though I haven't evaluated your first statement, the fact that I have to think about whether or not you are right at all is good enough.

We don't have to like Linus' communication... methods. I often don't. But no one is perfect, and I'm willing to cut Torvalds some slack for being a dick for past - and future - accomplishments. People think that greatness doesn't excuse demeanor... but it does when the greatness is that far along the SD curve.

> We all get X +/- N points to assign to various traits and skills. Linus maxed the intelligence and technical ability sliders and somewhat ignored the charisma and tact ones.

That's an unhelpful model. It implies that hitting the gym or studying or putting on make-up will make harm you in equal measures as it helps you.

>> It implies that hitting the gym or studying or putting on make-up will make harm you in equal measures as it helps you.

No, it doesn't, not for anyone who played D&D which is where it is derived from.

Also, your statement is at least partially true, given that you have finite time and resources to pursue skill development in any given area.

And some of the biggest assholes i have met have come in the gym and make-up form...