Thick skins

That might be true. I'm certainly willing to argue my point, loudly (which is a behavior that I've seen LKML encourage in the past -- "we want to make sure you really believe that the thing you're proposing is correct, so we'll force you to argue it"). In any case, I'm reasonably unlikely to be personally hurt or offended, and there should be somewhere in this skin analogy for the people who get hurt and give up.

Then again, I _have_ given up on LKML, so.

Yes, there's certainly a difference there.

I've been told, and told people, "This approach is completely wrong" couched inside "I value and appreciate the time you spent on it, and don't think that you are dumb." I've been in productive communities where you don't need a thick skin, because people value each other and their happiness in that community, and out of that, productivity and good technical decisions naturally happen.

When I'm lucky, instead of directly being told "This approach is completely wrong" (or telling people that, conversely), I'm simply asked enough questions to realize that for myself, and nobody is hurt. This approach requires patience and trust, and a general feeling that others in the project value you as well as the project. So it's understandable that in volunteer communities, people don't tend to go for that approach. But it's worth the long-term investment, where possible, and we should acknowledge that we're hurting people when we don't.

Seconded; you can tell someone in no uncertain terms that they've done something wrong and should make sure it never happens again, and even make it clear that they should have known better, without going on a tirade against them personally. *That's* what it means to keep things professional.

I also find Linus's behavior rarely if ever abusive, also because usually he's quite polite towards people that are trusted. (Frankly not always, but in general that is true).

But I was just now skimming the "recent patches" on LWN and found this:

> So you try to sell that qrwlock as a replacement for ticket spinlocks,
> while at the same time you omit the fact that we have an even better
> implementation (except for the last test case) already in the
> kernel. What's the point of this exercise?
>
> ...
>
> We really need proper explanations and not some guess based drivel to
> assess that.

The rest of the message contains many reasonable and insightful observations, but is it really necessary to talk about "guess based drivel" and accuse the poster of faking/withholding experimental data?

I call *this* lack of professionalism (and also the ARM example linked to in the article). I'd take a million swear words, it doesn't matter if English or Italian or Finnish, over this.

I am sorry, I go far beyond axboe, this is a storm in a thimble, Linus is often very rude and usually right.

This silly woman should STFU and stop working on the kernel and go back to sandal knitting.

I for one am sick and tired of special case identity politics and PC. from women. They want to play, then play don't start asking to change the rules