Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A regression is the kernel not giving the same result with the same user space

lkml.iu.edu

Post reply on HN
Print view View original

251–260 of 408 posts

Re: A regression is the kernel not giving the same result with the same user space

#251

Post by jordigh »

toomuchtodo wrote: post #233
tertius wrote: post #221

Earlier quoted context omitted.

> If someone prefers softer conduct The big issue I see people reacting to is profanity. You can be as hard on someone, given the nature of the infraction, without using profane language or denigrating someone. If they don't comply, threaten exactly what Linus threatened, no more commits will be accepted. While I don't disagree with Linus' right to speak as he does. He can be just as effective without the type of lan…

I don't see the problem with profanity, different strokes for different folks.

Profanity by itself isn't the problem. I don't have a problem with shit piss cunt fuck either. It's telling a contributor that their contributions are worthless, cannot be fixed, and other ad-hominem angry insults that just fosters a macho, jockish atmosphere in Linux that has demonstrably alienated capable contributors. I'm sorry that Sarah Sharp is no longer maintaining the USB modules because of Linus's attitude, because I've never seen anyone else who knows as much about USB as she does.

Re: A regression is the kernel not giving the same result with the same user space

#252

Post by sevagh »

Cshelton wrote: post #125
lmkg wrote: post #114

Earlier quoted context omitted.

Rust also uses a sizable fraction of all publicly-available Rust code as a regression suite for compiler updates. This allows them to check edge cases that are "breaking in theory" to find out if there's actually any code that would be broken, or if the edge case had literally not yet been encountered.

By sizable fraction, he is referring to Cargo.io, Rust's public package repository. They literally run all tests against every crate published to the package ecosystem. That is amazing.

Crates.io

Re: A regression is the kernel not giving the same result with the same user space

#253

Post by Winterflow3r »

andyjohnson0 wrote: post #109
dingo_bat wrote: post #81

I'm always eager to read Linus's rants. I wish more of my coworkers were like him. I hate the environment where nobody reacts strongly to anything! If somebody fucks up hard and refuses to own up, I (or whoever's in charge) should be able to shout at them. It's only natural.

I disagree. To me he usually comes across as an asshole, albeit an extremely smart and accomplished one. A working environment where the person at the top can be relied upon to react to human failings with an aggressive, sweary [1] rant is not somewhere I ever want to work. [1] I'm aware that use of profanity in everyday speech is common in Finland. I don't mind swearing per se, but I find its over-use a bit tedious.

Anecdata: I grew up in Finland. It's common if you're 15-16 and in middle school, but I've rarely heard adults swear in the workplace or in public (unless inebriated).

If Linus wants to use perkele management, then I guess it's his project. Doesn't mean it's the right way to do things.

Re: A regression is the kernel not giving the same result with the same user space

#254

Post by TallGuyShort »

res0nat0r wrote: post #215
TallGuyShort wrote: post #192

Earlier quoted context omitted.

He has previously been unnecessarily insulting, which I don't excuse, but in this case someone caused a problem, and after 3 weeks still does not understand how serious a problem it is. He's addressing it bluntly because that is what is required. He's not being unreasonable.

If it was anyone else we'd be calling them an utter asshole and then citing them as reasons why Silicon Valley has a reputation for . Linus is still a dick and it isn't funny, refreshing or cool.

I've called Linus an utter asshole many times before, I have no problem calling him an asshole when he is one. I've made big career changes because I was sick of certain toxic people in open source projects. I'm not defending Assholes In OpenSource(TM). I just don't think this is an instance. He isn't personally insulting the person, the person was defending a choice that anyone exposed to kernel development should know does not meet fundamental expectations. After 3 weeks, that requires some bluntness.

There's a thicker-than-fine line between being hostile to new contributors and being unkind, and tolerating bad practices from people who ought to know better.

Re: A regression is the kernel not giving the same result with the same user space

#255

Post by yeukhon »

I was recludant to agree woth Linus at first. Why should we avoid break hard? I understand we wany to avoid breaking user interface as much as possible - who would want to use an unstable API? But why is it better off to let things fail silently? For example he cited certain obsolete fields in /proc/stats return zeros.

Is it because Kernel is used by so many people and projects we can’t afford having some broken after upgrading to a new kernel. Replace Linux kernel with OSX or Android etc

Yes of course. But I feel the consequence of never break your interface and never blame userland is to force the users to write tests that can report changed behavior.

Re: A regression is the kernel not giving the same result with the same user space

#256

Post by WhitneyLand »

vorpalhex wrote: post #225
13of40 wrote: post #214

Earlier quoted context omitted.

I think a good leader could get the point across without resorting to bullying.

I wouldn't call it bullying. He isn't attacking the committer personally or due to anything outside the issue. He isn't hounding the committer. Bluntly telling someone they really screwed up and need to stop denying it isn't bullying. Someone being mean isn't bullying.

Dressing someone down publicly, in a tone based on anger and frustration, is a form of humiliation, and humiliation is a form of bullying.

Re: A regression is the kernel not giving the same result with the same user space

#257

Post by dmix »

vorpalhex wrote: post #225
13of40 wrote: post #214

Earlier quoted context omitted.

I think a good leader could get the point across without resorting to bullying.

I wouldn't call it bullying. He isn't attacking the committer personally or due to anything outside the issue. He isn't hounding the committer. Bluntly telling someone they really screwed up and need to stop denying it isn't bullying. Someone being mean isn't bullying.

Agreed, and if this a fundamental concept in kernel development, as he says it is, I believe he has every right to feel frustrated as the debate thread has been going on for three weeks.

Linus has been known to have some hostility to 'security developers' in the past, so I'm also going in with some skepticism. But I'm not seeing him be unreasonable here.

These are human processes, a lot of personality, dedication, and sacrifice is involved in these projects.

Re: A regression is the kernel not giving the same result with the same user space

#259

Post by musage »

opmac wrote: post #217
megaman22 wrote: post #164

Linus is such a breath of fresh air. I'll take an honest, emotional, impossible to misconstrue, ass-chewing over the alternatives. Too often there's an avoidance of straightforward, blunt speech that conveys exactly what is meant. I really have better things to do than try to parse nonspecific beating-around-the-bush and deconstruct word-castles.

"At least he tells it like it is" right? Forget that he's an asshole about it. Pretty much the same reason everybody voted for Trump.

> Pretty much the same reason everybody voted for Trump.

"Everybody"? Sanders also said a bunch of things like they are, and he said them politely. Especially factoring in grass roots support, he was more popular than Hillary and would have wiped the floor with Trump.

If you allow your media and "leaders" to sweep people like Sanders under the rug, you make people like Trump inevitable. And then you still don't learn. There can be worse things than Trump, so much infinitely worse; and they will come to pass if you insist on this trajectory.

Re: A regression is the kernel not giving the same result with the same user space

#260

Post by WhitneyLand »

adamnemecek wrote: post #223
13of40 wrote: post #214

Earlier quoted context omitted.

I think a good leader could get the point across without resorting to bullying.

The guy does seem like he was shifting blame around.

I think most people agree with you, he screwed up more with his handling of the situation. Occasional good faith mistakes shouldn’t get people in trouble, but he was weaseling.

The interesting issue is more of leadership.

1) Is it possible to get equally direct and effective results in a different style?

2) Does Linus care about the answer to question (1), or would he act out of emotion even if he knew knew there was a equal or more productive way?

Post reply on HN