News for hackers. Stuff that matters.
Re:We do not break userspace (2012) by overgard on Friday July 18, 2025 @10:34PM (#44612061)
When I first read this, back in 2012, I thought Linus' was kind of a dick for being so insulting about it. With more experience I have more empathy for Linus' though -- sometimes developers will get hell bent on getting their PR through no matter how bad it is, and you almost have to be mean in order to get through to them. If you just pick at specific issues instead of saying "This is all bad, start again", sometimes you just waste a lot of time.

Also, to be fair I don't think I would know about this "We Do Not Break Userspace" rule if this hadn't been so famous at the time.

Re:We do not break userspace (2012) by wahern on Friday July 18, 2025 @10:43PM (#44612102)
"Running a successful open source project is just Good Will Hunting in reverse, where you start out as a respected genius and end up being a janitor who gets into fights."

Credit: https://web.archive.org/web/20200909035546/https://diff.subs...

Re:We do not break userspace (2012) by sedatk on Friday July 18, 2025 @10:46PM (#44612119)
404
Re:We do not break userspace (2012) by wahern on Friday July 18, 2025 @10:53PM (#44612142)
Somehow I dropped the trailing 's'. Edited.
Re:We do not break userspace (2012) by sedatk on Saturday July 19, 2025 @12:44AM (#44612663)
Thanks! :)
Re:We do not break userspace (2012) by chamomeal on Friday July 18, 2025 @11:04PM (#44612200)
That is one of the best quotes I’ve ever heard lmao
Re:We do not break userspace (2012) by andrewflnr on Friday July 18, 2025 @11:02PM (#44612191)
You don't need to be "mean" to make clear that a PR is fundamentally flawed. You can literally say, "the entire approach here is unacceptable, and we won't merge it or anything like it." No profanity, no personal insults, not even picking at ancillary details, just facts.
Re:We do not break userspace (2012) by linusforthewin on Saturday July 19, 2025 @12:07AM (#44612509)
There are different management techniques. Some focus purely on the thalamus and some engage the amygdala.

For example, maybe you work as a program coordinator in a liberal arts college, then you may target the thalamus- the center of logical reasoning. You explain to professors how the program you’re responsible for will advance some goal that seems logical though no one feels emotional about it.

Or maybe you’re a drill sergeant whose job it is to ensure that when soldiers are under duress in war, your barked commands and insults will be burned into their amygdala, their emotion center that handles fight or flight immediate responses and related fast and deep emotional memories, so that they will respond as trained instead of forgetting everything that was learned and running away, because that may not be as effective. Logic doesn’t play into decisions under duress.

Linus may have later decided that hurting others was not as helpful, but he wasn’t being a bad leader when he did and said what he felt he needed to at the time.

Sometimes it takes a little tough love to make things happen. Even Jesus had to overturn some tables.

Re:We do not break userspace (2012) by overgard on Friday July 18, 2025 @11:10PM (#44612232)
I think there's a lot of people that would find saying "the entire approach here is unacceptable" to be "mean" though. But I generally agree with your point of not making it a personal insult.
Re:We do not break userspace (2012) by anal_reactor on Saturday July 19, 2025 @12:28AM (#44612598)
I have learned that number one rule of working in a corporation is "do not offend people". You can explain away a major delay. You can explain away an incident. You can explain away a project not being delivered at all. You can explain away all and any failure, but if you offend someone, you have a big problem. People's egos are extremely fragile, and it makes zero sense to risk even tiniest chance of conflict, unless your very job is at stake.

Seeing someone value a serious project higher than random guy's personal feelings is extremely refreshing.

> But you don't need to be mean

You don't need to be nice either

Re:We do not break userspace (2012) by andrewflnr on Saturday July 19, 2025 @01:16AM (#44612775)
> You don't need to be nice either.

...right, mostly. Being nice is nice, but the truth is more important. A corporation that could figure that out would run far better.

Re:We do not break userspace (2012) by _nalply on Saturday July 19, 2025 @07:44AM (#44614682)
Maybe. Maybe not. Given the human material (yes I know a dehumanizing formulation, but that's what Corporate does) it seems it's better not to step on anybody's toes except if it's abundantly clear that the person doesn't have a voice in things, but even then it's risky. Corporate is self-selecting: companies that don't generate profit die off sooner or later. Therefore we can conclude (if a bit shakily) from the norm in Corporate not to insult anybody that internal strife has been very damaging and therefore that Corporate self-selects in an evolutionary manner to avoid anything that leads to internal strife.
Re:We do not break userspace (2012) by andrewflnr on Saturday July 19, 2025 @09:48AM (#44615460)
That's giving evolution far too much credit. The same logic would tell you the appendix has some secret critical function.
Re:We do not break userspace (2012) by _alternator_ on Saturday July 19, 2025 @12:53PM (#44617130)
It does. The appendix is probably a reservoir for repopulating your microbiome with a “backup copy” after particularly nasty stomach bugs.

https://pmc.ncbi.nlm.nih.gov/articles/PMC12128343/

Re:We do not break userspace (2012) by andrewflnr on Saturday July 19, 2025 @11:06PM (#44621635)
I know. I said critical function for a reason. If you can remove it, it's not critical.

The point is, "but evolution" is a weak argument to justify any particular feature of an entity facing selection pressure. The laryngeal nerve detour would have been a better example but I wanted to stay accessible.

Re:We do not break userspace (2012) by anal_reactor on Saturday July 19, 2025 @06:16PM (#44619940)
I wouldn't say it has a good function on its own, but rather it's a mitigation strategy. People's fragile egos are a fact and this is not going to change anytime soon. So if you want to have a big organization, eventually you'll need to accommodate those who can't tolerate anything even remotely offending. Only organizations that aren't focused on growth can afford having a personality.
Re:We do not break userspace (2012) by andrewflnr on Saturday July 19, 2025 @11:34PM (#44621758)
> eventually you'll need to accommodate those who can't tolerate anything even remotely offending.

Nah, I don't want these people in my company. Speaking specifically, to be clear, of people who can't tolerate polite, precise correction. If I can't find enough reasonable people to grow, then that's fine. But honestly I think there are fewer of those people than you might think, given the right environment where (among other psychological safety factors) being proven wrong is fine and normal and genuinely does not require a counter attack.

Re:We do not break userspace (2012) by tempnew on Saturday July 19, 2025 @10:16AM (#44615653)
Should isn’t must.
Re:We do not break userspace (2012) by ants_everywhere on Friday July 18, 2025 @11:11PM (#44612238)
Sometimes being a parent is like this too.

But I think the "don't break user space" rule would be more effective if it was very clearly stated and you could point to a doc that made it obvious to the committer that it was against policy.

Linus seems to have a strong internal sense of what the policy means, and he heaps on Monty Python style verbal abuse to get his point across. But while that's good at indicating how angry it is, it's not clear to me that it's the most effective way of reducing violations of the rule.

Re:We do not break userspace (2012) by mycall on Friday July 18, 2025 @11:26PM (#44612328)
The main problem with "we don't break userspace" is that all general statements are false at some point. As what point is some specific instance considered by everyone incorrect that it doesn't get fixed? This is a strawman arg since I know it is simple enough to create new interfaces to bypass old problematic approaches and Linux is full of evolution.
Re:We do not break userspace (2012) by ksec on Friday July 18, 2025 @10:52PM (#44612136)
>With more experience I have more empathy for Linus' though

I am glad the pendulum is ( slowly ) swinging back.