Why This Hacker Stood Up Against ‘Verbal Abuse’ in Linux Land | Hacker News Reader

Why This Hacker Stood Up Against ‘Verbal Abuse’ in Linux Land

wired.com

69 pointsnicholassmith127 comments

I understand her viewpoint...

But I read Linus e-mail explaining his attitude, and I am more inclined to agree with him instead.

I am one that sometimes people called caustic or whatnot, including in my current startup, no matter how much I made a great effort to be polite and make clear I was not attacking anyone, every time I said the artist did something wrong, he would get pissed off and offended.

After a while I gave up, and started to write in my usual "caustic looking" style, and the artist started to understand me more, make less mistakes, and now we can make jokes and work with other without issues.

I personally hate the "corporate" style culture of being passive agressive, avoiding confrontations, being polite all the times and whatnot.

If you annoys me, I will tell you that you annoy me, and if you insist, I will throw a "fuck you".

I remember how much I hated every environment that I was, where "verbal abuse" was curtailed, because I never knew if someone was saying the truth or not, and in the end I started to behave like if everyone was a serial liar, and when you cannot trust anyone, it is hard to get anything done.

As someone who (I don't think) has any form of autism, or asbergers I still find it difficult to parse the intent behind overly polite messages, or those that try and avoid confrontation. I'd very much prefer to have messages that were particularly vicious, as it'd make the intent far clearer to me.

I totally agree. I believe if I can throw a "fuck you" at you and you can take it without creating much of a fuss, it'll just be easier for me to pick out your mistakes right up without resorting to fake politeness. And you could do the same. This would only make it easier to get things done. Also who doesn't like to get things done their way, when its their own project?

Also I can't help but agree to Linus' response mail.

Why does it have to be fake politeness? Why can't we have real politeness?

Not everyone is wired the same way right? Linus is being his way. If you are a polite person, you are not expected to swear around. Be polite and that'd be wonderful. The same goes for those who just aren't polite. Why expect them to fake it? Let them be themselves.

That's an interesting question.

Maybe all politeness is fake to some degree (you're interfacing with someone by being nice regardless of what that person does), but we say "fake politeness" when you're spending a lot of energy to keep the politeness filter on.

Tell me if you have a theory about real vs fake politeness.

If I had the choice between Linus' abrasive style and the fake politeness in Wikipedialand, where you can needle and provoke your enemy at will provided you don't use rude words I pick Linus a thousand times over.

Linus knows what a colleague is, and Wikipedia doesn't - over there they have enemies, not opponents. The headline calls it "verbal abuse", but they don't seem to have a grasp what abuse is.

Sure. I'll just take a third option though where people are direct but not abusive. Doesn't seem like too much to wish for.

Presumably you've heard the phrase that "we can disagree without being disagreeable".

I agree that I would much rather work with Linus than Wikipedians, but I'd also prefer working with people who I trust to give me their honest opinion without a needless verbal shelacking.

You could also ask me if i'd prefer to have my hand or my foot cut off (foot for the record), but personally my preference is to find a way to keep all of my limbs intact.

The point over which I absolutely support Sarah Sharp is that what Linus is doing is essentially hazing. He's trying to enforce societal norms by handing out verbal beatings, and trying to encourage others to enforce his norms via his methods.

That is a problem. Tart replies regarding criticism of code, or how code is managed is one thing. Trying to encourage a culture of verbal abuse is entirely another.

------------------------------------

Sidebar: There are organizations that do work these sorts of verbal dressing downs as a systematic tactic. The Marines are famous for it. The objective is to provide shared hardship to encourage bonding amongst a group of trainees. They're also famous for fucking it up (so much so that we've got films like Full Metal Jacket to memorialize the experience).

But FOSS and Kernel contribution is not like the Marines. There is no inherent unit of social support to fall into. If you're lucky enough to be part of a group, or to have someone say "oh yeah, that's how Linus is" then maybe you get over it and inure yourself to the abuse. Or maybe you just stop contributing to the Kernel.

Or maybe you just stop contributing to the Kernel.

I'm sure there are many potential contributors who have dropped out due to discomfort with communications. (Although it seems Linus reserves the most emotional tirades for long-time contributors whose performance he feels has declined.) However, does it make any sense to argue that the kernel needs more contributors? ISTM the kernel is moving as fast as can be expected or desired: some distros lag the kernel by increasing amounts.

Swearing is not a problem. In fact, people who swear tend to be more honest than those who don't. You don't have to be polite, you don't have to avoid confrontations. What you have to be is respectful. You should never attack a person, you should attack their actions. But only after you understand.

  > In fact, people who swear tend to be more honest than those who don't

[citation needed]

There's a whole lot of context involved here. Yes, there are very honest people who curse like sailors (fishermen especially [1]), likewise there are honest people who wouldn't throw about farthing filth about fornication willy-nilly[2].

I would add that ideas, in addition to actions, are fair game.

  [1] Deadliest Catch
  [2] Ex-Mormon colleague

Your sources are a TV show and an ex-mormon?

The TV show was cheek. The ex-Mormon, I've known for 12 years.

Which is what Linus did...

Is it not possible to both be, perhaps stern and clear in view without being abrasive?

"Hey John, thanks for contributing. I see you updated X. There's a lot of good there but this algorithm really needs to run faster and use less memory. Check out how Sarah did it here when working with Y. Additionally, here's some resources. Thanks."

That implies that it's ok. I would prefer: "Hey john, thanks for the contribution though, but it's not good enough - it's currently too slow and hogs too much memory. As an example of how it should be running see sarah's Y."

That would be the case I suppose if John's entire contribution was poor. I was thinking perhaps there might be some part of John's contribution that was actually quality and to say "hey, you know you're doing good work, it's just this part that isn't working well".

Obviously we're dealing in hypotheticals here and I guess if it is the case that none of John's code works well then your approach is probably the only decent approach. If there is some code of John's that does work well than I find a lot of times it's easier to give a little compliment prior to your criticism just so they know that yes, we do appreciate your work and contributions, but in order to be more effective here is what you have to do.

Have you ever managed or lead a team?

If you have, you'll know that your job in that situation is to get the best results from the team - over a variety of time-scales. Sometimes you have to crack the whip, but mostly its important to be somewhat diplomatic and respectful while still being clear and direct, because the team is made of people who (unless it has some kind of dysfunctional brogramming-type culture) have a need for fulfilment in the work they're doing.

Some cultures are more direct (eg. Dutch, Israeli, Finnish) than others (eg. Japan). If everyone on a team expects the same norms then they are going to get along well. The problem for Linus is that the "team" is a pan-cultural community with a spread of norms and (it seems to me) that Linus want to be the Linux benevolent-dictator-for-life while everyone else expects him to be a leader.

"Hey John, thanks for contributing" - This is only true if you are actually grateful to John. Otherwise, John is wasting your time (if his patch is complete rubbish).

"There's a lot of good there but this algorithm really needs to run faster and use less memory" - Another lie (unless there actually is a lot of good there). What if there's not a lot of good? Should we just prefix our statement with this banal clause?

"Check out how Sarah did it here when working with Y. Additionally, here's some resources." So now someone who's time is worth a lot (top tier linux maintainer) is responsible to LMGTFY for everyone who contributes?

Sometimes a little politics goes a long way. You can always be grateful someone is taking their time to contribute to a project even if they're not particularly good at it yet. Version control and branching means things don't have to be implemented immediately or at all.

If there's not a lot of good, I was assuming perhaps it was one part of his contribution, but if there isn't any good at all then how about "Thanks for contributing but in order to be an effective member of the team you're really going to need to work on X, Y, and Z. Unfortunately we cannot commit your patch to the code repository until this is the case. If you do work on X, Y, and Z though you should be at a point where you can contribute to quite a wide variety of areas within this project and will be a quite a valued member! We need people like you with dedication. And don't worry about it, if you look at version 0.a.b you'll see I made a bunch of the same mistakes!"

If you have the time to dole out harsh criticism of someones patch you can't take the time to provide a helpful comment instead of an overtly negative one? "Oh I spent all this time reviewing your code and it sucks." is better to you than taking the what, ten or fifteen minutes to compose a quality reply? I understand the top tier linux manager's time is valuable. But if you can tell the code is wrong, then give reasons for why it is wrong, tell them you can not commit until those reasons are corrected, and do so in a polite manner. In addition, by providing resources, you make your own job easier as the quality of patches continues to improve in quality, you retain talent, and foster new talent within the software community.

If you don't want to provide resources then perhaps a "Hey, if you have any questions trying to work through X, Y, and Z try to contact me here although you'll get far quicker responses on irc.freenode.net #linux".

>> But if you can tell the code is wrong, then give reasons for why it is wrong, tell them you can not commit until those reasons are corrected, and do so in a polite manner. In addition, by providing resources, you make your own job easier as the quality of patches continues to improve in quality, you retain talent, and foster new talent within the software community.

One of the problems with assumptions here is that kernel maintainers by definition shall know that their code is wrong and why it is wrong . If this is not true, they shall not be kernel maintainers.

Another one is that Linus does not need to retain talent that can not perform to acceptable standards.

Please note that acceptable standards are different for different people.

No this is totally true and I agree. But if you can mold someone into performing at acceptable standards than you have more talent which produces better products faster. That's basically my reasoning for my approach.

If you do not know why the code is wrong, then tell them what it is doing wrong in a similar fashion. "Hey John, I see you updated this patch. Thanks but it seems to be hogging memory. What can you do about that?"

Perhaps you point them to people who have worked in that area before. "Well you working on I/O, you know Tim over here has worked in that area before, he might be busy but if you can't figure it out on your own maybe you can shoot him a question and see if replies."

I guess a kernel maintainer may never make a mistake since you assume it is a tautology that all kernel maintainers by definition know when and where and why their code is wrong or they could not be kernel maintainers. Or you're implying that any kernel retainer will, upon being notified, no matter the language of the rebuke, immediately know how to fix it and will also be willing to do so.

There are also different levels of experience for different people. It doesn't mean you should throw them off the ship or so harshly rebuke them that they never return to the project.

And if there are different standards than you probably don't need to give everyone the harshest standard.

Alright but I've posted enough in this thread.

Your comments seem entirely premised on the patch being "complete rubbish". If that is not true (and, there's nothing in the GP's comment to indicate otherwise), and actually "there's a lot of good there", why should the maintainer not be grateful to John?

Of course, if the patch is helpful the response should be a positive one. I'm not debating that point.

I'm debating the fake politeness that is expected when a patch is in fact rubbish. It's more impolite to lie to someone and make them feel better by phrasing such as "It's good but...". It's much better to simply cut to the "but" part and tell it how it is.

Fellow developers are not clients who are our responsibility to make feel good at the end of the day.

You need to decide whether the examples given are outright lies or are just hypothetical. You can't have both.

But I agree with your point, not everyone has the skills and mentality to be mentors and/or teachers to others who may wish to learn. It's just too bad so many have to be complete assholes about going about informing people of this fact.

Furthermore, you can't assume that because someone is unwilling to be a mentor that it's always the mentor's fault for not having enough patience or the right attitude.

Sometimes the student just isn't worth investing your time and effort into.

Totally true, sometimes someone doesn't make for a good student.

But I don't think it should be up to a bad teacher to decide who is a bad student.

When I read that it really seems one step away from:

"Hello John, we recently received your contribution. It is clear you are aligned with our goal of leveraging core-competencies of a diverse group of synergistic A-players in our problem space. Your algorithm is proactive but we need it aligned to the specific performance/memory axis to support the rest of the product synergies. Sarah is a fantastic and robust resource that is empowered to impact this area of the value-add proposition. If you need to plug into any further best of breed enablement engines, please do as we are committed to our future-shapers building their capabilities to their fullest potential."

'Moving parts in rubbing contact require lubrication to avoid excessive wear. Honorifics and formal politeness provide lubrication where people rub together. Often the very young, the untraveled, the naïve, the unsophisticated deplore these formalities as "empty," "meaningless," or "dishonest," and scorn to use them. No matter how "pure" their motives, they thereby throw sand into machinery that does not work too well at best.'

I know I've seen that quote before, but I forget where.

I personally hate the "corporate" style culture of being passive agressive, avoiding confrontations, being polite all the times and whatnot.

Agreed. If only that were what she's talking about:

  To me, being "professional" means treating each other with respect.  I
  can show emotion, express displeasure, be direct, and still show respect
  for my fellow developers.
  
  For example, I find the following statement to be both direct and
  respectful, because it's criticizing code, not the person:
  
  "This code is SHIT!  It adds new warnings and it's marked for stable
  when it's clearly *crap code* that's not a bug fix.  I'm going to revert
  this merge, and I expect a fix from you IMMEDIATELY."

https://lkml.org/lkml/2013/7/18/340

  We're adults, not high schoolers.  We can figure out how to deliver
  harsh technical criticism without resorting to name calling, cussing at
  people, or personal attacks.

https://lkml.org/lkml/2013/7/16/554

  You are in a position of power.  Stop verbally abusing your developers.

https://lkml.org/lkml/2013/7/15/427

When it comes to fuck yous, I hate them because they don't communicate anything, although combined with a foreknowledge of the person's impulse control abilities you might be able to figure out their emotional state (which I usually don't care about.) It's just abuse, and if we're going to abuse each other with no content, I might as well punch you, which is fun. Swear words are often just more shit to wade through to get to content.

Sweary people like to direct their frustration outwards, but that doesn't necessarily make them direct. I see little relationship between people who throw hissyfits and people who are honest in the way I like. The honesty I like cuts down the search space for reaching pragmatic solutions; it doesn't simply tell me that Billy's blood sugar is getting low.

If interpreted correctly Linus' remarks about the value of having a panel discussion on the issues Sarah raised, he does think it might be possible to find an accomodation on lkml that suits both him and Sarah.

Sarah was not passive aggressive, and she did the opposite of avoid confrontation on this matter. It seems odd to me that she is accused of wanting these aspects of corporate style.

There seems to be a line of thinking that one can only be "fake polite" or "aggressive", and that there is no in-between.

It is perfectly possible to be assertive without being aggressive - to be certain, and strong-sounding without belittling the person you're talking to.

It does require more effort, to many people, than simply screaming at them. But if you want to get the best results out of them, and avoid putting off people who simply don't want to be around someone who screams, then it's worth investing the time.

Exactly. In addition, I think there's an unfortunate tendency among programmers to wear the "crusty curmudgeon with borderline autism" as a kind of pride, even emulating it to be more credible.

I agree.

There are genuinely intelligent people out there who look up to Linus and people like him. And the problem is that they look up to these people and identify with their behaviour because they want to be seen as uncompromising geniuses. They identify with a contrarian attitude and view conventional social idioms (for a lack of a better word at the moment) as ineffective in technical discussions. And so telling these people that the way they are behaving is not appropriate is essentially viewed by them as a personal insult. It all ends in a very divisive, two-sided argument.

The problem with Linus isn't Linus; it's that people want to be like Linus. They want to be seen as strong, assertive, and uncompromising technical leaders. They don't want peoples' feelings to interfere with their work. They view themselves as contrarians and will seek to identify with leaders in their community which reinforce their world-view. And the problem with that is one of misplaced attribution.

It's not wrong to use wit and a little strong language in a debate. Christopher Hitchens, a famous contrarian, employed such tactics from time to time with great effect. The difference however between someone like Hitchens and someone moderating a contributions to an open source project is that Hitchens was trained in rhetoric, elocution, and debate. He knew how to use insults and jabs in appropriate situations with practiced intent. Someone who views themselves as a contrarian and identifies with someone like Hitchens or Linus is prone to believe they understand these things when in fact they do not.

Very few schools teach such subjects today and I suspect even fewer programmers have ever bothered to take them.

However articles like OP's don't serve to forward the discussion. If we attempt to tell people who consider themselves to be intelligent contrarians that they are behaving inappropriately then they will immediately appeal to their heroes' success. This only starts the discussion off as a personal attack and devolves it into a polarized argument. Hardly useful.

Instead there must be a way to convince ourselves that losing your temper on a mailing list and denigrating the efforts and skills of other contributors is not the proper way to behave. There are other, more effective ways of being assertive and uncompromising that won't insult people and cause so many needless flame wars.

The problem with Linus isn't Linus; it's that people want to be like Linus.

This nails it.

It's not that Linus or his style are right or wrong, it's that that some people assume that his abrasiveness is the cause of his genius rather than an unrelated personal trait.

I don't understand why it even requires more effort. Here's one of the emails Sharp cites:

  <mchehab@redhat.com> wrote:
  >
  > Are you saying that pulseaudio is entering on some weird loop if the
  > returned value is not -EINVAL? That seems a bug at pulseaudio.

  Mauro, SHUT THE FUCK UP!

  It's a bug alright - in the kernel. How long have you been a
  maintainer? And you *still* haven't learnt the first rule of kernel
  maintenance?

  If a change results in user programs breaking, it's a bug in the
  kernel. We never EVER blame the user programs. How hard can this be to
  understand?

  To make matters worse, commit f0ed2ce840b3 is clearly total and utter
  CRAP even if it didn't break applications. ENOENT is not a valid error
  return from an ioctl. Never has been, never will be. ENOENT means "No
  such file and directory", and is for path operations. ioctl's are done
  on files that have already been opened, there's no way in hell that
  ENOENT would ever be valid.

  > So, on a first glance, this doesn't sound like a regression,
  > but, instead, it looks tha pulseaudio/tumbleweed has some serious
  > bugs and/or regressions.

  Shut up, Mauro. And I don't _ever_ want to hear that kind of obvious
  garbage and idiocy from a kernel maintainer again. Seriously.

  I'd wait for Rafael's patch to go through you, but I have another
  error report in my mailbox of all KDE media applications being broken
  by v3.8-rc1, and I bet it's the same kernel bug. And you've shown
  yourself to not be competent in this issue, so I'll apply it directly
  and immediately myself.

  WE DO NOT BREAK USERSPACE!

  Seriously. How hard is this rule to understand? We particularly don't
  break user space with TOTAL CRAP. I'm angry, because your whole email
  was so _horribly_ wrong, and the patch that broke things was so
  obviously crap. The whole patch is incredibly broken shit. It adds an
  insane error code (ENOENT), and then because it's so insane, it adds a
  few places to fix it up ("ret == -ENOENT ? -EINVAL : ret").

  The fact that you then try to make *excuses* for breaking user space,
  and blaming some external program that *used* to work, is just
  shameful. It's not how we work.

  Fix your f*cking "compliance tool", because it is obviously broken.
  And fix your approach to kernel programming.

                 Linus

329 words. Here is the same email with the dramatic noise removed, and patched up for clarity when necessary:

  <mchehab@redhat.com> wrote:
  >
  > Are you saying that pulseaudio is entering on some weird loop if the
  > returned value is not -EINVAL? That seems a bug at pulseaudio.

  It's a bug alright - in the kernel.

  If a change results in user programs breaking, it's a bug in the
  kernel. We never EVER blame the user programs. 

  To make matters worse, commit f0ed2ce840b3 is [broken] even
  if it didn't break applications. ENOENT is not a valid error
  return from an ioctl. Never has been, never will be. ENOENT
  means "No such file and directory", and is for path
  operations. ioctl's are done on files that have already been
  opened, there's no way in hell that ENOENT would ever be valid.

  > So, on a first glance, this doesn't sound like a regression,
  > but, instead, it looks tha pulseaudio/tumbleweed has some serious
  > bugs and/or regressions.

  [This is wrong.]

  I'd wait for Rafael's patch to go through you, but I have another
  error report in my mailbox of all KDE media applications being
  broken by v3.8-rc1, and I bet it's the same kernel bug. [Your
  patch doesn't work] so I'll apply it directly and immediately
  myself.

  WE DO NOT BREAK USERSPACE!

  The whole patch is broken. It adds an incorrect error
  code (ENOENT), and then because it's wrong, it adds a few places
  to fix it up ("ret == -ENOENT ? -EINVAL : ret").

  Breaking user space, and blaming some external program that
  *used* to work, is wrong. It's not how we work.

  Fix your "compliance tool", because it is obviously broken.

                 Linus

209 words: a savings of more than 33%.

I personally find people demanding Linus change to fit them to be the rudest thing I have ever seen.

Let's look at the history quickly: Linus sits down to write an OS as a hobby. It's a huge success and he shares it for free with everyone who wants to use it. Some big corporations like Intel see how good the thing that Linus has made is, and they employ people to help him out.

They then turn around and say Linux is now a big corporate project, and Linus needs to watch his manners? Fuck them, they wouldn't have Linux without him. It's beyond rude for these people to demand Linus fit their corporate culture. I'd say they are welcome to communicate privately with Linus - with a lot of respect owed to him - and convince him to the merits of behaving differently. But to go to the press and make a public statement on how Linus needs to change himself? It's simply disgusting.

Linus is free to act however he wants, and anybody dealing with him is free to walk away. They won't though - Linus has shown incredible ability in managing contributors and also great 'taste' in keeping the kernel working smoothly. This article, however, is plain rude.

The "Linux is Linus' baby, he can do what ever he wants"-argument is getting really old. For a very long time it has been a collaboration project where people work together. When collaborating you have to have respect for your peers or, at least, act in a respectful manner. If you can't do that then you are both childish and unprofessional. Name calling and personal attacks are only bad for morale and hinders collaboration.

The proper response to old arguments that no longer convince is to fork. If all the people who claim to be tired of working with Linus put their efforts where their claims are, they would have a viable fork.

Shorter version of your argument: "How dare you tell your parents to change the way they treat you, you're alive because of them!"

Living beings have pressured each other to change for one reason or another since the beginning of time. Linus is an effective leader and an asshole. Could he be more effective if he's less of an asshole? That's an open question. It's not any more rude to suggest it than him calling people stupid, idiotic, retarded for introducing kernel regressions.

Ps. She didn't go to the press. The tech press has a tendency to come to you. Sometimes it's better to clarify than stay silent.

I'm obviously communicating my argument poorly then.

Shorter version of my argument: "How dare you tell other people's parents to change the way they treat them - publicly in an interview?" Some parents might discipline their children more harshly than others. If I went to tell the press about how a parent gave their child a smack for doing something stupid, I'd say that would be very rude to that parent. Linus isn't killing anybody here - he's shouting at people who fucked up, and telling his staff to be more rude in sending away bad patches.

If she went to the press or the press came to her feels very irrelevant. She chose to answer in an interview she knew would be very public - she started this on a very public mailing list.

I would agree if this were a company project, or private organization, but its not, its a public dispute on a public project with a vast audience, and a press hungry for political drama-fueled clicks.

I don't think she said anything in this interview that was disrespectful to Linus, or anything other than color commentary. It's sort of like the press conference after a sports game where a player disagreed with the ref during the game and it was picked up on mic, and the press asked her about it after.

Her crime is that she took a stand against an authority figure. She might be wrong in the end, and talking to the press CAN be risky, but this specific incident doesn't seem to be all that bad.

Are Intel and other corporations the ones saying he should watch his manners? It seems more that it's developers or maintainers, not corporations.

I just don't recall seeing corporations telling Linus to essentially be nicer, just developers. If you have links to the corporations doing it, please share.

I suspect most organizations are even more cutthroat than Linus: if we're paying you to contribute patches that are of strategic benefit to the company, and for whatever reason those patches are not accepted, we'll find someone else to contribute our patches.

No, you are wrong, and Linus is wrong and you are both worthless assholes who should just fuck off and die. Oh, you don't like my tone? Yeah, just like a loser to play the victim card. Learn to accept my different style of communication already!

The disgusting things here are Linus' behaviour and your rationalizing of it.

I don't personally mind your tone, but your message has no content. If you read any of the 'insulting' posts by Linus, you'll see they're insulting for a reason: someone messed up badly, and he always explains exactly where and how.

The difference is that Linus resorts to such communication for a reason versus you doing this just to prove your point.

Oh, and what is Linus' reason, if not to prove his point? He pretty much outright says so: he thinks it's necessary to swear to get people to listen.

>> he thinks it's necessary to swear to get people to listen.

That is correct, but ONLY when people (specifically kernel maintainers) produce much below acceptable standards. In case I am missing smth - can you point to an e-mail where Linus was cursing at a newcomer (i.e. not kernel maintainer)?

I was lost as to context, following the mailing list thread (which was essentially just Linus' response to something I hadn't seen.)

I was in the middle to start with, and this pushes me right onto Sarah's side. Especially this:

> "I don’t think we’re going to hate each other. I think we’re going to go to Kernel Summit and eat cookies and discuss this."

You can definitely give negative feedback without being a foulmouth, especially when you're talking to people who are at work. You can be emphatic, rigorous and uncompromising without swearing or insulting anyone.

Sarah, in fact, is giving exactly this sort of negative feedback to Linus. He just needs to understand that, and that everyone is always improving how they work, no matter how many years they've been doing it, and no matter whether they invented the field.

Sarah is essentially saying: "everybody can love each other, and when there is a mistake, we will just kiss each other harder until we get the problem solved". This, in my opinion, is much too optimistic.

Linus is saying: "there are some people I do not like, and I will never like. But I do not want that to get in the way of being productive. We can still work in the same project, maybe not side by side. And whenever you make a mistake, I will let you know without the tiniest doubt. And you should do the same, too." I think this is clever, and allows for very diverse teams, where people are able to work together even though personally do not fit at all. People are different - let's find a way to overcome that!

Basically, Sarah is worried about the social side of running the kernel - the egos of people - and Linus is (surprisingly for a manager!) worried about the technical side, sidestepping personal issues. He just does not care if you are feeling hurt. For that, you have your mum.

For him, the only worthwhile goal is to be productive, and to communicate clearly what he is thinking. Sarah expects him to understand that she is pissed by this. In this case, the whole context makes it very easy to understand it, but being polite is very often the wrong way to make things clear. It just takes much longer, and sometimes you never realize there was a problem to start with. Since she is trying to solve this issue "politely", it will just take forever, and no solution will be reached - because there is no solution! People are not going to start loving each other just because she would like to. She is trying to create a kernel lah-lah land, where everybody is happy and at peace with each other. Not gonna happen!

At the end of the day, it just a harsh email that you get. I would say, get over it!

You put quotes around your (biased and wrong[1]) summary of what Sarah said, and then quoted Linus word for word in the same style. Dude, you're presenting parody-fiction and evidence on a par. Not. Cool.

Sarah and I do not think that people have to like each other in order to act civilly towards one another; the entire weak-tie economy proves this to be false, in fact.

She is saying (and I agree) that you can be robust and challenge someone without turning off your brain-mouth filter for trash talk. I think you'll find people can be just as productive and technically awesome without all the swearing.

[1] and probably sexist, since you implied that the girl is talking about kissing. Seriously, dude? That's all female programmers can ever be talking about?

>> Dude, you're presenting parody-fiction

gonvaled shared his interpretation of the context and his opinion whilst you are engaging in the personal attack.

I just wonder what would be Sarah's reply to your style above [dude, brain-mouth filter for trash talk...].

>> Sarah and I do not think ...

Are you Sarah's colleague, by the way?

> "gonvaled shared his interpretation of the context and his opinion whilst you are engaging in the personal attack."

umm, what? I didn't use any insults, I just talked about why the user's behaviour/arguments were bad. It's criticism, yes, but it's not ad hominem, it's on the behaviour and the ideas.

> "I just wonder what would be Sarah's reply to your style above"

If two (almost certainly) male people on the internet can't address one another as dude, I don't know what the world is coming to[1]. Some people might object, yes (I remember a particularly pompous twitter politico who balked at me calling him "dude" on twitter in an argument) - but this is how I talk out loud, and it's not obscene it's just informal. I don't believe it conflicts in any way with what I've said about being politely robust and emphatic.

> "Are you Sarah's colleague, by the way?"

no, never heard of her before this incident. I find it kind of sad that you thought I might be some kind of stooge. Feminist boys do exist, you know.

[1] I do occasionally use "dude" as an exclamation when talking to girls too, but it is usually in person and quite obviously sarcastic (I am British, and I say "dude" in a vaguely Californian accent, a result of listening to too much Blink 182 as a teenager I think.)

The problematic thing about Linus's behavior is not its direct effects but the example it sets. Even if one can justify his behavior in terms of his responsibilities at the top of the technical hierarchy, that doesn't justify worse behavior by those lower down. Others see what he does and take it one step further. Then, in the absence of a correction from him, they take it another step further, until they're just as nasty as their own pathologies allow them to be, and newcomers start to imitate that right off the bat. Just one "Al, that was over the line" comment from Linus would damp that feedback loop. The continued absence of leadership on that front ends up affecting the overall tone far more than the occasional one-on-one outburst.

P.S. To cheery, since you've been hellbanned: that's a false dichotomy. I'm not suggesting that people should be artificially nice. I'm not a nice guy myself. I'm just suggesting that discussions should be kept focused and fact-based, not made personal or nasty for the sake of entertainment or sheer bloody-mindedness. That's what I see both on LKML and here, and it serves no purpose. It's not honest, it's not efficient, it's just being a dick.

This is why Linus may want to consider toning it down or fin another way to make his point:

"But 80 percent of the kernel contributors are paid by companies. So that means that the Linux kernel really has a lot of people from corporations on there. Some of them, like Linus, are paid by nonprofits. But it’s still becoming more of a corporate environment, and a lot of corporations have codes of conduct. And those codes of conduct often say things about how you conduct yourself on public forums or social media. The code of conduct for those companies also applies to the [Linux Kernel] mailing list.

So we need to figure out how to be respectful and civil to each other in the mailing list because this is no longer a volunteer project. People are paid to work on this. This is about making a workplace more civil and more respectful."

It's not just a good ole boy OSS project anymore.

I do not understand this argument (Side Note: I am fully supportive of the idea of being nice, encouraging, and forgiving towards people in Open Source development).

Unless The Linux Foundation (which is what pays Linus himself) states a code of conduct, and says that it will dismiss Linus unless he follows it, I don't see why he has to be respectful of, say, Intel's code of conduct. Also, what about an unpaid volunteer? Why should they respect corporate codes of conduct when they are not being paid a single cent?

Two Intel engineers working on Intel's internal kernel fork may well have to do that. But the precise reason it is an Open Source project is that it is unaffected by corporate policies of any sort. This applies to "codes of conduct" just as much as it applies to "keep your [computer] code 100% private".

It is still 'just' an OSS project.

And Python and other OSS conventions and projects are adopting Codes of Conduct and working through these pains.

Linus might want to consider that Intel is playing the salaries of professionals that are dealing with some tough hardware driver challenges. If he feels that unpaid volunteers can commit the same level of quality while enduring four letter words, more power to him.

Linux may be OSS, but pull the corporate support and see what happens over time. Corporations are currently willing to deal with these small issues of personality because they are mostly isolated from them and having Linux around is good for business.

This has been true for years, though.

Ignoring the actual assessment of "Is Linus too abusive?"...do those who defend Linus think there's no middle ground between harsh and benevolent forms of dictatorship (let's also ignore evaluating the goodness of dictatorships in general)?

In other words, does harshness positively correlate with success? Sure, you can cite Steve Jobs. But others could cite Woz, who while delivered plenty of harsh technical criticisms of his own, still seems to have avoided doing it in a way that denigrated both the person and the work.

The middle ground is Linus continuing to be who he is, and the corporate types can be as professional as their employers require. I strongly disagree with Sharp's premise that the entire kernel team should act professionally just because there are corporations paying kernel developers, who should somehow have elevated status because they now account for 80% of the development. The corporations are the ones who chose to play in a volunteer sandbox, and they are in essence volunteers of the project themselves; it's just their employees who aren't. That relationship is between the paid developers and their employers, not between the paid developers and the project.

In part of the conversation Sharp spoke up in, Linus was instructing another maintainer, Greg, to learn how to yell at people because he is currently too much of a pushover. Sharp's point is that maintainers should not encourage each other to verbally abuse submitters. There's a difference between being nice and being a doormat, but Linus is acting like there isn't.

If Linus is too abusive, perhaps this developer should consider joining the OpenBSD project. I hear Theo is a really nice guy to work with.

Is he actually nice or are you being sarcastic?

Theo makes Linus seem like Mr. Rogers.

Your answers lie in his quotes. http://en.wikiquote.org/wiki/Theo_de_Raadt

I especially like the one where he states that Linux is garbage.

People defend abrasive guy because they claim his abrupt communication style maintains quality code. But even more abrasive guy claims abrasive guy's project is garbage. Therefore, can we assume abrasive guy is full of crap because even more abrasive guy says he is?

I'm just trying to understand the logic behind defending such behavior.

The whole world is a Blub Paradox.

(Note: I like both these people)

Given the mailing list posts I've seen, at least with Theo you know the range of attitudes you're likely going to get. Heck, I got a pretty vitriolic email a while back for asking if OpenBSD wanted a donation of some unsupported RAID cards from a vendor that wasn't very supportive of OSS projects.

Linus's range of responses is much wider and more unpredictable, and he has a much larger audience.

As someone who's borderline ADHD and who frequently struggles with American social conventions (to the point where it's a significant stressor in my life), I'm much more threatened by people like Sarah than by Linus.

Sarah is probably more pleasant to deal with if you're socially skilled, but she demands the same thing in return, and will call you out---and apparently even go to the press---if she doesn't like how you interact with people. People like Linus will just pay attention to your actual point, and will look past the cultural subtleties of how you said something.

Every time Linus says something like, "shut the fuck up, Mauro", he's signalling that he wouldn't derail a conversation just because someone was slightly rude. For me, this means that I wouldn't need to stress over the wording of every email I send to Linus, because he's not going to care.

I'm not saying that politeness is bad---it's great if you're good at it---but random contributors enforcing politeness is toxic. Now, if I want to contribute to the kernel without getting publicly shamed on lkml, I have to be good at kernel development and at wording things in a way that's acceptable to people like Sarah. I don't think that's an improvement.

Ok, now I don't want to make this a gender issue, but this article really can't decide, at one point they say:

>I also got some really awful hate mail that tried to drag my gender in. But I don’t think this is a gendered topic

then a couple of responses later they say:

> I think that my perspective is somewhat colored by both my gender and my age

Which is it?

> So I think they picked up a little bit of the brogrammer culture and kind of said: “Yeah it’s OK to yell at each other because we’ll obviously just meet over beer and pound each other on the back and it will be all fine.”

> Sometimes, that just doesn’t work for people who aren’t men.

I also found this to be confusing....

N.B. I have no dog in this race, so what I say below is meant to be neutral to the issue.

I don't think there is any contradiction in what she says.

Gender, age, upbringing, education – all those things will influence someone's perspective on an issue.

But the issue itself is not about gender. Plenty of men don't want communicate via a lot of "Shut the fuck up, this is shit", and obviously women can swear and put people down, so it is not a gender issue. It is a style of communication issue, and the argument is:

Linus: I hate fake politeness and game-playing, so I'm just gonna let rip, especially on people who waste my time (implictly, to discourage that strongly).

Sarah: You can be assertive without being rude, and politeness is important in a professional context.

Her viewpoint might be different, and her gender might influence that, but that doesn't make it a gender issue.

Hum... How about her gender colors her perspective but is not the issue ?

Gender affect a lot of your environment and your education so you can't escape it's influence. But in this case gender is not a significative parameter in one's perception of Linus and his strong-mouthness.

The fuck do you mean, which is it? Those are two different fucking things, dipshit.

The fucking topic is whether communication on the KML should be a bunch of stupid assholes giving each other shit. The topic has nothing to do with fucking gender. She's just saying (as is obvious to anyone with a BRAIN) that who she is as a person influences her opinion on the topic.

Get it now, dumbass?

[this has been an example of how Linus would like us to talk on the KML]

Just because it's not a topic which is gender-related doesn't mean gender can't color your opinions of it.

I find it fascinating that "gender" completely replaced "sex" in American English.

I have zero dogs in this fight, so it really doesn't make a difference to me either way.

But communication isn't some binary thing. You don't need to either be an abrasive honest asshole OR a slimy political wheedler. It's perfectly possible to be honest AND polite.

I wonder how many more articles on the internet are going to be wasted on this topic.

Linus swears a lot is basically a "dog bites man" story.

I can't disagree with what Linus said in his mail and the rest of the media reaction seems like unwanted unnecessary attention towards this issue.

Pretty much. This is the biggest tech non-story I've seen in months.. anyone who follows kernel development knows that Linus drops the occasional precision "F" strike. This is neither surprising nor newsworthy. The only reason it's getting as much play as it is, would be because another completely unrelated developer (and a female one at that, a rare person) decided to take issue with it this time.

More surprising is the number of people (including the developer in question) who can't seem to grasp that because someone is swearing in a conversation with you doesn't mean they are swearing at you.

Go re-read Linus' message if this is unclear. He never once demeans the intelligence, ability, or anything else about the person, only calling out their mistake as egregiously bad.

The gist of the message is this:

You screwed up. Here's why you screwed up. You know better than that. Don't do it again.

Saying someone did something stupid and saying they themselves are stupid are two hugely different things. One of these thing is merely being acerbic, the other is actual verbal abuse.

There are several comments about how "corporate culture" is completely negative. Maybe some are. There's plenty of start-ups with a negative culture as well. But this is not a big business problem. It's a people problem.

Good communication makes everything better. EVERYTHING. It makes people happier. It helps people understand each other. It lubricates the transition of ideas into implementation. You can not underestimate how important it is.

Bad communication has the opposite effect. It makes people feel bad. It creates barriers to prevent people from accomplishing goals. It can spread negativity like a virus throughout a community.

When you lash out at people, you're creating a few problems. First is that the person feels insulted. If they are insulted, they won't want to work with you, and will actively campaign against you, unproductively. Even if you have a good idea, a person that you're verbally abusing won't give it a second thought, most often because emotion overrides intelligent thought. Finally, it stimulates a kind of general negativity in your community that simply spreads bad feelings and bad attitudes everywhere.

Nobody likes feeling bad. Nobody wants to fight over something stupid. If you want to get work done, the last thing you need is to be frustrated, and deal with frustrated people.

Furthermore, insulting someone does not make your point more right. It doesn't make it less right, either, but it's not helping at all. You might be able to bully someone into agreeing with you, sure. I've had people argue with me on every single possible point 'til i've given up out of exasperation. That doesn't make for good solutions to problems.

If you feel excited, agitated, defensive, or angry, write a letter in response - but send it to yourself. Wait 10 minutes and read it, and then consider sending it. Ask yourself if you can edit it down to be more concise as well as friendly. You'll end up finding people agree with you more this way.

Lol Gnome and Canonical a project that never listens to the community and tries to make their software less customizable and a project which for it size is contributing a very small amount back to the kernel.

I'll take Linus yelling at me because my code sucks over them just politely ignoring my contribution and not telling me what i did wrong any day.

A method I've found that helps when dealing with all sorts of people -- don't parse people's output according to your standards, parse it according to their standards. (It helps when they've published their standards, as Linus has)

This approach has made it quite easy for me to get along with people from all sorts of cultures, as well as all types of people within a culture (aspergers springs to mind) -- where others think "he said something that is offensive in my culture, he must hate me", I think "he said something which is acceptable in his own culture, he must think I'm doing ok"; or around my aspergers friends "she's being rude and insensitive, what a bitch" vs "she's being far more empathetic than she normally is, she must be trying really hard because she cares"

This could have been handled privately I think. She could have privately said "Linus, I understand you don't like my code but I would appreciate if you wouldn't use foul language".It seems to me she played the victim card. But whether or not Linus cussed at her, her code probably still sucks.

She's the USB3 maintainer, and as far as I know, there's nothing wrong with her code; Linus never even directed any vitriol at her.

I really like to read comments that start off with "Linus should do this and Linus should do that, he should be more accommodating, he should stop swearing, he shouldn't be so offensive, etc."

NO! It's perfectly fine to bring up an issue like this on a mailing list, private conversation or anywhere else... What is not fine, is still beating a dead horse when Linus says that he will not change. Who do these people think they are to tell someone else to change? Seems rather arrogant and offensive, doesn't it?

Not sure why this was left out of TFA: https://lkml.org/lkml/2013/7/15/329

That does seem like quite an overreaction to a mild suggestion from Linus.

You're only seeing the tail end of the conversation. The one that started it all seems to be this:

https://lkml.org/lkml/2013/7/13/132

That's interesting but doesn't seem to be what she is responding to. She's overreacting to comments that are tongue in cheek

"Linus Torvalds is advocating for physical intimidation and violence." is her response to goofy comments about the physical size of the patch reviewer.

Well, said goofy comments can be seen as childish attempts to fend off a real discussion.

I'm sure you could link us to dozens more examples that match this one exactly in every relevant detail.

However, this example is completely innocuous. He doesn't say anyone is a piece of shit, rather he identifies a particular piece-of-shit patch as being a piece of shit. Which, if Linus says it, is probably true.

She's equating verbal intimidation "don't do this ever again/this should not have happened" with advocating "physical intimidation and violence", or advocating "verbal abuse". On her blog, Sarah Sharp is "Standing up against verbal abuse". I feel that this is just as un-helpful as telling people "if you can't take the heat, get back in the kitchen".

Let's start from the facts: LKML is (unlike Wikimedia, or Ubuntu, or other 'pleasant' places) fundamentally open, allowing everyone to open their mouth and potentially be heard. To make this possible, it's necessary for Linus and others to sharply rebuke those that do non-constructive things, be it well-intentioned or not.

The desiderata here are: 1) the rules of what is considered constructive or non-constructive should be such that really everyone can follow them. 2) the people whose authority is trusted should be able to sharply rebuke those doing "shouldn't happen" things in a clear ostentative way, but without the whole thing into verbal abuse. This is clearly a cultural issue, and I feel that some people take reasonable feedback as intentionally abusive, and some people produce abusive language because of a lack of "stop for 10 seconds and let it cool down" when they feel their domain of operation has been encroached upon.

Let's not make a war among cultural groups (be it male/female, small-town/big-town or whatnot, with the whole playbook of us-vs-them rhetoric creeping along) out of what should be a joint search for the culture and conventions that are most effective for solving the task at hand.

I don't like abusive comments at all, and I'm sure that an abrasive environment will keep some good hackers away.

On the other hand, I'm also sure that an environment where you always need to bite your tongue and force yourself to be very polite, even when you're pissed off, would keep other good hackers away. Especially among those that do this for passion, and not because they're paid to.

I'm always baffled by this response.

There are many ways to tell someone that they are wrong without calling them a fucking idiot or without having to bite your tongue and forcing yourself to be very polite.

The thing is that there will always be people who have different thresholds for what is abusive, what is polite, etc. For example, here in Rome swearing is very common and isn't usually considered abusive if it isn't directed to a the person.

Example: "For fucks sake, why the hell did you name that variable X instead of Y, as our naming conventions say?" is not polite, but it's a critic directed at what you did, not what you are. On the other hand "I'm beginning to think that you're never going to be able to follow our naming conventions. You probably just aren't good enough" sounds polite, but it's really a demeaning personal attack. What's more abusive?

Between your two examples, the second one is by far more demeaning. Cursing is not always an example of being demeaning to another person. Your first example has two curses in it, to some just one, but they weren't directly aimed at the person being spoken to. That's just being rude by using language that many people may not like to hear in a public space.

If your first example was an abrasive comment directed at the person, such as "You are a fucking idiot because you can't follow simple naming conventions that a slightly trained monkey could follow!", then it would be demeaning towards that person.

I would also say that your second example isn't exactly polite either.

Between your two examples, the second one is by far more demeaning. Exactly. Being "professional" and being respectful are almost orthogonal.

Attitude is enough to be constructive.

When someone says something, replying "F*ck that" is as constructive as replying "I think there is a problem with your point". And it is really important when it comes to a critical area.

Being polite? Well, when things don't get messed up, yes. Otherwise, no. A polite vice-pilot is usually one of the main causes of a crash.

Abuse? Important lesson that most people of my generation and prior learned at about age 4:

"Sticks and stones may break my bones, but words will never hurt me."

Is the self-esteem movement of subsequent generations really putting words on the same level as fists, knives, etc?

It's a totally bullshit saying. A kid who's constantly told they're a worthless piece of shit by their parents is far more likely than average to wind up with psychological problems. Words have power. We can read great writing and be deeply moved by it. Why would you think they don't have the power to harm?

Because we're in control of our emotions and our reactions. We are not entirely in control of the laws of physics. Ergo, physical abuse is not even remotely in the same category as verbal "abuse".

On behalf of the kids you bullied at school I would like to say the following: fuck you.

Well, that's a constructive comment. The strange irony of what you said is that it was exactly the intense bullying I received as a computer nerd in grade school that helped lead to the understanding that I am in full control of how others "make me feel".

If you think that's intense bullying, you should be glad.

Another strange response, seeing as how I never even specified anything about my bullying experiences, other than their effects.

Right, I misread your post. My mistake, sorry.

It's probable that you had stronger psychological fortitude in the first place than many people. Not everyone comes to the feel-control realization on their own, some require intense relearning (therapy, counselling, etc.) to figure it out. Many people never make the connection.

It's also possible that one person's "intense bullying" is another's "gee, I wish my bullying was that low-grade".

Oh, in that case bullying is ok as long as enough people agree that it's low-grade?

It's also possible that one person's "low-grade bullying" is another's "gee, I'm glad my bullying is not that bad".

Not at all. I'm simply pointing out that without details on the bullying, it's entirely possible that @throwit1979 has little or no experience with the sort of viciousness that causes psychological problems for those verbally abused, resulting in the seemingly ignorant opinion that everyone has the power to fully control their emotions and behaviour.

Sounds like the intense bullying you suffered damaged you. You think of this as a good thing; many others would not.

> Because we're in control of our emotions and our reactions.

Again, bullshit. People with severe OCD, bipolar, schizophrenia, etc. can clearly have little or no control of their emotions and reactions. Abused children deprived of social interaction can lack the ability to even develop language. For example, http://www.tampabay.com/specials/2008/reports/danielle/

Your emotions are a physical process occurring in your brain, and you are not in control of them any more than you are in control of whether or not a bruise forms after you are struck.

I like how you put "abuse" in quotes like that.

Verbal abuse is recognized as a phenomenon in medicine, religion, and in law. It is not at the same level of unacceptable civilized behavior as physical violence, but it is still unacceptable nevertheless, and can carry severe consequences, quite rightly.

Many people are actually not deeply self-aware or in control of their emotions or reactions. This often is due to the environment they grew up in, their age/maturity, and can be overcome. It also may be due to illness that requires treatment. Some are victims of abuse, that hurts their career, enjoyment of life, etc. Yet abusers themselves are often the most egregiously acting without self awareness or emotional control, as they are often lashing out what they're feeling without restraint or contextual nuance, attempting to inflict pain on others.

Yet other people also are deeply self aware or in control of their emotions but due to community or societal norms are subject to systematic verbal abuse. Sexual harassment, threats against geeks/nerds, racial discrimination are all examples. There's often only so much any well-adjusted person can take.

You seem to be confident that words should not be able to inflict pain. Yet it's clear from history that they do.

"Sticks and stones may break my bones, but words can make me think I deserved it."

http://xkcd.com/1216/

Formal Sweatpants fixed that saying for you:

http://formalsweatpants.com/stick-n-stones/

"Sticks and stones may break my bones, but words will cut to the core of my beliefs and form the foundation of my existence".

IOW, Many people wind up using fists, knives, etc. in part because of the destructive words in their life. (And lead poisoning, apparently).

I hate corporate style.

Linus has organized thousands to work together to build something amazing. We all benefit. That's social genius.

After reading the lkml archives I find it hard not to conclude that Linus has done his best and it's time for him to step down. Linux is just too important to have such lousy leadership.

Why should he step down if he doesn't want to? If people are unhappy with him, they are free to fork it and set it up the way it pleases them.

Linux is not a technically impressive project. What makes it impressive is the fact that it is built, collectively, by many different parties contributing for their mutual benefit, from tech companies to academic researchers to hobbyists. The fact that Linux ended up being the kernel that got this massive social support seems to suggest it has good leadership with good social skills.

Over the years I've read some really asinine comments on HN, but this one takes the cake. Linus could do whatever the hell he wants. It is is project afterall. People are also more than free to take the source and do what they'd like with it.

Are you going to say that Linux is too important and needs to be nationalized next?

Reply on news.ycombinator.com