No more verbal abuse

I’m standing up against verbal abuse on LKML.  I will happily stand alone, however you can also support this cause.  Please speak up, either by resharing this post, or commenting on this post with words of support.  If you dare, you can also reply to my LKML email.

“Where do I put this fire? This bright red feeling? This Tiger Lily down my mouth? He wants to grow to 20 feet tall… I’m so tired of being shy; I’m not that girl any more. I’m not that straight-A anymore.”


Examples of verbally abusive behavior on the Linux kernel mailing list:

287 thoughts on “No more verbal abuse

    1. “Take a stand” against, what, exactly?

      A group of people getting things done? Oh yes, it’s a female proclivity to…stop people from accomplishing things by imposing one person’s views on an ALREADY-WELL-WORKING system?


      1. Do you have evidence that verbal abuse leads to productivity? Did it even occur to you that you might need any?

        (To say nothing of citing evidence about what a female proclivity is and isn’t. That’s not an invitation, btw.)

        1. Evidence that verbal abuse works:

          – Linux kernel development
          – Apple, ruled by abusive Steve Jobs
          – Microsoft, rules by abusive Ballmer
          – Oracle, ruled by abusive Larry Ellison
          – HP at the time of Mark Hurd

          Many of these were also accompanied by much more soft spoken individuals, e.g. Woz, Gates, … There are also plenty of counter examples, like Hewlett and Packard.

          So management by shouting around and management by walking around both work. But there is clear evidence that shouting works.

          1. The Linux kernel development is a prime example that verbal abuse does not work, at least in a free project.

            80% of its development is by paid pros. In other words, a lot of volunteer contributors have been driven away. Make a comparison with eg. Wikipedia, which produces more value than the Linux kernel with magnitudes less paid personnel.

            Did it? Now imagine Jimmy Wales behaving like Linus. What will happen with the Wikipedia contributors, and where will Wikipedia go very quickly?

            To me, this is a clear evidence that shouting may work in a corporation, but does not work in a free project. Never had and never will.

            1. Actually, I think these two statements are unrelated: “80% of its development is by paid pros. In other words, a lot of volunteer contributors have been driven away.”

              I think what happened is that corporations hired anyone who had experience with the Linux kernel. However, we can’t be sure whether all volunteers are turning into corporate employees without Greg’s master database (which he doesn’t want to share for privacy reasons).

            2. Uh, you don’t contribute much to Wikipedia, do you? It can be just as abusive as the LKML at times, only without the profanity.

              And Jimmy Wales should either participate more — or just go away & let a leader emerge. A lot of things are going wrong on Wikipedia because there isn’t anyone to provide leadership & build real consensus. (Versus the Foundation’s approach of making changes unilaterally in software & policy, then wondering why the community is angry about it.)

    2. Something tells me you have no idea what any of this is. This has nothing to do with women’s rights and everything to do with how a person conducts his own project. Direct your support elsewhere.

  1. Thank you for standing up for civilized discourse in a community and treating (potential) contributors as human beings. There is absolutely no reason to shout down people, insult them, intimidate them, etc, etc. Especially when they’re trying to contribute to, and help a community. If their contribution isn’t a good idea, or is unhelpful, there are plenty of ways to let them know this without using these tactics.

  2. I dunno, the profanity in your two samples seemed ever so slightly creative, but mostly stupid.
    Like, maybe, being flame-balled by a fetid illusionary dragon. I myself have been asked by Linus if I was on drugs because he decided to do something pointless with git and I objected. I thought that was pretty stupid and was severely underwhelmed. But I have not paid my dues, earned my stripes, or whatever with the linux kernel. You have. Go for it. You have a vote, I don’t.

    But I have seen it often said that Linus has only this one tool to chastise people. Give him a another tool that actually works. Don’t invoke the adainitiative, speak for yourself, you earned your voice.

    Keep up the good work, thanks!

    1. Perry, I don’t think it’s anyone else’s job to ‘give Linus another tool’ to replace his ‘be a douchebag’ tool, surely he can work out how to object with civility along with the rest of us. I’m not sure where anyone mentioned invoking TAI either, that seems a bit out of left field.

      1. I’ve read Linus’s responses to Sarah on the LKML thread, I suggest you do to. Other to mention that, I’ll pretty much shut up.

        Sarah invoked TAI at one point in this, that wasn’t left field.

        Having read more of the thread now, and thought even more-before-responding, I see Linus as rather comical, not especially in a good way. I’d hate to see him sharpen his wit and be actually devastating.

        Me? I’m being misunderstood again. I’m for Sarah.

        1. Perry: I put TAI on my original tweet because I hoped they could share it around. I would have also put in other organizations, like lwn and devchix, but I ran out of space. Tweet lengths are too short. 🙂 I’m in no way asking TAI to step into the conversation, only to spread the word about me standing up to Linus.

  3. Sure Linus may seem abusive on-line, but that’s just crap. Its just pixels on a screen. No one is being hurt or abused. I grew up in Boston, MA insults and verbal abuse is an art form and thats what we do. I have offended people with different cultural backgrounds as well.

    “No one can make you feel inferior without your consent.”
    Eleanor Roosevelt

    I think you are being abusive and aggressive. You step in to a situation that has been working for two decades and say it has to change because *you* are a self appointed spokesperson for the oppressed and abused and demand it so.

    The most vile and ruinous censorship begins as a call for civility and any such request tends to be quite dubious of intention.

    1. You are admitting to being an abusive person yourself here, both directly and by engaging in abusive tactics. You should be ashamed of your behavior; it’s socially poisonous at best.

    2. I hope this is satirical or ironic, because if not it displays a heroic level of cluelessness.

      1. Censorship isn’t at issue here, that comes from the government clamping down on dissent.

      2. “No one is being hurt of abused.” We’ll let that pass as the unsubstantiated bullshit that it is.

      3. “The most vile and ruinous censorship begins as a call for civility…” Are you seriously suggesting that asking people not to behave like dicks is oppressing people but that actually being a dick is not?

  4. As a person who has led a few small projects at my university and at work, I’m sorry to say, but: niceities don’t work.

    Whenever I tried to be nice and understanding whenever someone fucked something up, they didn’t learn. I HAD to be an asshole to get results. You sometimes HAVE to do break someone down and humble them to build them back up.

    Programmers are lazy. They need someone to whip them into shape. You have to have a thick skin if you want to stick around in this business.

    1. As someone who has lead teams of all sizes up to 40 people professionaly developing both hardware and software, I can assure you that being nice works — and works best. When you manage with the idea that everyone is a contributor whose role needs to defined in a manor which makes best use of her/him for all, including growing the developer, people naturally work together as at team. No one in the group want to be seen as holding the team/work back.

    2. Maybe the problem lies in your inability to express thoughts clearly. Being nice and clear are orthogonal concepts.

      Very glad that I will never work for you.

    3. I agree with you and have had the same experience. I think Linus tirades are called for. That is not to say its the only style that works, but its definitely needed for a lot of people. In fact I would say you’d need both the Greg (Teddy bear style) — good cop and someone to play bad cop every once in a while. That is not to say Linus always plays bad cop, but he knows when its needed.

      He’s yelling at people he’s known for years and has a fairly good understanding of their personality, more so than I would guess Sarah does. So the whole argument about Karate that Sarah gave is kinda pointless to me, but eh I’m not qualified to talk about karate 🙂 and there is the public looking in argument.

      1. Hmm I must have mistyped my first sentence by accident.

        I meant to say I DON’T AGREE WITH YOU.

        1) First up this one: that admittedly is over the top, but even Linus apologized and said that was not one of his proudest moments.

        2) The second onslaught you mention —

        Frankly I see nothing wrong with that.

        The reason is simple
        You say in your threads — just tell me to fix it and I will do so or something like that. don’t call me names.

        When Linus writes these very focused “insults” — he is doing so to change behavior — not to change the result of one incident.

        I must add being a black woman (though this isn’t a about being female or minority as you say), I’m pretty pissed about your behavior. The reason is simple — I work with a lot of guys as you must too, and when they talk with each other they are pretty blunt — like “Hey that’s a pretty stupid idea”

        When they talk with me they are never blunt and are always nice (a little too suspiciously nice in fact) — they’ll say thngs like

        “I wouldn’t have done it that way.”

        So I have to translate that to “What the FUCK were you thinking?”

        I don’t need these layers of confusion in communication. I much prefer the directness of men than the sometimes empathetic sugar-coating of women.

        1. So you think community disfunction isn’t a “real problem”? Ok, go back to working on your solo projects.

          This discussion has nothing to do with the gender of the people involved. Several male kernel developers have publicly spoken up for a more civil LKML communication style. It’s not just “women getting their feelings hurt”. So stop bringing up the gender card, and find a real argument against civil communication styles.

  5. Those kind of abuse need to stop. Newsletters are like a dark corner of the internet and those things need to come out more often

    1. A publicly archived mailing list is neither a newsletter nor a dark corner of the internet. Given that the list is read by every single Linux tech site writer going, even if she hadn’t re-tweeted it, it would have been picked up and dug into.

      FWIW, I tend toward Stephen Fry – swear words are powerful because we censor them, not vice versa:

      “The sort of twee person who thinks swearing is in any way a sign of a lack of education or a lack of verbal interest or -is just a fucking lunatic… I haven’t met anybody who’s truly shocked at swearing, really, they’re only shocked on behalf of other people.”

      So then, there’s the question of abuse. Leaving aside whether an individual message is an indicator of abuse (I tend to suggest it takes more than 1 and there’s no evidence of systemic emails to the same people) and the fact that in many large corporations, curt, bordering on rude and uncivil emails are fairly common in the upper echelons, the question has to be asked. Was the comment directed at the person or their actions or their code?

      The next question is whether the solution exists. Does resorting to fake comments and sarcastic remarks save anyone anything?

      Finally, it’s worth pointing out, that, as noted different cultures take different attitudes towards this. For example, as a Brit I’m aware that we make a ton of comments in discussions which aren’t immediately clear or might be misinterpreted as to the prioritization of them. Someone whose natural grasp of language isn’t English (or just someone with a different background might not see the subtext).

      As an aside, I really doubt anything will be solved before the Kernel summit, email is a very impersonal and terse medium. Nor do emoticons/emoji go any way to solve this.

  6. Linus is interested in the Linux kernel, not your feelings. He created it, cares for it and takes personal responsibility for it. You create, care for and are responsible for your feelings. You volunteered to help out on the Linux journey. He never volunteered to care for your feelings, nor did anyone else. It’s an opt-in community and you can always opt out at any time. Caveat emptor.

    1. Linus is interested in getting the best code out of his developers. Being verbally abusive is not helping some of those developers (including but not limited to myself) to work on the Linux kernel.

      No, I can’t opt-out. I’m paid to work on the Linux kernel. 80% of the Linux kernel developers are paid to work on Linux. This is not about a volunteer project anymore. This is about a person paid to manage the kernel (Linus) verbally abusing other kernel developers who are also paid to work on Linux. This is about making a workplace (the Linux kernel community) a safe workplace, even during heated technical discussions.

      1. Yes people should make an effort to be civil. But killing projects by continuously asking questions, deferring decisions, tacitly agreeing while privately resolving not to support something, all of these things I see daily at work, which are really pretty shitty when you think about it, and nobody needed to swear to do them. I could do with more candid discussion and a lot less politicking from the people I work with.

        But I also feel pretty terrible about this whole thing, because this is obviously the last straw for you in a long pattern of being disenfranchised, and you’re attempting to learn how to speak up for yourself.

        1. Sounds like you are mistaking a managmement problem for a worker problem. All the stuff you’re seeing at work daily and attribute to “killing projects” is bad management. I’ve worked at companies which resemble your environmnet; the reason for the workplace being crap was trickle down from top management.

      2. So, apparently, the problem is that you don’t like the job you have an obligation to do, because Linus is an ass. Well… he doesn’t owe you anything. As long as the kernel is being well maintained everyone should be happy.

        Not that I don’t think he isn’t being abusive (he is), but going on a crusade against someone’s personality isn’t exactly a good thing. I’ll conclude with a message:

        1. That’s an interesting take, which reveals an angle on the shot. For reference, the deal with it idea is that the cars are going to have to deal with it, because the swan isn’t going to move.

          The idea that you take the position of the cars and that running roughshod over the swan is the standard response is, as I said, interesting.

        2. “…he doesn’t owe you anything.”
          Actually he does owe her; she’s why Linux had USB 3.0 support before any other O/S. And he does owe the other developers courtesy by the very fact that they are all working to keep the kernel up-to-date while improving it; he needs to be aware that they are working on the kernel, not because of him, but because there is a whole lotta engineers and other folk that strongly believe in open software.

      3. This. People are most productive and creative in positive and supportive environments. You can be uncompromising about quality while still being a positive influence. I had no idea the linux community was so abusive. I love linux as much as the next guy, but just from reading this I can tell you that I will never touch that codebase with a ten foot pole.

        1. Some people are far, far more productive in a challenging, competitive environment.

          1. I don’t see “challenging and competitive” and “positive and supportive” as mutually exclusive concepts. Abusive or harsh language is certainly one way to enforce a challenging and competitive environment, but it’s not the only way to create one, and arguably not the most productive way to create one either.

      4. Exactly! It’s companies and ppl like you who further the Linux-Kernel-cause … and someone bite me if you had to sign a clause in your contract with Intel that you have to take abuse from a Scandinavian long-term-teenie who thinks collaboration is a rap-battle on 8 Mile with a lot of hissin’ and dissin’! And most certainly the guy from Redhat didn’t sign anything neither.

        It’s really a shame that so many think that attitude outweighs the overall goal of furthering a rock-solid OS that actually would have it in itself to rearrange the OS-ecosystem for desktops. No wonder many corporations are very reluctant in getting involved deeper.

  7. Just wanted to add my voice of support. There was some stuff that I wanted to do to contribute to the kernel, but was turned off because of the tone of LKML.

    I dislike fake politeness as much as anyone, but I only have a limited amount of time and effort to give. I gave up petty schoolyard politics in high school. I don’t want it to infect my hobby.

  8. While I think that Linus goes overboard pretty often, I don’t think the LKML is ‘too hostile’ for the average contributor. At least when I read it, I don’t think what’s posted is degrading in a way that really sticks to you / humiliates you / discourages you from participating.

    I also think there’s value in expressing emotion on the list, as well as veering off topic sometimes or making a lame joke about pot brownies 😉

      1. “shut the fuck up” or “please do not speak again for a while” might sound differently and deliver a different message.

        Probably the second might have more impact on people used to the first expression or none at all.

        In the end of the day you just want that person to shut up, which is the best and most effective way to achieve that has to be proven.

  9. Your post got me thinking a bit: does being direct correlate with being verbally abusive?

    Let me back up a bit. I’ve worked in a few companies over the years. Most have kept the internal communications professional and criticism tempered — often with the result that the “yes-men” rise through the ranks and the company teeters.

    My current company, by contrast, prefers to keep communications direct — managers are even coached on this. The directness is an aspect I find refreshing and has helped keep us nimble and focused. On the flip side, a decent plurality interpret this as allowing verbal abuse (in other words, they think they have license to be an asshole) — a huge turn-off, and I’m certain we lose a few good candidates and employees this way.

    Clearly, the two don’t *have* to go hand-in-hand, but my observation is there’s a definite correlation. I’d like to think there’s an easy way to break this chain.

    1. There’s definitely a distinction between directness and verbal abuse – and I agree that directness can come across as harsh, impersonal, and inappropriate in a lot of situations. Poor or ineffectual communication deserves to be called out too, but it’s not the same as abuse.

      Abuse is usually delivered with an intent to manipulate, diminish or control – it’s a non-egalitarian approach to interpersonal relationships. Unfortunately, it’s a lot easier to detect once you’ve been exposed to a lot of it. I’m hoping these sorts of discussions help us get a better idea about these distinctions though, because it’s definitely hard to have a discussion about abusive behavior when not everybody’s on the same page about what it looks like.

      1. How do you know he intends to “manipulate, diminish or control”. I’ve used such language a lot in the past and it is always meant as a joke intended to defuse a situation which could easily get truly hostile. I welcome people who will interact with me in a similar manner because I want to know how they feel. When they communicate in a level even monotonous form, I can’t hear any distinction between the sentences. I can’t hear which part of their argument is the important part and which is grammatical. I can’t hear the tone in the words written on the screen in front of me. I can’t tell if they are agreeing or disagreeing. It leads to a great deal of confusion and much more pain down the line when we realize neither of us understood the other.

    2. Linus has real authority. He doesn’t have to insult people to get things done. Screaming reduces the respect people have for him. It doesn’t magnify his power.

      Talking calmly, being respectful, and having your facts straight are the best way I know to assert intellectual leadership.

      1. “Linus has real authority. ”

        He can’t fire people at other companies. There’s a limit to his ability to really effect someone’s livelihood, unlike an actual manager.

        1. Linus decides whether he will pull from someone. He’s “fired” top-tier maintainers before. If that someone’s job depended on them being a top-tier maintainer (or even getting their employer’s code into the kernel) they could be let go from their job. For example, Andi Kleen was blocking the x86 and amd_64 merge. He got “fired” as a maintainer by Linus, and three new x86 maintainers took over. That incident is something employers remember when trying to hire new kernel developers.

  10. I disagree, based on the two examples you listed. I wish more development leaders took this approach (confined to cases similar to those you quoted, not as a general way of behaving).

  11. With the way linus is communicating, he probably cares more than other people.

    And he did not only said what has to be said no he also described the issue well.

  12. This popped up on my radar and I wanted to suggest something. You’re probably not going to like what I have to say, but stick it through and really think about it. I’m actually on your side. I believe we met briefly at OS-Bridge or somewhere else while I was mucking around PDX. Hi again.

    Linus wasn’t advocating violence. He was commenting on a man’s immense size, likening him to a giant that might accidentally squish someone, like I’m sure most of us have accidentally squished a bug here or there. That is miles away from threats or advocacy of violence. You read into his statements and pulled something out that just isn’t there.

    As to getting rid of verbal abuse on the mailing list. Why are you okay with shouting matches in person, but not okay with it on a mailing list? That doesn’t make any sense to me. People shouting at me in person often triggers an adrenal dump and begins the fight or flight response, which ends any possibility of elevated discourse.

    I don’t see any problems with those two posts that you cited. They call the statements stupid and question the maintainers for their competency, but the patches broke things and were bad. If someone’s work breaks things and is bad, they should be called out for it. Sure, it sucks to be called out for making a mistake, but that’s the only way that people can get better at doing anything. Public shame is a powerful motivator. And in a volunteer project it’s the only extrinsic motivator available.

    1. Kevin, please read the emails linked at the bottom of the post. This isn’t about the original thread. This is a complaint about Linus’ (and others) past verbal abuse of other kernel developers.

  13. You’re no one compared to Linus. Start being Alan Cox or Theodore T’so first to criticize him for his behaviour. Not everyone is fucking American with your bullshit about professionalism equals political correctness. Nordic people can be nasty, and this is a good thing.

    1. No one said professionalism == PC. The way you speak makes you sound more like an idiot than honest. It’s also an insult to Nordic people to say they are noisy and ill-mannered, because they’re not.

      Since you don’t care about what people say, I’m sure you won’t mind when I tell you to fuck of — no one is interested in what you have to say.

      Have a Great Day!

    2. I have to concur somewhat. Some of this is about male hacker volunteer engineer culture vs non-gendered professional culture. But some of it is also about US professional culture vs other international cultures.

      Here in Australia, I’ve heard and used language in the workplace (from men and women) that I would never, ever use when working with colleagues in the US. And for an Australian I’m a polite, quietly spoken person. Nordic colleagues combine a sort of Germanic directness (which can be quite shocking when you first encounter it) with the same relaxed attitude to swearing that I find in Australia.

      Male privilege is a thing. US cultural imperialism is also a thing. I can see good reasons to try to shift Linus’ mode of communication, and thereby the overall culture. But for you Americans on this thread, please try to back out of your cultural bubble, and remember that things are very different elsewhere and your cultural expectations have no particular right to overrule those of others.

  14. Any adult could switch on their TV and hear language a hundred times as bad as any of the examples you posted.

    I appreciate that you’re offended by this, but I think you should consider that this simply may not be the show for you. Sure, it’s R rated – and so are many movies, TV shows and stage shows. If they’re not to your taste, you switch off and go someplace else.

    Getting out of your seat and telling the director he has to change the show – a show which thousands of other people are enjoying, people who revel in its frankness and bawdy language – isn’t a mature way to deal with being offended.

    1. Working in the Linux kernel isn’t like watching TV, so your analogy is faulty. This is about standing up to the director of a *project* and telling him to stop abusing *his employees*. The fact that someone else is paying the employees paychecks is irrelevant. 80% of the Linux kernel developers are paid to work on Linux. They don’t deserve to work in a community where they are verbally abused.

      1. I’m sorry, but it’s not like that, because the fact that someone else is paying the paycheck is 100% relevant. Especially since the companies who pay the developers don’t have a contractual relationship with Linus Torvalds. Linus is who he is, and he’s been acting the way he does for 20 years. He’s not going to change, nor is there any reasonable way to compel him to. Feel free to tilt at this windmill, but your only realistic option is to find a job where you don’t have to deal with him, or learn to live with it.

      2. Actually you are wrong about that. The fact that someone else is paying you is very important because it means you have no obligations to Linus (e.g., to be a competent employee) and he has no obligations to you (e.g., to treat you fairly or behave “professionally” and courteously).

        Calling Linus out personally for his interactions with others is just another adversarial approach.

        1. So, just because two people work together on something, but get paid by different employers, they shouldn’t be civil to each other? *Snort*

          By your argument, political leaders who are paid by different countries are under no obligation to be civil to each other. Teachers should be allowed to cuss out students from different schools. Seriously, the world does not work this way. Verbal abuse and person attacks are not appropriate in any setting.

  15. I fully agree with you. Linus uses very abusive language. He does it a lot. If the Linux kernel is his private property and he recruits people to work on it, then may be he can foul mouth people. It’s as if he owns the code. This guy needs to learn some manners, no matter if he some kernel god. He could simply, clearly, and openly say that he does not want to accept the pull request, because of so and so reasons, without having to use foul language. Blaming it on the culture is no excuse.
    I am not basically a kernel developer. I seriously wish GNU/Hurd to catch up. Or systems like Haiku, where people are very friendly. Till then, may be we have to put up with his arrogant behavior.

  16. Hi Sarah,

    Linux Kernel development process is distributed enough so you don’t need to interact with people you don’t want to.

    However, you have _no_ right to push what you deem to be appropriate behavior onto anyone. Not even Linus.

    Although I see as a valid initiative to bring some cordiality to LKML, I wonder why you did not stand up against other people with similar behavior in there. Choosing Linus seems more a political and symbolic movement than a genuine one.


    1. Limiting my interaction with Linus is a career limiting move. It means I can’t be a top-tier maintainer, or a stable tree maintainer. Opting out of communication with Linus is not an option.

      I do have a right to ask for professional communication in Linux, which is now an 80% corporate-backed project. I have a right to ask for a safe workplace, a civil community. We may not agree on the baseline of civility, but even users on slashdot and hacker news realize that Linus is being a jerk.

      Yes, there are other people in the Linux kernel community who are also verbally abusive. Linus sets the tone for the project, and other people pick it up. If he changes, other developers will have to take a look at their own behavior and decide whether it is acceptable anymore. If you want change to happen from the top down, you have to start at the top and work your way down.

    2. What is “deemed appropriate” is indeed “pushed” onto everyone, everyday, everywhere — it’s called society, and we all live in them; if people set out of what’s considered appropriate, there is a kind of backlash.

      Pick up your copy of Rand if this thought makes you uncomfortable, but reality is something you cannot completely escape.


  17. Why don’t you play the “Intellectual Honesty” Card?

    Yes, let’s move this conversation into the “how to work together DESPITE
    people being different” realm. I would be happy to have that
    discussion. As Linus said, some people work together better than
    others. Some people have different expectations of appropriate ways to
    interact with co-workers. Sometimes that means that people only work
    with certain other co-workers, like Greg and I.

  18. I’ve been waiting for someone to stand up and say something. You did the right thing and you seem to know it.

      1. Calling a woman a “whore” — hmm, original. And you are a contemptable person.

        1. I’ve always understood the phrase “attention whore” as being unisex.

          Of course, it has also attained the status of a thought-terminating cliche, and an ad hominem one at that.

          1. I think the correct term for a male “drama queen” would be “attention manwhore”. A more polite (and non-gender specific) term might be “drama llama”.

      2. And you Sir, shouldn’t have open your mouth and shown that you without doubt are a misogynistic idiot who hasn’t reached the 21th century, like the rest of us.

  19. The LKML is not a place for easily offended girls to be. Get over yourself.

  20. You undermined yourself by ranting in a reply to good-natured banter. The tone of the e-mails clearly went miles over your head. Your accusation of Linus of advocating for “physical intimidation and violence” is beyond ridiculous and is quite unreflective. Everything about your e-mail is fantastical in the thread’s context.

    I understand that what you are actually upset about is Linus’s earlier e-mails that you have linked to. Of course there is legitimate discussion to be had about those, but in your eagerness to get a word in about things that happened months ago you just kind of lost perspective here.

    By the way, those older e-mails have been discussed to death, and you haven’t added anything new. You basically walked onto a battlefield where the battle ended months ago, saw that a battle had occurred, took sides, and started shooting at those who were involved and have moved on.

    1. The door was opened by this particular thread, which is why I replied to it. If I hadn’t replied with such strong language, and I had just asked politely for the kernel communication style to change, nothing would have changed. Linus is bull-headed, and everyone else would have ignored my reply. Now that people are paying attention, and I can go back to being polite. If you read the thread, it’s been slowly deescalating, and now we’re moving on to constructive responses, like putting this on the KS agenda, and talking about what behavior is and isn’t appropriate.

      Those older emails haven’t been discussed to death. No one spoke up. The maintainers involved moved on without criticizing Linus for his tone. Someone needed to speak up, and I guess that someone would be me.

      1. “If I hadn’t replied with such strong language, and I had just asked politely for the kernel communication style to change, nothing would have changed.”

        You have just fully conceded Linus’s point. You seem to justify his use of strong language. Read what he wrote:

        “The fact is, people need to know what my position on things are. And I can’t just say “please don’t do that”, because people won’t listen. I say “On the internet, nobody can hear you being subtle”, and I mean it.

        And I definitely am not willing to string people along, either. I’ve had that happen too – not telling people clearly enough that I don’t like their approach, they go on to re-architect something, and get really upset when I am then not willing to take their work.”

        And the older e-mails might not have been discussed to death on the list itself, but they indeed have been in the community at large. At this point, it is developer lore. For example:

        There are also a lot of discussions about mean behavior in general, such as

        Or try:

        This is all just that one website. There are numerous blog posts and endless threads on other e-mail lists.

        1. I don’t read ycombinator or hacker news. These threads have never been discussed *by the kernel community*. A re-hashing on news sites does nothing to change the community culture. Change has to come from within, and it starts with a member of the community standing up and saying, “No, this isn’t fucking cool.”

          1. Hi Sarah,

            I have read the mailing-list thread and the previous threads you linked.

            I think Linus is sometimes rude in his responses. I do not, on the other hand, think he is abusive.

            The worst words used in the threads you linked were “shut the fuck up”. From your message above, I assume the “fuck” word is ok to use in some instances.

            As for telling someone to shut up, it is rude but I don’t think it constitute verbal abuse.

            I don’t have access to a reputable dictionary to quote the definition of verbal abuse, but none of my search results match your definition:
            (I don’t cite the current version of wikipedia because otherwise people will surely modify it)

            As for the rest of Linus’s rants, he basically called out the idiocy of his lieutenants acting how they did. I think it is ok to say something someone did stupid. It is not an attack on their character.

            I don’t quite understand how you can talk about emotional breakdown and such as you have shown to be quite emotionally strong in this very campaign.

            1. You’ve been looking up “abuse”, not “verbal abuse”. Here’s a good reference for symptoms of a verbal abuser. Note things like: belittlement, demeaning statements, hysteria, name-calling, raging and violent statements, and mocking sarcasm.

        2. I don’t think Sarah is against that particular point. She has argued that you can get you point across without verbal abuse. Yes, she used strong language but she didn’t verbally abuse anybody. And she argues that Linus has (and I agree). Heck, even Linus isn’t proud of some of his replies to the mailing list. Even though I like the man and they guy is an old dog, it’s for him to learn some new tricks.

  21. I, for one, would be humbled if Linus had something negative to say about me in any way, especially if it were about a skill of mine that I could improve. Perhaps this is more a sensitivity issue, and some folks don’t agree with a healthy berating… I wouldn’t treat people the way Linus does, but it certainly wouldn’t bother me if I were on the receiving end. Maybe you ought to find a different way to enjoy Linux as a hobby, or just don’t work as much with Linus. In any case, good luck.

    1. Linux isn’t my hobby. It’s my job. It’s career limiting to *not* work with Linus, because that means I will never be a “top-tier” maintainer, and never maintain one of the stable trees. So I will continue the conversation with Linus and the other kernel developers, and hope we can come to a middle ground on communication styles.

  22. Being direct, unambiguous and letting your feelings known is part of lkml culture. Didnt you get the memo?

    1. I’ll assume you know basic C and spell it out to you:

      true = ((Being direct || unambiguous || showing feelings) != (emotionally tearing someone to shreds || cussing || verbal abuse))

      LKML is the way it is because of Linus and people who follow his example. That doesn’t mean it’s right.

      1. Do study the C language order of precedence rules again…

        != has a higher precedence than logical OR thus your code does not work properly.

        Or maybe it was sarcasm.

        1. Ah, thanks for the reminder! I wrote that fast, and didn’t think too much about correctness. The comment is fixed now. 🙂 Hooray for open source code review on blogs!

  23. You’re my new hero.
    While I do “get” that swearing is kind of Torvalds thing, swearing and insulting isn’t the same thing. But mostly I just can’t get over the fact that he’s arguments are at the level of a five year old thinking he’s the only person in the world.

    1. “While I do “get” that swearing is kind of Torvalds thing, swearing and insulting isn’t the same thing. But mostly I just can’t get over the fact that he’s arguments are at the level of a five year old thinking he’s the only person in the world.”

      Being explicit is talking like a five year old? Do you think Torvalds is five years old?

      ” swearing and insulting”

      Please do enlighten us with al Torvalds “insulting”. The chances are bigger Sarah will get caught high of brownies than Torvalds going ad hominem on his beloved kernel devs.

      1. I am pretty sure that GR was saying that Torvalds’ arguments in defense of his behavior are like a 5-year-old’s. Which they are. Something like: “I have to be rude to get my way, because it helps me get my way!” Basically the idea is that I’m allowed to be a dick because it benefits _me_. No need to consider the other person (let alone consider the effects on the people watching).

        People like Linus, with very high profiles (almost celebrity status) should actually be held to a _higher_ standard of behavior, because their social influence means that more is at stake than just the people immediately involved. But free software culture is backwards in that respect — the degree of tolerance of bad behavior is higher exactly where it ought to be lower.

        1. And, “I’m allowed to do this because it benefits _me_” is a clear sign that the behavior is indeed abusive. The argument for the benefit of the community is so much harder to make – those making it seem to have abusive tendencies themselves.

  24. I agree with you about this, Sarah, and I wish I had clout in the kernel community so that I could stand with you in more than just spirit. What troubles me most about what I frequently hear from open source developers on this subject (and this includes some of my co-workers on the Mesa team) is the belief that by being abusive and dismissive to each other we are actually improving productivity. Based on my mostly closed-source experience, the correlation actually runs in the other direction (groups that are more respectful are more productive), and I believe the Linux community is successful in spite of the abusive elements of its culture rather than because of them. But I don’t really care whether I’m right or wrong about that, because IMHO some things are more important than productivity (gasp!). Being abusive and dismissive creates a hostile work environment for everyone, and alienates an important subset of our community. We need to put a stop to it, not just because it is contrary to the inclusive spirit of the open source movement, but because it’s simply no way to treat our fellow human beings.

    Thanks for fighting the good fight.

  25. I don’t know why people can’t understand that being direct and being rude are 2 different things. And people should stop branding things as “politically correct” because it is something they don’t want to hear.

    A thousand years ago, when I first started working with upstream, I was working in an area outside of my expertise, so I posted a question to one of the upstream kernel lists. I got basically ridiculed, and told I was an idiot and given a ridiculously harsh response to my question. I emailed the maintainer privately (this was davem), and basically said “hey, that was uncalled for”, and he told me he *had* to react that way because of email archives and he wanted to make sure nobody ever dared to ask that question again. Now, obviously, this abuse didn’t make me walk away from kernel development, but seriously. Someone else might have just gotten afraid to ask *any* questions, and is that really what we need as a community?

    I have given training to people at work on how to interact with the kernel community, and in my training I actually have to add a section talking about how *other* people flame and given personal insults, but how *we* should never react in kind. It seems to me that we might need provide some training to the maintainers at KS – how to provide negative feedback without personal insults. Something I learned at work on day 1, but apparently has never been taught to some of these people.

    1. ” Someone else might have just gotten afraid to ask *any* questions, and is that really what we need as a community?”

      The problem is that, as a practical matter, it’s a problem if your mailing list gets filled up with noobs asking RTFM questions, only some of whom will really be serious about contributing.

      If that happens, real work will be harder to do due to the noise, and people needing to do real work will move to other routes of communication. And then the people who really do need questions answered will lose access to those who can answer the questions.

      Frankly, mailing list based communities, that are actually trying to make progress on a project, don’t scale very well.

      1. If the problem is noobs asking RTFM questions, obviously there’s a problem with the ‘M’. Maybe the manual isn’t complete or up-to-date. Maybe it’s hard to find, and isn’t in Documentation. Maybe someone needs to write a tutorial, or write some “step zero” documentation on how to set up the toolchain. Maybe there needs to be a TODO list, along with a list of things that should never be done. Newbies asking questions is a failure of the maintainer to provide adequate documentation to answer those questions.

        1. > If the problem is noobs asking RTFM questions, obviously there’s a problem
          >with the ‘M’.
          From several years of helping people on IRC, I can tell you that that your statement does not always, nor even usually follow.

          > Maybe the manual isn’t complete or up-to-date. Maybe it’s hard to find, and > isn’t in Documentation. Maybe someone needs to write a tutorial, or write
          > some “step zero” documentation on how to set up the toolchain. Maybe
          > there needs to be a TODO list, along with a list of things that should never
          > be done.

          Maybe; however as with all things in FLOSS, there remains the question of who will write them: typically the person who identifies a problem as meritorious of someone’s work gets to do it, or at least coordinate it. But that’s not always seen as leading to the “top tier” career path.

          I think this also bears reflection:
          “Programming is a /constructive/ activity. How can a constructive, inventive activity be taught? One method is to crystallize elementary composition principles out of many cases and exhibit them in a systematic manner. But programming is a field of vast variety often involving complex intellectual activities. The belief that it could ever be condensed into a sort of pure recipe teaching is mistaken.
          “What remains in our arsenal of teaching methods is the careful selection and presentation of master examples. Naturally we should not believe that everyone is capable of gaining equally much from the study of examples. It is the characteristic of this approach that much is left to the student, to diligence and intuition.” (Wirth)

          And that’s the point wrt documentation and newbies: the simple fact is that there are an awful lot more people who want to get involved with Computing, and even with the same proportion or greater of successes (ie people who learn to program well enough to spend the rest of their lives doing it, irrespective of how much they get paid, since it’s what they *do* and how they define themselves) there will be an awful lot more who decide it isn’t for them. A large majority of those will, in my experience, simply be dipping their toes in the water, and not really enjoy having to self-start with codebases, even their own.

          > Newbies asking questions is a failure of the maintainer to provide adequate
          > documentation to answer those questions.
          Not at all. Quite often it’s that they don’t bother to look up the documentation, even when it’s given to them, but rather expect others to do the work for them. (After all, that’s what computers are for, right?) The people who ask questions based on actually trying things are much fewer; but that can be nurtured, assuming they’re not going to give up in 5 minutes.

          That’s why we have webpages telling people what they’re getting into:

          are both a bit friendlier than eg “smart questions” but ultimately if someone doesn’t want to take the time to consider what’s said to them, nor to read any of the relevant documentation, there’s not much that anyone else can do. They won’t get very far, and sometimes that’s enough to get them to realise what we’re on about; sometimes it’s a cue for a tantrum, as their unreasonable demands that everyone else do their work for them, come up against the reality that no-one cares.

          None of which excuses berating newcomers or belittling them. However, so much of programmer activity is based on evaluating what is in front of you and rejecting much of it as rubbish; rubbish that you wrote that now needs changing. So there’s also a tendency to attach that label to stupid ideas or designs that make no sense, and further to resist any perceived acceptance of it. Not to do so leads to code that does not work; and experience based on prior bad judgements leads to aversion.

          That tendency will never go away with computer programmers, since it is _inherent_ to the activity. GIGO always wins. But then, no-one likes to be told what they wrote is garbage either, except a few who appreciate the consequent elegance when they ditch a large chunk of code, in following a better approach.

  26. Well said, Sarah. Sorry I missed the window on g+, sounds like it got ugly. I have often been faced with the decision whether to jump into the fray and speak up, and too often I keep my mouth shut. The fact that you received that level of grief speaks volumes about the health of the community.

    This isn’t about cursing – it’s about kindness, compassion, maturity, and all those other parts of human interaction that seem to get lost in a lot of technical discussions, particularly on LKML. The tone of the list is set, and thus the tone of the group of maintainers is set, and I don’t think they even realize how limiting that is. Imagine what Linux would be like, if only.

  27. i agree with you sarah, you are right. i think that is obvious and i doubt anyone sane will disagree with you regarding infantile behavior on LKML. that said, linus doesn’t appear to be interested in moderating his behavior and in the end this is a volunteer opt-in organization and linus is afforded the freedom to be a prick if he’s posting in the US.

    that said, trying to turn the LKML into a long-running political discussion may not be useful to you. this topic has been covered before, everyone understands who they are dealing with and chooses to participate because the advantages outweigh the disadvantages. my assumption is that as this goes on, you will simply be placed on a filter list.

    my advice is to simply fork the kernel and start a new org with ground rules you can assent to. someone will eventually fork the kernel with success, you might as well try your hand at it.

    1. You miss two points:

      1. The Linux kernel is not a “volunteer” project anymore. 80% of the Linux kernel developers are paid to work on the Linux kernel. As we transition from a “college project” to a “serious corporate-backed project”, abusive behavior simply cannot be tolerated any more. Opting out is not an option when you are paid to work on the kernel.

      2. Forking the Linux kernel is not just a simple matter of putting a new repo on git hub. A fork will die unless there are backers of the fork to work on. The Android fork survived because it had the weight of Google and the phone hardware vendors behind it. Unless a new fork has similar corporate backing, and a body of developers to work on it, the fork will simply die.

      1. I think you can point out these facts to the employer that is forcing to you to collaborate to the Linux Project, in order to find out another project or task , you can go to your Human Resource department try to get a solution for you, a new project in other unix .

        You have two possible actions:

        1. Ask Intel to fork the kernel in order to became a corporate solution.

        2. Destroy Linus funding money with Linux Foundation making the actual Company supporters stop sending checks.

        I hope you won’t win if you already though in the second option.

      2. Firstly, respect for standing up for others who may be too cowed or inured to the abuse to stand for themselves.

        I fear however that you’re in a losing battle, Linus has had the luck of never needing to grow up, and most likely won’t do it now, and by the special nature of the kernel development i don’t think he can by reached by anti bullying laws.

        I wonder though if there aren’t any reasonably mantained Linux forks that could be used as a distro core. If there is enough demand, distros could offer a “Linus free” distro for those who care. It’s the only way I see he could be pressured into behaving like a grown-up.

  28. I am reminded of Alan Cox and Linus’ insults on his TTY patch. Alan Cox split. Then there was Con Kolivas’ scheduler upon which Linus just shat great toordes. Con split.

    So, the linux community has lost talented kernel devs and prolly frightened many potential ones off. It’s not a new thing, though. He’s been like this since Tennenbaum and Usenet days. But it’s Linus’ kernel. Just like OpenBSD is Theo De Raadt’s baby and he’s FAR worse than Linus. We’re free to fork the code and roll our own if playing nice is that important.

    1. For Alan himself:
      “I’d like to assure everyone that while I frequently think Linus is an asshole (and therefore very good as kernel dictator) I am departing quite genuinely for family reasons and not because I’ve fallen out with Linus or Intel or anyone else. ”

      but please do not let facts get in the way of a good argument…

    2. Theo is not “FAR worse” than Linus. Theo doesn’t flame innocent, well-meaning people. Plus, his flames are more clever than Linus’

  29. Hey Sarah,

    I was just following up on what was said yesterday about trying to convince Linus with data showing that improving the atmosphere on the ML can improve the productivity of the maintainers of (and thus the quality of) the kernel. Have you or anyone looked into the research yet? If not, I’d be happy to take a look and send some resources your way.


  30. shit, just what we need – a bitch running around crying about how hurt her feelings are.

    1. Shit, just what we need, another cowardly troll hiding behind anonymity to shit on others.

      1. Just watch we need – another paladin ass wiper ready for everything to get closer to pussy.

        1. Shit, just what we need — another overgrown 5 year old who thinks the only excuse for acting like an adult is to get laid.

          1. Shit, just what we need – another reply to this thread. Don’t I have anything better to do?

  31. Supportive for Sarah. Constructive criticism doesn’t require being shamed or abused. The fact that Linux kernel is alive and well doesn’t mean that the management style has no room for improvement.

  32. Thanks for speaking up about this. I have long wondered why people put up with Linus’s abuse. Mostly I try to ignore it, on the grounds that it’s not worth my time to put much emotional energy into jerks, but you’ve reminded me why it’s important to speak out about this kind of stuff. It needs to stop, and Linus needs to understand why it’s not okay.

  33. In a vacuum, civility is preferable to incivility. But there is plenty of context here: 1) Were Torvalds’s comments off the mark? Unnecessary? Overstated? 2) Is the subject matter important enough to merit strong language to make a point? 3) Would the alternative of a more-democratic process to close debate on a subject be better, or worse? Or is the Linux community getting as close to optimal value from this process as it is likely to get? 4) Is there a sexism case to be made here? Are we losing a whole group of potential maintainers by way of Torvalds’s behavior? Is he excluding women? 5) Are there forked kernels running some code that shows that Torvalds is obstinate? Unwilling to admit a mistake? Or do the best ideas and implementations get mainstreamed?

    You point out that Linux is now a cooperative effort of commercial ventures and governments with $10s of millions at stake, at least. Is it a bad thing that a system with very few incentives or disciplinary measures have an environment where bad mistakes result in a public shaming?

  34. Oh, look what someone sent to my work email address. A threat that my job at Intel might be in jeopardy.

    1. It is wrong and it is abusive. And definitely you don’t deserve it.

      But congratulations, now you just managed to turn Linux, Linus, LKML into a feminist-debate stage.

      I hope you are feeling really proud about yourself 🙁

        1. No, but it’s a good first step. Kindly don’t accede to the agenda of those who want to make ‘feminism’ a dirty word, please.

      1. I may be blind, but I ain’t seen anything related to gender anywhere. Or at least not from anyone with a seemingly positive IQ, i.e. not a troll. So, I don’t really understand your point.

    2. You’re nothing like Adria Richards, what you’re saying has actual merit. You’re not whining about a stupid joke, but making a valid point about something that is an issue for more people than just yourself.

      Keep it up. We need people like you, not people like Adria.

      1. The Adria-bashing is totally unnecessary here, and really unfortunate to see. Anybody who calls for better professional behavior within our communities deserves support and not vitriol.

    1. This comes from a Finn, but: management by perkele works. Point of it is not to insult, but to maintain clear focus on the task, goals and chain of command. At the same time, it is a leadership strategy that works only when people believe in merits of the leader. In Finnish engineering environment, that is always a question of technical merit – never a question of verbal exposition. A leader that’s abusive is not a leader for long, but on the other hand: you have to break the eggs if you want to get your omelette actually done, instead of theorizing how it would be performed with least strain on fragile participants.

      One thing that many commentators here forget is that LKML is truly international platform for development. Neither small niceties nor hints of dissatisfaction translate well over cultural boundaries, even if English would be the shared language. Trust me, communication between Finns, Poles and the French can be hard enough, even on a video conference. What when there are thousands of people form several continents on a single mailing list, most of whom you have never met? There has to be a culture of quality and style of leadership that doesn’t leave margin for misinterpretation. Otherwise quality can’t be maintained – and then nobody is happy. It’s better to have a grumpy captain on the ship than one that hides in his cabin, and under above circumstances, it’s better him to be direct and unambiguous.

      LKML is a platform with a goal: a goal of developing one of most widely used operating system kernels in the planet, in a way it stays that way. By definition it can’t accept hobbyist quality of commits if that goal is to be maintained. In my opinion, it’s way more important to enforce this culture than be nice to *anyone* just as a social factor.

      If some of you are distressed by this situation, there are millions of other activities in the world to participate. I don’t do bodybuilding, I’m not a chess grandmaster, and I don’t sing at the opera. Also, I don’t think I have a position to comment on their attitude towards reaching set goals. I definitely enjoy more an opera singer who has put effort on learning her skill to perfection rather than one trained by someone ashamed to “upset” his student by telling the harsh facts in on-your-face attitude. I most definitely respect the teacher that has achieved bringing perfection on stage; it tends to demand making unpleasant truths clear to the student.

      BTW: In those circles where I work, the best way to lose gained respect is to get upset of something not directly related to professional goals at your own position. Sticking to doing good work is thousand times better way to gain respect and reputation – at least the good kind.

  35. Good for you, Sarah. Rationality and civility do not interfere with productivity, and if anything should enhance productivity. It’s just much easier to be a boor and a bully. No excuse for such obviously smart people to be so lazy about applying their brainpower to interpersonal skills.

  36. Linus’s way of communicating has worked for decades, it is direct but not abusive although you seem to disagree. I don’t see this a big issue nor do I see it as an unprofessional way of communicating, it’s just a different one.

    1. Being rude on a mailing list is one thing, posting e-mails someone sent to you in private might be more rude.

      1. Perhaps you could help me understand why semi-pseudonymous strangers should feel that they can send obscene, meritless abuse to anybody in the world with the expectation that their abuse will be kept perfectly private?

        Because that sounds to me like a defense of the culture of abuse that she’s trying to change.

      2. Ah yes, that good ‘ol playground favorite “nothing is worse than TATTLING!”

        The idea that unsolicited haranguing deserves to be treated like it was under some sort of legal privilege is abuse culture.

      3. Where does sending pseudo-anonymous abusive hate mail land in your complex scale of morality??

      4. pseudonymous strangers who use email for verbal slaps in response to a public debate have no basis to expect privacy.

    2. I understand that posting screenshots of coarse hate mails is relevant to the discussion and may also help ease off the unpleasantness of receiving them.

      May I suggest, anyway, that you take care to mask the email addresses of the offenders?

      Showing them doesn’t add much to the debate and just gives some other random immature pothead the chance to retaliate with counter hate mails to the offenders.
      Which I am pretty sure is not an outcome that you’d support.

  37. I don’t have much to say other than: thank you for fighting the good fight and taking a stand here. The prestiege of the Linux kernel leads it to often be looked at as a model example of how to run a community… it’s not pleasant to see the network effect of the rest of free software taking on that poisonous attitude.

    It’s frustrating to see the kind of backlash you’re facing for doing the right thing… all I can say is, thanks for doing it, and you have my verbal support, even if I don’t have much more to give.

  38. There’s never any excuse for verbal abuse in any open source project. The LKML community is the outlier amongst open source projects in this respect, the majority of which will not tolerate bad behaviour. It is well overdue for LKML to hold itself to a higher standard for the greater good of all its community participants.

  39. I read through all the comments here and some of your previous post too, but all I can feel is sorry for you. You are looking at LKML with your blinders on. But LKML is not your project to expect your “corporate” rules. You are definitely taking your feminity coupled with corporate bullshit to LKML and make it work for you. But unfortunately LKML is not a social media for your antics. If this statements are so rough for you to take, you can very well moderate this comment.

    There is always some disaggrement with any kind of mangement in this world. If Linus has made sure he acts this way to ensure quality and if he was sissy, we probably won’t be talking about this whole project itself.

    The objective question is, will it help to code or work better, if he tones down a bit?

    I don’t think so. Programmers becomes lazy and the code quality nose-dives are very much expected. Was steve jobs ever nice to every one?

    I know a bunch of people is asking you to switch jobs if you don’t want to deal with Linus. But the reality is there will be a Linus everywhere for you in this world and your career will be riddled with controversies acomplishing nothing.

    If you sincerely believe you can make a difference in your life, better make contributions to any project so that you have control over the project. If your expectations are going to be this way, even with your own project you will face troubles with flame wars.

    I don’t think Linus suffered mentally a bit with this controversy, but I can tell you did.
    I did read his reply even though I was not entirely happy with his comment for this blog but I already knew what he meant.

    1. ” But unfortunately LKML is not a social media for your antics.”

      Expecting simple decency is not an ‘antic’. Your statement is an example of a person who lacks empathy.

    2. There’s a difference between being uncompromising about quality and being abusive. I clearly don’t have the history behind these interactions, but the messages linked in the article are downright abusive. All that was necessary in that message was “your patch breaks user functionality, and therefore can’t be accepted. Further, your use of (I forget which exit code) was erroneous because of x y z.”.

      Attacking people’s self esteems is a perfect way to get people to disrespect you (as well they should), and to be less productive and creative.

  40. 1. Unless Linus is the one paying that 80% for their work he doesn’t _have_ to change. If you have a problem with the environment you work in file a complaint with Intel HR.

    2. For someone who seems to care so much about being a stable tree maintainer you sure seem to have chosen the wrong way to accomplish it. IMO

    Good luck on your quest.

    1. 1) This is a conflation of two different sentiments which happen to be expressIible using the same English sentence.

      p: “Linus does not ‘have’ to change” (Linus cannot be forced to change)
      q: “Linus does not ‘have’ to change” (Linus’ behavior is already appropriate)

      It IS true that because there is no clear legal chain of responsibility in play, Linus is not *legally obligated* to change. This is COMPLETELY ORTHOGONAL to the untrue statement that Linus is not *morally* obligated to stop abusing people who are in effect, if not law, answerable to him in their professional career.

      2) Oh, goody. The argument from retaliation.

      1. 1.
        p: Glad we agree on the point I meant to make.

        q: Does Sarah’s opinion decide what abuse is in this case? I don’t see anyone in her examples coming forward to say they felt abused. Why does her opinion of Linus mean more then any other dev?

        2. If 1.p is true then no, it’s not the argument from retaliation. It’s common sense. 🙂

      1. Pardon me but I don’t think that this attempt at pillorying is either appropriate or productive.

    1. Heh. I like the tag line he’s using.

      Thank you for standing up against that nonsense, and godspeed. (I’m thankfully not doing any kernel work, so I’m not tempted to get involved. Just resharing this on G+ attracted more than enough noise. I’m good for now 😉

    2. That’s Lame. First, all three of those things have involved small-minded girls without the faintest sense of humor; some of whom I’ve known for 30 years. Second, I don’t think you’re small-minded; just a little naive about how easy it is to get people to change their behavior when they have no motivation to.

  41. I have to side with Linus on this one.

    His nonsense intolerant style of management elevated the kernel from a side project to the world class software it is today. I myself lay it down pretty hard when a top developer produces code that is not of the expected quality level or when a FLOSS project is attacked. Hacking is not just about writing code, is also about changing management paradigms and styles, and Linus proved this beyond a doubt, managing one of the bigger (if not the biggest) software development project in history.

    1. This is exactly the problem: too many people confuse “nonsense-intolerant” with “abusive.” Linus’s insistence on high-quality contributions is exactly on target, but his approach to dealing with mistakes (or, worse, differences in philosophy) is not.

      It is possible to take a no-nonsense approach to leading a software development project and succeed without resorting to petty insults and invective. It really is. All it takes is a little emotional self-discipline.

  42. I stumbled upon this stuff. I had not checked my privileges, nor accomplished anything of valuable (atm) in the IT world. I will nonetheless steal some speech rights on this comments dashboard to state that the master Linux kernel is built on a filthy bunch of blood, verbal violence and bashing.
    Please make your own fork with a community committed to politeness, if you think that Torvalds is too rude. I like the guy as he is – grumpy, short-tempered, yelling to people and red hat drones. I don’t think that politeness is for pussies, but I support unpoliteness as a weapon of the free source against the open source, the closed source and the companies.


  43. Bravo to Sarah.

    As a friend said to me, Linus doesn’t care about his abusive attitude, and why should he?

    My answer: because linux would improve so much faster if he decided to act like a decent, reasonable adult. His childish, asshole attitude screens out 90% of people willing to participate in kernel development.

    What’s ironic is that Linus and his disciples truly believe that a deliberate culture of verbal abuse keeps things “efficient” and “on track” and “is just humor”. Amazing. Maybe they’re just all secretly afraid of having too many contributors and losing control.

    1. 90% of the people willing to participate in kernel development shouldn’t, including many who get paid to. (not a shot at Sarah.) No doubt Linus’ harshness loses a few who should and still lets in a few who shouldn’t, but the numbers are in the noise.

      I don’t think Linus cares about or has commented on “efficient” or “on track” in the context of his behavior. I know that he doesn’t fear losing control.

      1. Couldn’t (presuming your claim is accurate) 90% of people willing to participate in kernel development not being good fits be a result of “willingness to tolerate a toxic environment” being a terrible filter?

        I have no doubt that this approach does filter out some unqualified developers – but it’s filtering out a whole lot of qualified developers, too. Which honestly makes sense – I see zero correlation between development skill and willingness to tolerate abuse.

  44. Oy vey you poor goyi…girl. You need to teach these sexist boys that being racist is wrong. Think of the wonderful things that womyn have done in the IT field. Clearly Linus is a rape apologist who fosters negative views of minorities.

    Keep up the good work and check his privilege.

    1. I think you are a brand new tea. With coconuts. And cranberries. Little sour cranberries. Sugar-coated cranberries of death from hell.
      You should not hate the old tea.

  45. Perhaps the reason that 80% of Linux kernel developers are salaried employees and not volunteers is you would have to be a masochist to be abused for free.

  46. I’m perplexed as to why many commenters can’t seem to grasp the difference between directness and abuse. We can agree that directness is OK and even valuable to help move a project forward. But that doesn’t preclude us from agreeing that abuse is not OK.

    Cursing and insulting your professional colleagues is abusive, by almost any standard. You can argue against that all you want, but as far as the rest of the professional world goes (and Linux maintenance is a professional activity nowadays), you’d be in the minority.

    1. My guess as to the reason: it’s not so much “can’t grasp” as “refuse to grasp”.

      Upton Sinclair said, “It is difficult to get a man to understand something when his salary depends upon his not understanding it!” That’s definitely true about a paycheck, and even more true about avoiding the pain of admitting they’ve spent years being a jackass because they liked being a jackass, or the work of figuring out how to interact in ways that are both respectful and effective.

    2. this exactly. The idea that you somehow need to insult someone personally to let them know their work is bad is just bizarre to me.

      And the idea that insulting people is some “cultural” thing is even more bizarre. I’m German, we’re supposedly rather direct, and sure, German’s find the idea of taboo four letter words a bit silly, but no, nothing in German culture makes it OK to berate other people. I’ve interacted with enough Scandinavians to not that they don’t find insulting people to be a-OK, either.

  47. I’ve thought that Linus’ verbal excoriations of people were over the line for a long time. I don’t work on the Linux kernel but have been an observer for a while. He is far over any acceptable line.

    You are right, Sarah, in speaking up about this.

  48. This is why women are viewed as pathetic jokes, especially in the tech world – because you’re weak and ineffectual, insufferable pansies who expect the world to cater and accommodate your thin skin and easily offended hyper-sensibilities. Grow the fuck up bitch. It’s real cute how you’ve tried to paint yourself as some gallant Joan of Arc, crusading against “muh bigotry” and “muh intolerance.” You’re a feminist joke, one in a very long line.

    Quit trying to be a damsel in distress, pull the victim card out of your ass, and do your fucking job.

    1. The whole “I’m offended” meme does seem to mostly show up amongst women more, or at least makes the news more often when it’s a women, but it’s obviously a small percentage of a small percentage.

      I seem to remember some similar complaints during the whole “fuck nvidia” incident, but they never received this kind of attention. I think places like ars may be selecting for “women in tech” stories like this, making it seem like women are more likely to complain about being offended.

    2. The reason why I support people like Sarah Sharp is to make the tech world safe for people like my daughter.

      Do you have a daughter? Imagine for a moment someone treating your daughter the way you’re treating Ms Sharp.

      Go ahead, I’ll wait.

        1. No it would not be OK. The original LKML thread is about basic civility towards humans. This is NOT a gendered issue. Men don’t want to face verbal abuse either.

    3. Did you actually read Sarah’s article, or did you assume that because she has ovaries that her objections to Linus using obscenities and insults with people is a feminist issue? Take your woman-hate elsewhere, and take own advice and grow the fuck up.

      Suppose Linus was instead a woman named Lina, and Sarah was a man named Steve. Would “Lina’s” words be considered rude, and would it make you decide “What a bitch!”? Would you say “Right on, Steve, this is unprofessional”? I bet you would.

      Stop embarrassing the sane men here.

    4. Let me correct that for you : that’s why YOU view women as pathetic jokes.
      Because personally, that’s just a case of someone bringing out an old, already much debated subject. The gender of the person bringing that old subject out again is totally irrelevant, since it’s been brought before a lot of time by people from other genders.

      Hence you are saying bullshit, you are an idiot, you should feel bad about yourself, become a depressive suicidal alcoholic and die sad and alone in your own piss :P.

  49. Hear, hear. I really hope the day comes soon when men in tech will begin, collectively, acting our fucking ages.

  50. I’m not certain how much of the misogyny is actually genuinely misogyny, and how much is just people trying to be as insulting as possible. I suspect that if you were in the situation without your gender apparent, or were male, you’d get a similar percentage of insulting comments. They probably wouldn’t use gendered insults like they are now, but they’d still be asshats.

  51. Thanks for doing this, Sarah. I appreciate it.

    I also note with amusement that all the jackass respondents here are anonymous or pseudonymous. Apparently their dislike of “weakness” doesn’t extend as far as being strong enough to own their words.

  52. I support you Sarah. By sheer coincidence I found your blog on the very same day that a friend sent me a link to this:

    Whilst I may actually agree with Linus’ sentiment in wanting (demanding, even) only the best, most perfect code to go into the Linux kernel, I think communications should certainly be kept on a professional level. One can be firm without being rude or obnoxious, but sadly Linus seems to revel in making a conscious choice for the latter most of the time. In fact, I think he’s something of a loose cannon and might be doing more harm than good for the advancement of Linux. I work as a Systems Specialist in a corporate environment which is primarily a Wintel shop, but through considerable efforts on my part over a number of years, I have slowly, ever so slowly, convinced the powers that be of the merits of Linux, and I am now happily putting in Linux servers here and there. Historically the prejudiced attitude of those in the ivory towers was that Linux was just a two-bit hobby operating system “cobbled together in someone’s bedroom”. Thankfully they don’t read the LKML, but I can just imagine that if they did (or, for example, if they were to see Linus unleashing his venom on nVidia:, all my hard work would be undone in an instant and they would see their criticisms and prejudices about Linux as fully validated.

    I don’t view Linus as a good ambassador for Linux. He makes me wonder why FreeBSD isn’t more popular:

    Oh, and I know you didn’t want to make this into a gender debate, but you seem to have attracted a good deal of misogynistic condescension here (I’m sure you’re used to it in your line of work) so I’d like to say that I once worked with a female Unix sysadmin colleague whose capability and intellect could run rings around the macho so-called techies in the department. I know you are quite capable of standing on your own two feet and fighting your own battles, but I thought this was apt to mention.

    Anyway, it’s refreshing to read your blog post about this. Keep fighting the good fight!

  53. “My culture is cursing”

    unacceptable from linus. linux community should sack him. 100% totally unacceptable. sooner or later he’ll run into a strong kernel maintaining dude who won’t have much tolerance for his idiotic behavior.

  54. The ridiculousness of this “scandal” is now officially beyond any resonable measure.


    Based on what standard are you juding the morality of behaviour of people on the LKML?

    You see, if your opinion is that someone’s acting immoral, then that’s absolutely fine, but if you want to actually make a point, get it across and make them act differently, you have to back it up by a reference to a moral standard that is held valid and respected by both parties.

    I’m not saying you’re wrong, but you can’t just impose your own personal moral standards on other people, because people perceive morality differently. Instead it would be nice if you could specify according to what standard, codex or even law is the way of communication on the LKML wrong.


    Even being right can go horribly wrong. I think this whole thing is a perfect example.

    I’ve encountred aggression and violence numerous times, either IRL or online, verbal or physical… but one thing was always the same: More violence did not help.

    Sure, you can “stop being nice” and throw around phrases like “this is not *fucking* cool” and “the shit I have to put up with” and similar beautiful usages of the English language. Go ahead. Try not being nice anymore, see what happenes.

    Or lemme spare you the trouble: It will not help. It will only make the matters worse. People *cannot be forced* to be nice to one another.

    The only way to make a point with people who are violent or aggressive in some way is by not losing your cool under absolutely no circumstances, by being consistent and assertive and by leading by example.

    You say you’re “standing up against verbal abuse”. What this ultimately leads to, at least the way you put this “standing up” thing in practice so far, is more verbal violence (either explicit or covert), more flamewar, more hatred and no improvement whatsoever.

    What are you really after? An improvement? Or some kind of revenge? I hope it’s the former.

  55. So many abusers defending abusive behavior on this thread. It’s shameful, despicable behavior, and while it’s expected that they’ll double down on it when they get called out for it, it’s about time that we shine some light on this issue.

    The only people that an abuse-intolerant culture hurts are abusers. I’m perfectly okay with that, and thanks for helping us move in that direction.

  56. Thank you, Sarah, for standing up for a basic standard of courtesy. I agree with what Jonathan has said: “Cursing and insulting your professional colleagues is abusive, by almost any standard.”

    There’s no need to sugarcoat things, but removing obscenities and insults from Linus’ emails would make his criticism come across as harsh yet constructive, as opposed to rude and obnoxious commentary it is. He could flat out tell people “This is bad coding, I’m honestly surprised that an experienced coder like you came up with something so poorly thought out, you need to fix this” and that would be direct and reasonable.

    To all the men bringing Sarah’s gender into this or making this about feminism, stop embarrassing all of the rational people on here with XY chromosomes. YOU are being oversensitive by assuming that valid criticism of Linus’ public behaviour is a women’s issue rather than an everyone issue. Go whine about feminism elsewhere. Or better yet, realize that the reason feminism is still a thing is because of unenlightened fools who make broad general assumptions about half the world’s population.

  57. I’m sorry. The two examples you posted are clearly because someone is fucking up. You don’t get them to stop fucking up by being nice. Simple as that. There is always the option of just toughening up. Your version of vulgarity, which is at worst what we’re talking about here, is not the same as mine. Clearly these are recurring issues, and for someone like Linus to have to chime in on it I would imagine it’s something rather recurring. Hell, he even says he has to do this every release window, how many times do you think he’s said the same thing over and over and over? Also, it appears both of those people were in the wrong. I suppose revoking their commit access would be better? All I can say is growing up is hard to do. As well this has nothing to do with women v men, I didn’t see any sexist comments in there. And even if I did… who the fuck are you? You don’t police the mailing list, pretty sure fucking Linus does, you know… the guy whose shit you’re working on…

    1. Actually, in the second example, Linus was _wrong_. He took a sentence out of context and misinterpreted what was going on. The regression was the result of a fix which enabled code that should have been running but wasn’t — reverting it would have just broken the entire functionality.

      Either way, being right is no excuse for being an asshole.

  58. Right is right, even if everyone is against it, and wrong is wrong, even if everyone is for it.”
    — William Penn, British statesman and philosopher

    And you are right!

    1. It’s hilarious how many anonymous cowards think anyone wants to hear their opinion. You’re the one that came here to peep out a barely coherent sentence as if it was relevant — who’s the loser now?

      1. The trolls don’t even understand that they’re only getting their messages through to prove Sarah’s point.

    2. Wow, that’s great that you feel comfortable enough to post this from YOUR WORK COMPUTER. I wonder what HP management would think of this…

  59. LKML posts are google searchable and what Linus is doing is public abuse. The language in which he abuses is despicable in many countries including mine – India. If he wants to develop the culture of abuse he can turn LKML into a closed group. Not being professional doesn’t mean you have to be a naked (or well, in a bathrobe) beast roaring around (BTW, no one even wanted to know whether he was in a bathrobe. He just added that to emphasize his points). He should learn to control the way he writes. If he has some pet developers who are OK with the abuse Linus can call them up over the phone and abuse them. But not in public. The world is a much bigger and important place than LKML and many people are trying to keep it clean for the future. Do NOT pollute it with your cheap, gross and vulgar language.

  60. I too have to question your motives since only in the greatest stretch of the imagination is this latest “gem” from Linus
    “advocating for physical intimidation and violence”.

    Sure Linus is abbrassive (and there are many examples to back that up). My question to you is what do you hope to achieve by changing Linus? If the Linux Kernel as a project was failing and you were somehow able to demonstrate that it was a result of Linus’ leadership style, then I would be more inclined to agree with you. The Linux Kernel as a project is however not failing, and one might easily conclude that Linus’ management style, although certainly not for everyone, seems to be working.

    If you have a particular issue with his “style” (lets continue to call it that for argument’s sake), then why not take it up with Linus? If you’re offended than tell Linus that. Everyone on LKML has their own voice, I’m not sure you need to step up on your soapbox to defend the rights of the downtrodden just yet.


    1. I think you haven’t been following the LKML thread. I DID speak up to Linus about his management style.

    2. You’ve conflated two different things, the success of the overall Linux project and some poor management choices by Linus, just because Linux is successful doesn’t mean that Linus is perfect. If success worked that way we’d all be working for Microsoft.

      One doesn’t need to be abrasive and insulting to be direct when giving negative feedback, they just need to be clear about what is wrong and what the consequences are, reverting a patch for example. It’s not that hard. Most people who are adults and work with other people do this successfully every day. In my work environment we can have a frank exchange of ideas, sometimes even heated with raised voices, but never with insults or hard feelings.

      As you say, everyone on LKML has a voice, including Linus and Sarah so I don’t understand your subsequent statement recommending that Sarah chose not to use hers. Why not?

      I am somewhat saddened by the number of people here who are presumably out of their teens who refuse so obnoxiously when asked to just simply behave like mature, responsible, adults.

      1. Sorry, should have been more clear…. Everyone on LKML has their “own” voice. (emphasis on “own”).

        Or to put it another way “it’s admirable to stand up for those that can’t do so themselves, it’s arrogant to defend individuals that are perfectly capable of defending themselves.”

        also … unicorns; mature, responsible, adults; common sense. I’m not sure they exist.

  61. Hi Sarah.

    I stumbled upon this entire discussion in hacker news this morning and found this post via proggit at night.

    I’m certainly sympathetic to the obviously misogynistic comments

  62. I have no idea why gender is entering this dialog. From my perspective, gender is irrelevant to Mr. Torvalds’ behavior, and citing it clouds the issue at hand.

    There are many points worth considering:

    – When in Rome, do as the Romans do. If you don’t like it, you can revolt and attempt to spawn an uprising of your peers. Do you think the people who matter will aim to join and please you, or passively side with the benevolent director of the project? (hint: the later)

    – It’s one thing to publicly correct a peer. Linus is _not_ your peer, and you are not giving him the respect he deserves. Your contribution to the project is not close to his. You’re speaking to him as if you are on the same rung of the ladder. Reflect. You _arent_.

    – Telling Linus you won’t tolerate his shit is ridiculous. It’s _his project_. If you don’t want to contribute code to the project anymore, don’t. Someone else _will_ take your place. You are not a unique snowflake that is irreplaceable. On the other hand, if he chooses to no longer tolerate _you_, he has the power to ignore you and your contributions going forward.

    – He isn’t going to change. Does he sound like a rage filled buffoon spouting off tirades: yes. Do most people process it in the context that that is who Linus is: yes. Will he change at his age: absolutely not. Is it an amiable goal to try: sure.

    Will it _ultimately cause you more grief than gain_: yes.

    1. Gender is only in here in so far as mysogynists keep brining it up — “little girl” and “get back in the kitchen”.

  63. My condolences for all the terrible comments that you must be getting (some of which I have seen).

    Could anyone here explain to me why swearing in a mailing list is unacceptable, or even calling someone a moron or other names. I understand this might be unacceptable behaviour in the US (or perhaps on in office environments in the US). However after thinking about things for a while now I see no reason to see why a culture could be different in that people communicated with brutally honesty.

    I read your blog from time to time, and I’m going to guess you’re American. What I don’t understand is why a mailing list should set standards that fit the views of professionalism that Americans expect and not the professional behaviour of some other culture that perhaps is more direct.

    Calling someone a moron after they make a terrible mistake is something that refers to a person at that time. I’ve got friends who are exceptionally intelligent (well maybe not as intelligent as Sarah and other kernel developers) but I’d call them an idiot if they suddenly came out and said that 2+2 is 22. He’ll I think I’m a moron when I make mistakes that are beneath me. It’s not a remark on some immutable aspect of a person’s character. it’s a remark on a specific incident involving them.

    As someone who doesn’t work on the kernel, and reads no more than 80 lkml messages a year, I’m definitely missing something, so I’d appreciate anyone who could clarify the issue for me.

    1. You would call your friend who said 2+2 = 22 an idiot to their face. Calling them a moron on a public forum, and verbally dressing them down to assuage your own frustration is not civil.

      If your small child was with you when you dressed your friend down, you might choose not to cuss your friend out. Or you might turn to your child and say, “Cover your ears”, or “Don’t ever repeat what I’m about to say.” The problem is that people see Linus’ verbal abuse, and they emulate it. Newcomers never understand that Linus is dressing down people he trusts, because they violated his trust by breaking is maintainer rules. Newcomers see these rants and think it’s OK to verbally abuse anyone in the community. That makes some people leave the community, or never join. Other newcomers will develop a “thick skin” to avoid feeling verbal abuse, and in the process, they will learn to be verbal abusers themselves.

      1. Are you saying that the issue is with people attempting to imitate Linus’s communication style without an understanding of the cases in which this communication style is acceptable?

        Or are you saying that Linus’s communication style in and of itself is unacceptable (irrespective of what behavior that may cause in the wider community)?

        1. I’m saying that people attempt Linus’ flame-worthy style in cases where it’s not appropriate to flame someone. I also think (unrelated to newcomers modeling their behavior after Linus) that kernel developers should focus on criticizing code, not people. It’s fine to call code crap. It’s not fine to call a person crap.

  64. I just want to voice support for Sarah’s message: the best way to get good work out of people isn’t to berate them, and the best way to encourage new members isn’t to publicly berate the ones you currently have.

    Sarah, your calmness and clarity of message is really refreshing. Thank you for taking a stand.

  65. Most of the responses here are appalling. I now know what codebase I’m staying away from at all costs. My suggestion to you would be to go work for an organization that treats its contributors with respect.

  66. You’re being irresponsible. If you’re such feminist, you should know that being woman and feminist is not about just having rights, it’s about having proper responsibilities. If you can’t fullfill your responsibilities as developer, please, go. Run away from this scope, because you will bring great headache to yourself and LKML community, and maybe even to Intel.

    Tl;dr: you should DEAL WITH IT.

    1. This is not about feminism, it’s about abuse. Since you seem to enjoy using abusive tactics, I’m not surprised you’re trying to make it about something else.

  67. Anyway, good riddance. As we say, “when a woman leaves the cart, the mare feels easier”.

  68. This is all strawman.

    A professional person, paid, and demanding from others ‘professionalism’ – delivered code that was not just substandard, but totally unfit for its purpose. This should never have happened. Period. Never.
    Now, this has been turned into some left field behaviousrist sudo scientific argument based around verbal abuse. Its a cover and a smoke screen and its bullcrap.
    You’ve made much of the fact that you are paid to do this, and that its all professional. In many companies shipping this level of fail would garner a tough repost’e. The patter cake patter cake petting someone and telling them everything is ok wahh wahh bullcrap isn’t real world.
    This wasn’t about emotions, or verbal abuse, or violence. All of that was a side effect of the origins of the whole subject.

    Unprofessional code was submitted. It wasn’t fit, and should never have been put through, and you’ve now been told by the man who runs linux some home truth about it.

    You should go away and fix your stuff. And while you are at it, you should do your work, the work that you are paid for, and not spend time jerking aroud playing petty politics and making up cases about behaviour, and verbal abuse and whining about how tough life is.

    Its not. You have a wicked gig working at Intel and working with some top level minds within Linux.

    I’d have more credence for your case if it had not been raised in a response to you screwing up. You did not write this to try to fix things – you wrote this because you screwed up, were unprofessional, and got told straight up.

    You are not a junior now. School is over. In the real world, people get told the real deal.

    Maybe instead of you holding a summit at KS where you bring your sudo bullshit, someone should sit down with you and make sure you understand how to ship your code professionally. That seems to be much better use of your time and theirs.

    This very public stomping of your little feet because you got told straight up is all very emotional and careing and sharing and sweet. Its also complete bull.

    Its now become the cover for your screw up and is bigger news than your error. Right? Thats how it works? No. Smokescreening doesn’t always work.

    1. You fail at reading mailing list threads. I did not screw up. Linus yelled at Rafael and Mauro and the x86 maintainers, who rolled over and took his abuse. Linus has never personally yelled at me, which is why I used this thread to tell him I would not take it if he did.

      And yes, it’s time to go back to doing my job. I love my job at Intel, I love being a kernel hacker, but I hate what the community does to each other. It’s time to move on, and save the discussion for Kernel Summit.

  69. I’ve used Linux since my first year of university, and in the spirit of open source, I’ve always wanted to get involved with kernel development. But this reputation of verbal abuse on the mailing list, is definitely why I decided not to get involved.

    Obviously, there’s a steep learning curve when first getting involved with any development project. Thus, as a potential newcomer, I feared that I would be likely to make mistakes, and that I would be ripped to shreds for those mistakes.

    Because I really like to contribute back to the open source software that I use, I’ve even considered giving up Linux completely and using a different open source OS, such as FreeBSD, that I would feel more confident getting involved with.

    Unfortunately, I feel that this reputation of an aggressive working environment is likely to harm Linux kernel development in the future. For potential newcomers (like me), I think many will be too intimidated to initially get involved. For existing contributors (for whom kernel development may be a hobby, in addition to their regular 9-5 job), I think they may start to feel that they would prefer to avoid the additional stress and hassle.

    1. What’s not clear from the threads is that Linux kernel developers (especially Greg) are generally very friendly towards newcomers, especially students. It’s just when you start working for a company, or have been in the community for a while, that they think you can put up the verbal abuse. I would really recommend you do try contributing as a student. There’s a good tutorial on how to get started cleaning up staging drivers. If you’re brace, you can start there. Don’t let this argument between friends on LKML change how you use and contribute to Linux. If you do, they win. 🙁

      1. Thank you for your advice Sarah, I really appreciate it. And thank you for confronting this issue; I can only imagine that a less aggressive working environment will result in happier, and better motivated, developers and thus an even higher quality of work.

  70. It’s worth remembering that the production of software of this type is a social process. It is possible to make utilitarian arguments along the lines of “it doesn’t matter how people treat each other so long as the end result is working software” but that’s a recipe for short term success and longer term dissolution. It’s not only important to have users but also to be capable of sustaining collaboration.

    I’m not a kernel developer myself, but on a personal level I have a ridiculously low nonsense threshold, so if I was part of a community in which language such as that in the examples began to be used routinely then I would simply leave that community without any further discussion. Communities which aren’t able to self-regulate adequately will lose good developers and eventually fall into obscurity.

  71. 1) Misogynistic and sexist writing on mailing lists is both morally wrong and counter-productive to any reasonable project. No excuses.

    2) You need to be clearer (i.e. explicit, on this blog post!) that you’re not implicating Linus in said activities, unless you really mean to.

    3) It’s totally legitimate to say that you don’t like Linus’ attitude. Good on you for speaking up,

    4) “Abuse” is a wide-reaching word that needs adjectives to characterise it’s extent and severity. Snapping at someone when you’re tired and systematically breaking someone with targeted insult are both “verbal abuse”. There’s a term for abusing generic words like that but I can’t remember it….

    5) Threats of physical violence? Really? Sorry, that sort of overreaction is what loses you the support of people who would otherwise be totally with you.

    1. 5) I brought up the “threats of physical violence” because (although it was a joke on their part) they were trying to use it as an example of how Greg KH should be more initimating. They jumped the shark with that joke, and they know it. But the joke is systematic of the larger cultural push: to be the jerk maintainer, that doesn’t care about developers feelings, and tears people apart to prove a point or spread the word about a behavior they don’t like. You didn’t read the whole thread, but that particular part of the argument got dropped because everyone recognized the relevant issue was the verbal, not physical, abuse.

      1. Nobody’s care about your feelings, woman. You are paid professional which should do job you’re best at, not cry and whine like a little girl. Get over it.

      2. Fair enough. Indeed I didn’t read the whole thread, I’m glad that point got dropped.

        Good luck 🙂

        P.S. This post made me realise how pleasant the corners of the internet I visit are most of the time. The level of misogynistic trolling in these comments (and I do mean misogynistic, for any dictionary fascists out there) is quite frankly appalling.

  72. Well done Sarah! You are right. These emails are from the leader of a community to members in the community – members who have worked hard to become part of the community and recognized by it. That kind of language and embarrassment in front of your peers must really hurt. If I was the recipient I would be crushed, demoralized and probably even apprehensive about continued participation – alienating devs diminishing the promise of open source.

    You go girl!

    An appreciative end user.

  73. Sarah, I saw your posts to LKML linked on reddit. I’m very happy to see this, and to see so much positive response. There is almost a kind of frat-boy hazing culture in computer programming (esp. free software), and that has got to go.

    The real problem is not so much Linus’s behavior in itself, but the general attitude of tolerance of such behavior — as if the anointed hackers of the in-group are so intellectually superior that they ought to live under different rules from the rest of society. The problem is not at all limited to LKML, but Linus’s famous rants — and their defenses and defenders — are emblematic of a dysfunction permeating the whole culture. Linus is not just hurting people’s feelings, but perpetuating this dysfunction (which he certainly did not create in the first place, of course), causing others to believe it is acceptable behave in the same way. Thus, although I am not involved in kernel development, I still feel involved in this as a free software developer.

    There is a long-established “cycle of abuse,” which I think actually derives from competitive culture in CS programs in elite universities, but in any case which must be broken. And all we need to do to break it is just what you have done. Just let people know that it is not OK to behave that way, every time it happens.

    I have already seen, over the last 10 years, a significant positive shift away from the elitist hacker mentality. We are all better off as a result of it. Our communities are healthier, and we are held in better regard by outsiders.

    I just want to thank you for helping push this forward.

  74. Mostly, the two examples show developers reminding others that their behavior and opinions go starkly against established kernel policies, such as regressions and breaking user space, and not accepting code without review.

    What I will agree with you on is that there is a lot of obnoxious not-getting-to-the-point-because-angry-ranting going on. That’s where the unprofessional and dumb stuff seem to come from.

  75. Isaac Newton, Pythagoras, Benjamin Franklin, Thomas Edison… great geniuses who were also total and utter arseholes.

    Let’s add Linus Torvalds to the list.

    (Somewhere near the bottom though. Compared to the law of universal gravitation, Linux isn’t that good. But compared to electrocuting animals for fun, firing off a few wankerish emails isn’t that bad either.)

    1. Seriously? I can’t even consider Dennis Ritchie part of that list, let alone MINIX thief.

  76. Linus playing the ‘minority’ card in response to you shows a complete lack of understanding. That’s not an argument, that’s “what about MY suffering?” angst. Has he *done* anything that would invite me to care about him?

    I hope something good comes of all this. Not just on LKML but the whole OSS world. Good luck, and thank you.

    (I’d join you publicly, except I know I don’t want, um, “recognition” from a public that includes Linus or many of your trolls here. I didn’t really enjoy interaction with any OSS communities when I was trying to provide useful patches, either.)

  77. This type of abuse shouldn’t be tolerated and thank you for standing up; as a paying customer to RedHat, Novell and Intel I will use the power of ‘money’ if they decided to cut you off Linux Kernel Development.

    – just another concerned citizen

  78. I really like what you’re trying to do Sarah! One comment,

    Could it be possible to find some sort of compromise?

    – Linus wants to be open and honest and allow people to FEEL the full force of his oppinions. Cursing is a big part of that and he puts forward some advantages of it
    – You want to end “Verbal abuse” which is aimed at someone.

    For example you said this “You don’t need to SHOUT, call me
    names, or tell me to SHUT THE FUCK UP!”

    Imagine if instead of “SHUT THE FUCK UP” linus torvals said “THIS IS MAKING ME FUCKING ANGRY AND YOU NEED TO STOP”

    One of them is cursing and anger directed at an individual. The other is linus merely angrily making the full extent of his emotion clear followed by a clear (but not abusive) order.

    Could a compromise like that work? Or would that still be too much?

    1. It sounds like a good compromise to me. I think both Linus and Rusty Russel agree that saying “SHUT THE FUCK UP” is not appropriate. I agree that people need to be told, publicly, that they messed up, in a way that other developers will pay attention to. The x86 email was a good example of him expressing his displeasure without making value judgments on the developers involved: I’m not sure if Linus is willing to change his communication style though.

    2. Imagine if instead of “SHUT THE FUCK UP” linus torvals said “THIS IS MAKING ME FUCKING ANGRY AND YOU NEED TO STOP”

      so, use 10 words to convey the same semantic content than 4… yeah that sound like a winner to me.

    3. “THIS IS MAKING ME FUCKING ANGRY AND YOU NEED TO STOP” would make Linus sound like a child who has a five word vocabulary.

  79. Hey, I don’t want to further any more controversy because there is plenty of that on the internet, but I will say that I applaud your stance and your effort. The internet could be this fascinating place of learning and collaboration where the greatest minds can meet from around the world and solve problems that make the world a better place… and then we’ve got people who just make the internet a continuous rant of “I’m wrong and you’re right.” And don’t forget the holier than thou of people who 7 years ago were struggling with their C… never mind you watched them through the very learning phases they’re hazing you about. Without the common decency that should be afforded to any human. Thanks Sarah! I’m tired of it too.

  80. I don’t really understand the argument here.

    First of all, “violent and threatening”? Really? Being told to “SHUT UP” from the world’s biggest nerd over the Internet is *hardly* a violent and threatening situation. Do you really feel threatened, or did someone’s feelings get hurt? Secondly, if you don’t like how the maintainer of a project maintains, go maintain your own project. Thirdly, from the examples you provided, it seems like they needed to be yapped at.

    We should probably tell our military officers to follow this philosophy too… seems like a winner.

    1. It sounds like you might not understand the argument because you don’t see abusive behavior for what it is, in part because you seem to utilize some of it. Dictating how others should feel or respond to someone else’s actions – i.e. that they should not feel threatened – is a common abusive tactic known as “gaslighting.”

  81. As an minor point, all of this is covered in the FAQ for the mailing list linked at the bottom of every thread. Not sure the debate is really about that on the list, per your last – which indicating you’d moved on to the expectations of subsystem maintainers re the #1 rule etc.

    * Be precise, clear and concise, whether asking a question or making a comment or announcing a bug, posting a patch or whatever. Post facts, avoid opinions.
    * Be nice, there is no need to be rude. Avoid expressions that may be interpreted as aggressive towards other list participants, even if the subject being treated is particularly relevant to you and/or controversial.
    * Don’t drag on with controversies. Don’t try to have the last word. You will eventually have the last word, but meanwhile you’ll have lost all your sympathy credit.
    * It’s very easy to criticize someone else’s code, but when you write something for the first time, it’s not that simple. If you find a bug, a mistake, or something that could be perfected, don’t immediately post a comment such as “This piece of code is crap, how did it get into the kernel?”. Contact the author of the code, explain the issue, and try to get the point across in a simple, humble way. Do that a few times and you will get a lot of credit as a good code debugger. Then when you write a piece of code people will pay attention to you.
    * Don’t flame beginners that ask the wrong questions. This adds noise to the list. Send them a private mail pointing them to a source of information e.g. this FAQ.

    So your points re “abuse” are covered, someone on the list mentioned Documentation stuff re rule #1, Jeff Liu’s points re mentoring new people vs writing the patch yourself as an experienced dev is definitely covered.

    Obviously it’s a bit dated – “As of 27-MAY-2002 (S1.21)” shows the age pretty nicely

    Of course YMMV on what is rude and aggressive, but if we want closure on that we’ll be here well past any summit.

  82. You are absolutely right. Linus Torvalds is a total dick. He’s acted this way from the very beginning.

    Maybe if his culture is cursing more people need to jump down his throat when he acts that way. I wonder if he has some sort of borderline autism from the way he treats people.

  83. Man, I don`t want to get into this. The problem is that Linus Torvalds is trying to manage a geographically-dispersed group of very intelligent people. These people need a kick in the ass, not a polite chat, because these folks aren’t going to listen to a polite chat. Acting as a drill sergeant is not necessarily the only way, but it has proven effective, and the risk created by changing the environment now is too significant. The core maintainers are people who are, on the whole, fine with this system. If you change the environment in which they work, they will probably lose discipline or drift away. If that happens, the void will not fill fast enough for Linux to survive. We cannot as a society afford politeness here, and frankly, your feelings, my feelings, Linus’ feelings are way less significant than the continued success of Linux.

  84. As someone who runs a reasonably popular open-source project (, I can only commend you and thank you for doing this. I’ve had this precise conversation with some colleagues over the years, where I’ve said that I explicitly make a point of trying to keep a culture in our mailing lists, help rooms and meetings, that balances directness and frank discussion with consideration. I always use the LKML as my contrasting example of what I do NOT want to do.

    There are aspects of Linus’ concern about politics that are very valid (we see that both in the corporate world and in open source mailing lists). I also believe strongly that allowing for frank disagreement and clarity is important to cut down on the noise and ambiguity.

    But I also think it’s perfectly possible to balance clarity, open disagreement, direct discussion AND civility. A civil (I’m deliberately avoiding the word “professional” for some of its corporate connotations) discussion can still be very frank and direct, and it can even tell people that they just wasted a month of their life on something that doesn’t have a chance of ever getting accepted. But it can also do it without insulting them and threatening them.

    I do find some of Linus’ rants funny on occasion, as I’m sure many of us do from the outside. But I would much rather lose that source of sporadic humor if it meant having the kernel be a more welcoming community that appeals to people because of their talents and interest in the kernel, not just because they are willing to be verbally abused and insulted.

    In summary: yes to directness, frank discussions and avoiding the bullshit that often masquerades as politeness. But most definitely NO to abusing and insulting people just because they made technical mistakes, no matter how egregious.

  85. My original problem with your post had to do with the word professionalism which allowed debate hijacking, while I preferred the term civility, a basic need for any kind of communication between humans.

    My 2nd was with the fact that so many of us knew exactly how this was going to play out culminating with a post 6months down “Was it worth it?”
    But yeah, yeah,… it was the right thing to say. If we allow less than civil behaviour in the mailing, we can definitely be so broadminded as to entertain the idea of asking these questions. It scares me less when some younger child asks “Does the king have no clothes?” every few years than those who get offended at the mere idea of asking such a question. That scares the crap out of me.
    Blind devotion, true believers.
    You are going to piss these freaks out.
    And worse of all, there are a lot of miserable people out there… yup, even in Linux land. All kind of psychopaths who psych their 5yr old with ‘no pain, no gain’ and others who see life this way: “Whenever I tried to be nice and understanding whenever someone fucked something up, they didn’t learn. I HAD to be an asshole to get results. You sometimes HAVE to do break someone down and humble them to build them back up.” We’ve all seen people like this at work, guess what? To them Linus doing it ‘his way’ is simply awe-inspiring because tahts how theyd love to be at work. Geez, you got taht Spiers whack job comparing kernel coders to military officers!!

    I would like to know how the Canonical developers (wait are the any kernel ones?) feel about this controversy since they have been made to swear allegiance to the Buntu Moral Code when they join by Saint Jono of Bacon.
    He likes to wade into topics in comment boards all over the net, dont see why he woudnt pop in here and enlightern us what his 3 books on internet group dynamics says about this.

  86. We’ve become a society that gives our kids trophies for coming 13th place. People need to learn to take criticism of all sorts, polite and not, to handle people of all sorts. This is *the* most important asset in the corporate world. The corporate world has only recently bought into the “social standards” recently, and it’s because the government has enforced it. I think everyone would agree that industry is far more efficient and successful than government (have you ever seen what government employees do all day?!). We have become a society with soft hands.

  87. I was kind of puzzled by all this hubub so I read a few of the replies above and a few threads that Sarah linked to. My comment: Obviously there’s no need for Linus to talk like that, and it’s lame, and it diminishes him, etc etc.

    Or to put it another way, if I was working for someone and they spoke to me like that, I’d definitely leave. If someone is cursing at people in a professional environment, it says more about their personality than it does about the ineptitude of the people being cursed at. (In my life/work experiences, I’ve noticed that people who belittle others have usually turned out to have overbearing or abusive fathers).

    Hopefully it’s clear where I’m coming from.

    All this said, I want to offer one observation to Sarah – despite being undeniably right about the core issue you raised (cursing at devs), other comments you’re making are missing the mark widely, and really reducing the strength of your core argument. A couple of examples: 1) Your interpretation of the email where Linus jokes that a dev should be scared of a much larger dev. There’s no way that any reasonable observer would draw the conclusion that Linus thinks that people should actually be scared of submitting to devs who are bigger. 2) Your interpretation of the swan photo as some kind of death threat – when it’s fairly clear that the poster was suggesting that it was good for Linus to act like a diva-esque swan.

    My 2 cents: Ignore the trolls, they’re not worth it. Ignore the people who just don’t get it, and ignore the people muddying the water. You know you’re right about the cursing thing, so fuck ’em. If they can’t see it, they have some issues that aren’t your problem. And most of all, just because people are dicks, don’t let it cause you stress. There are lots of dicks in the world, we can’t change that. But at least we’re still lucky enough to plentiful lives typing our ideas into the ether for fun/profit/satisfaction. Interesting times.

    be well

  88. As I see it, the “civility distinction” is the one between saying “that is a terrible idea” or “this is a bad situation” and “you are a terrible person for having that idea” or “you are a terrible person for getting us into this situation”.

    Bad ideas are inevitable (Sturgeon’s Law applies to ideas, just as it applies to everything else), as are procedural mistakes (especially when processes are manual rather than automated). Taking out our frustrations about those bad ideas and mistakes directly on the people responsible for them is usually going to be counterproductive in the long run.

    We all get angry and frustrated, but we can vent that harmlessly at the universe at large (and, in a development context, preferably offline), before knuckling back down to the task of resolving whatever it was that angered us in the first place.

    While, as Linus notes, there do come times when we have to go “for whatever reason, I cannot work with . We need to either work on completely separate things, or else others will need to act as a buffer between us.”, a conscious effort to be civil (even when we’d prefer to be anything but) can really help to keep things running smoothly, rather than jamming up the works by making people feel hurt and start getting defensive.

  89. Go Sarah – yes Linus has a right to demand high standards without dealing with political BS. He is WRONG to think it is OK to act unprofessional.

    All the dweebs on who commented that it is OK to verbally abuse to get a point across have never dealt with the consequences of saying what ever they wanted to the wrong person at the wrong time – they and Linus need a HUGE lesson in manners and politeness. If anyone thinks it is weak to be polite or that it is being like a doormat are way off-base. There is a difference between being aggressive, assertive and passive-aggressive in communication.

    I see a lot of aggressive communication online but passive-aggressive in person among super smart- techy nerd types. Assertive communication is the only acceptable communication. There are classes available on how to communicate in an assertive way that is respective but allows one to stand up for themselves without having to cuss and cut down the other person.

    Linus and and rest of the world intellectual bullies need to mouth off to a real bully in person and see where it gets them(or a Judge or a Police Officer).

    I think it is great we have Linux and Open Source but why are so many of the folks that work on the code such real-world pussies? Will get in someone’s face online but never look someone in the eye and tell them what they really think face to face man to man.

    The need to get out of their comfort zone and learn how to communicate and get some help for their asperger’s syndrome symptoms.

  90. Started my job developing in investment banking – was called various swearwords, and people shouted to get stuff done *right*. Fast-forward two decades into less toxic environments, and thankfully I can put that world beyond me.

    Civility is something worth fighting for, and regrettably not even seen as a base-standard in some social environments. Sarah, congratulations for calling out the elephant in the room on this one. Stay positive, and I hope it leads to positive change.

  91. Sarah,

    I applaud you for standing up and sharing your feeling. Telling Linus that you and others feel offended is the right thing to do.

    At the same time (and it’s not a paradox), I side entirely with Linus here. His management works. It’s not as if he had a bad character or was holding grudges. His style is efficient. When you drive a F1 car at 300kph, you have to accept that it’s not a smooth ride. So if you ask the driver to please make the ride more comfortable, you may be missing the point of the driving style entirely, which is “go fast”.

    My contributions to Linux are so miserable that I won’t talk on LKML. But I have created my own enterprise-grade kernel, and I’ve been working in kernel land for years in what you call a “professional” environment. I believe my colleagues appreciated my humor, but they also knew I could be temperamental at times. And I was only driving a school bus, not the Linux F1.

    Why is it efficient? Here is an analogy: imagine your 4-years old kid is crossing the street in front of you. Will you say “Dear kid, would you please be kind enough to return to the sidewalk”, or shout “Stooooop” out of your lungs at the risk of humiliating the poor kid in public and scarring his ego for life? Comfort comes after the problem has been fixed. And it’s not just for life threatening situations, it’s also when the kids behavior is socially unacceptable, e.g. if he bites others.

    That does not mean your fight is not a good fight. I think you will ultimately win, but only when Linux development slows down and becomes more mundane.

    Just my 2e-6.

    1. I disagree completely. Kernel developers can be blunt, direct, and work “fast” while still being respectful and civil.

      1. I could be wrong but I’m guessing that by now you’ve already hesitated about which is the better reasoning to apply to an issue that you feel so strongly about.

        You’ve been switching arguments to try to get your point accross

        (note: I’m not Ingo, just think he’s the one that expresses the particular problem better)

        That (switching arguments) is okay but I believe you should consider that by switching your focus from efficiency (professionalism) to morals (respect) you are entering a realm that you should not possibly have any hopes in (p)reaching universalisms.

        About this issue, you need not worry, there is centuries of good literature and history.

  92. It’s just ridiculous how Linus and his advocates try to justify his sociopathic behavour as a way of managing code quality in the kernel. What about his so called rants against Gnome or OpenSuse? I guess he is just so concerned about raising the overall quality of the OpenSuse distribution that he must the devs tell to “please just kill yourself now. The world will be a better place”.

  93. It’s amazing how trolls bring gender into this issue. This isn’t about gender. It’s about human beings treating each other civilly.

    1. I think the trolls are referencing the issue with Adria Richards a few months ago. Either way, this has nothing to do with that.

      My question to you, is what do you stand to gain from all this? I get it, working with Linus seems tough and god knows I’m not volunteering my time and energy and blood/sweat/tears just to be cursed at over a few lines of bullshit code. But are you expecting him to change? Are you expecting people to just stop contributing to the Linux kernel because he’s a prick to a few people?

      Besides the fact that I and a lot of other programmers now know your name, what are you really trying to accomplish here?

      1. “The standard you walk past is the standard you accept.”

        Someone has to speak up in order for change to happen. Maybe it’s not Linus who changes, but the rest of the developers who emulate his communication style. You can bet people will think before flaming, at least for a while.

        My end goal has already been achieved: to open a dialog about this issue. Kernel developers will talk about this one on one at conferences and in private email. This will be discussed at kernel summit. Let’s talk about the elephant in the room, rather than ignoring it.

        Personal change happens slowly. Community change happens slowly. Nothing will change overnight. But change won’t happen if people who speak up for change are silenced.

  94. It has been more or less the same for 20 years and while I think that it can be daunting and scary for new developers who are unaccustomed to it it does have some merit.

    The two cases is actually well deserved of some crap… That does not mean that it could not have been toned down a bit.

    I very much prefer to unvarnished truth with all the human emphasis and emotion than what is much more common in most companies where appearances are more important than results. No its not logically either or but I have come to see that it very often works that way.

    I am not saying you should not voice your objections but neither should you try to flush the baby out with the bath water. I think you should seriously consider that there may be something too it (maybe a tad too much) when it actually spawned what arguably is the best piece of software ever produced over a long period of time.

  95. From Linus’s December, 2011 rant against Novel/SUSE/openSUSE bug trackers (this blowup due to Linus’s daughter’s laptop not being able to sign onto her school’s network printer without a password):

    Main thread:

    Linus comment:

    You’re a distribution. Your *ONLY*GOAL*IN*LIFE* should be to make something
    that works.

    If you say “We ship shit, so you need to be an expert and fix it up in order
    for it to be usable”, you have failed at your job.

    And seriously, that is exactly what you said.

    OpenSUSE 12.1 network configuration *IS*NOT*USABLE* in real life as-is.

    Don’t tell people to edit their polkit privileges to individual needs. Make a
    usable system, or at least expose a big and visible button saying “make this
    system usable”.

    As it is, the only people who can fix it are people who know more than the
    average bear. That’s a disaster.

    This is not about “security issues”. A unusable system is always secure,
    because nobody *cares*. It’s crap. It is, as somebody commented elsewhere, like
    making everybody have their shell be “/bin/false”. That’s really secure, but
    since it means that people can’t get any actual work done, who the hell cares?

    That kind of “security” isn’t security, it’s just stupidity.”

  96. Man, It is unbelievable what we have seen on this subject. One person, independent if she/he is male, female or alien is just asking for people put in practice what their parents (supposedly) taught them: be polite. By politeness I am not saying “always agree” or “accept anything”, but just disagree or refuse a merge request with *only* technical arguments, not treating people . What we have seen here is a bunch of posts of the kind “I have a bigger dick, so I am right”. I am not saying sometimes the discussion can get very though, but in any healthy community (of anything), this should be the exception, not the rule. Messages of the kind “go back to kitchen” in plain 2013 just astonish me. Sarah, you are completely right, congratulations for your attitude!

  97. I teach art, and a primary means of reflection within art instruction is the critique. For generations, many felt that the more direct “read: brutal” a critique was, the better for the student. Well, most of us don’t think that way anymore. A great short book on best practices regarding giving feedback is Liz Lerman and John Borstel’s “Critical Response Process.” At 60 pages, it is well worth the read, and is applicable to any field.

  98. Hi,

    while I’m not directly involved (I’m not a kernel hacker, I’m working on userspace bits for a driver) I fully support your stand and hope you will make things change in the right direction.

    I think abusive behavior is hurting development in the long term : the community is likely missing some potential volunteering hackers because these hackers feared some LKML reaction, and I think most current paid Linux dev were recruited among the volunters.

    And on a more personal side, I dislike when people start to make comments on me based on patch I send, it literraly kills my motivation to fix the concerned patch and submit future ones.

  99. Sarah,

    I am no programmer, hacker or anything else but I say good for you. I spent years of my life in the military defending and promoting the rights granted to each of us by the Constitution that include the freedom of speech. Your female, Ok, whats the point? You do what you have to do and don’t take shit from anyone. Stand your ground, Speak your mind, though be humble and admit when your wrong.

  100. Sarah,

    This is an amazing topic you’ve touched off! Stand your ground and believe in the good cause!

    For my part, I’ve only ever contributed one backported patch to the kernel, and after some nice people on the lists pointed out where my submission was going wrong, it finally got accepted. Had they been dicks, there’d have probably been no patch.

    I think it will be very, very important to make a clean distinction between blunt truth and verbal abuse. It’ll be hard for some people to accept, or even understand. Blunt truth can be to say: “This code is junk for reasons A, B, and C.” Verbal abuse is when it rages into “And you’re a stupid idiot.” People make mistakes; we’re human. But unlike the common dog, we ought not rub each other’s noses in it. Respect is for certain the absolute foundation of all communication.

    I’m a bit fascinated by Linus’ various responses. He seems a very angry person, hell-bent on projecting an amazing hostility. Is it possible this is a projection of hostility he receives? Or the manifestation of the stresses of being The One And Only True Master of Merges? Someone made the argument that kernel dev was like driving an F1 car, but hey, even light has a speed limit. Sometimes the best thing to do is slow down. Not everything can be fixed all at once. Not everything should be merged immediately.

    Linus seemed concerned that people would be angry with him when they didn’t listen to him and his warnings against implementing things wrongly. Why should their anger even concern him? Why is that part of his justification for behaving the way he chooses to? Nothing evidently stops him from rejecting code he doesn’t want in the kernel. You can’t tell people what to feel, so in that respect he shouldn’t base his actions on their (to-be-received) anger. Reject, explain in a zero-emotion manner, and walk away. Nothing says self-control like that.

    The sad fact is that those who stay and take the abuse end up enabling it, whether they want to or not. Those who profess “I’m a dick because that’s who I am – deal with it” I think have much deeper problems than just being an dick. Worse, even, when that someone is (a) very very smart, and/or (b) very much in control. I’ve had my share of (very painful) personal experience with the like, and it’s not something that should be taken lightly by anyone. It isn’t about culture, and isn’t about getting things done. Those are two cop-outs if ever I heard them. And that being said, if Linus’ methods “work,” it may be purely by coincidence.

    As my wife would say, there are plenty of hotel guests that get just as good service (or even better) for being nice instead of being douche-bags. Where those who hide behind the nasty tendencies are concerned, there will be a great challenge. I think only leading by example will change the way things are in the long-run. A collective effort. No one can look to people like Linus or others like him to be better people. WE must be the better people, and make no excuse for those who chose not to be. We must encourage others to do the same, by not tolerating disrespect, by shutting the door or turning off the car when anger takes the front seat. Sooner or later I think they will realize it’s better for everyone to play along.

    A disclaimer: I only mention Linus here as he is so prominently featured in these debates. This problem of people riding their emotions is systemic.

  101. Found you from twitter. Thanks for standing against abusive words which can be as harmful as abusive behaviour whatever the excuse. I remember the good old days when even “damn you” sounded too harsh. Things are not the same anymore. Even network tv suddenly turned to full-time bleeping a few years ago! It’s so sad the youth of today are expected and encouraged to abusive speech like never before. Alone, the individual lacks the power to stop the bully’s barrage of curse words. Only a network of like-minded people can empower us mentally and stop the maniacs who support, condone and promote verbal abuse everywhere they are. I am speaking in general and not particularly against any individual.

  102. Welcome to “Hell’s Kitchen” or were you looking for 12A?

    M: Well, I was told outside that…
    Q: Don’t give me that, you snotty-
    faced heap of parrot droppings!
    M: What?
    Q: Shut your festering gob, you tit!
    Your type really makes me puke, you
    vacuous, coffee-nosed, maloderous,
    ARGUMENT, I’m not going to just
    Q: OH, oh I’m sorry, but this is abuse.
    M: Oh, I see, well, that explains it.
    Q: Ah yes, you want room 12A, Just
    along the corridor.
    M: Oh, Thank you very much. Sorry.
    Q: Not at all.
    M: Thank You.
    (Under his breath ) Stupid git!!

    — Month Python “The Argument Skit”

  103. …allthough it is not about gender…i encounter this kind of ego unManagment on a daily basis inside my gnu-, cyber- (& not so -anarcho) -feministic activism & advocacy; inside of a local hackLab… \o/ :-/
    This is actually an old thing, which should be managed from perspective of psychology & sociology…maybe culture…
    I read the facebook & efyTime entry on this subject…i”ll try to followItUp & write
    an article on this on maybe VOX Feminae & Femen groups…
    Sarah, do you have links on exact reference to your issue…? where mailingList is reffered to you personally…
    It is about transFiguring ego to creativity… 🙂 \o/ <(")



  104. I’m with you Sarah; not because I have anything against Linus, but because I have had the same issues with a few other FOSS-ed projects myself and I had to unsubscribe from their forums, their mailing lists, and their IRC channels, because I went depressed with their behavior, even though I was offering my QA-ing skills to detect serious bugs that would crash the entire application AND answering the same questions over and over for new forum users.

    I just can’t stand the abusing discrimination that happens on (almost) every project out there that its main development team tries to either underestimate or cyber-bully users and/or any candidate developer for who-knows what stupid reason they would use to excuse themselves for such behavior.

    It’s like having to do with boys that cry when you take their *toys*. I remember this kind of aggressive behavior back in 2002-2003 when I first installed RedHat 9 and went on IRC to ask questions…I got blacklisted within 15 minutes from almost every channel I used, for asking naive, innocent, novice questions. I went like “those guys have no life!” and started crying for getting offended that badly. It was a horrible, painful feeling even for a man like myself.

    Is this the way you treat your people *Mr.* Linus? Then why should we give any form of donation to keep any project alive since various projects don’t keep our spirits as such?!

    Just because someone has one of our many intelligences quite high, does not make him *smart*, so he or she should be abusing people for not being able to enter his or her stupid mind and think the same.

    My advice to you Mr. Linus: start learning to respect yourself, so you can finally learn how to respect others.

  105. Saw a post about this on Planet Blue, and came here to comment.

    I applaud you for standing up for something you believe in, with the goal of making a situation better for everyone, even though you knew it might end up a personal disaster (especially for your inbox). You did the right thing.

    A lot of programmers/developers (and everyone else…) act like being “professional” is some kind of plague. It’s not. Being a jerk doesn’t make you a better programmer.

    I believe the opposite of Linus, that everyone should be more professional all the time. The world would benefit from more civil discourses and less yelling. At the very least, people might not feel comfortable wearing pajamas to the mall anymore.

  106. In college (1997-2001), I red LKML every day, every post. I had discovered Linux and though it would be a good place to learn about software. I was right, but I lost a lot of respect for Linus in the process.

    Go Sarah, hope you are still finding time to do PSAS stuff, and hope you are enjoying the rest of your job at Intel.

  107. I just wanted to commend you for your post on abusive language on lkml. Women need to speak out more on the women unfriendly environment of much of the Linux and tech community. Men need to be more aware of the fact that their posts are read by real people. I have had many women friends and colleagues who have left the technical side of the industry because of the boys club attitude towards women. And I for one, enjoy working with women.

  108. It’s my opinion that your crusade has an honorable purpose, but that you’re trying to supplant an open and free culture with one that puts value in limiting how people interact with each other.

    It’s definitely distasteful to be met with scorn and words that offend me, but I’ll gladly accept them and the freedom to respond in kind if that’s the culture I keep. I think where we differ is that I reach a point where I can ignore what I dislike without feeling like I need to change it so that I don’t feel hurt or upset.

    An open culture is better than a closed one.

  109. Verbal abuse is real and I had my fill of creeps in my career, but context is important. Young boys will physically fight for a coveted toy but if they are buddies they will then go on to happily play together. They may be sizing each other without intention to harm each other. It becomes abuse when one of the boys hides or breaks the toy or hurts the other child with the intent of keeping control. Context and intent are important and not taking things personally too.

    Among the creeps in my life there were a couple of smiling faces who rolled their eyes in faked disgust and spread rumors when they thought it was safe for them to do so. I know how to deal with the overt aggression but the covert one is an entirely different ball game.

    There is a folklore among leaders of yelling at people based on the observation that compliments typically are followed by lower performance and shouting is typically followed by increased performance. It turns out this observation is very real but it is entirely explained by regression to the mean. Good people management requires good statistical thinking skills.

    Best of luck and stand your grounds.

  110. Sincerity is more important than civility because the former implies respect but the latter.doesn’t. The manager who harps on me just for the heck of it isn’t sincere, the manager who sweet talks me while pulling the carpet under my feet isn’t sincere either. I will not volunteer for someone that I don’t feel is sincere.

  111. We are human, not animals. Stop any form of abuse The world begins worse and worse, people is struggling, Be nice to people.
    I support you.

  112. I agree with both you and Linus. You’re right that no one should put up with abuse/bullying, but he’s right that this is only email and in the examples you provided he was justified to “raise his voice”.

    I believe you should continue to point out bullying and Linus should continue to administer his flock of developers how he see’s fit.

    If either of you actually prove the other wrong then we can settle things. Otherwise, as I started with: both of you will continue to have a point…

  113. Sarah:

    Some time ago I filed a kernel bug. For an average non-programmer Linux user, this is a very intimidating thing. I prepared for days, gathering data and testing. Finally I posted the bug.

    It was assigned to you. You went through testing procedures very thoroughly with me over the course of several days. You even patiently walked me through applying a kernel patch, which is child’s play to you but was very new to me. You treated me with respect and approached my issue with professionalism, concentrating on the hardware and software and not the fact that I’m a novice in way over my head.

    It was a nice welcome to the community of Linux developers and made me feel like I was making a contribution in some tiny way.

    If, however, the bug had been assigned to someone like Linus I would have deeply regretted ever disturbing the delicate genius with my menial problem. I would have shut up for good and if I ever found any kernel-level bug ever again I’d be sure to be quiet about it, hoping someone else finds it.

    Linus could learn a thing or two from you, but his attitude shows he thinks he knows everything already. You have my support.

Comments are closed.