After Years of Abusive E-mails, the Creator of Linux Steps Aside (The New Yorker)

I usually try not to conflate Causality and Correlation, but in this case, my best guess is: Yes. It seems like Linus and the LF suspected what was coming from the New Yorker, and took preemptive action.

"I do take issue with the characterization of Linux as "the basis for other operating systems like Google's Android for phones and Chrome OS for computers" instead of noting that these are varieties of Linux."

That is an extremely correct description of what Linux actually is. Linux is an OS kernel, not a complete OS.

> What am i missing here?
The fact that a significant proportion of the population is now pathologically sensitive to perceived criticism.

Are you saying that all instances of the word “fuck” are equal?

Clearly, when it's directed towards someone in an angered or negative way, it's very different from “fuck, I'm hungry, let's go have some food” at some college campus. If people at the colleges you frequent go say things like “Mauro, SHUT THE FUCK UP!” to each other, I'd suggest you go to less abusive colleges. And Hollywood movies are not role models for how people should behave against each other in real life (and thankfully, people generally do not behave as in Hollywood movies).

If Linus were responding to patches with "this is fucking great, thank you", I don't think anyone would mind. It is not a particular four-letter word that really matters in this case -- despite the New Yorker article starting with a list of those they found. It's the tone and content of the message in general. That said, sprinkling obscenities into it doesn't help.

College campus, Hollywood movie, city centres. Of these, timing the interval between the F word would be extremely boring, except for Hollywood movies. Where I live, you could probably spend a day without hearing that word on a campus or in a city center. Because surprise: not the whole world is the United States of America. And not all Linux kernel developers live in the United States of America, even though most of the online communication is done using the English language.

If Linus just used the word "fuck" as a general-purpose expletive, it would not have been a very big deal (although it is unnecessarily offensive to some people, and thus an unnecessary impediment to getting along with strangers).

The problem is his sweary rants attacking people and their work e.g. "you are a fucking moron". That is what triggers the complaints about abuse.

> And, as the article points out, women are not allowed to adopt that sort of attitude (some tried).

This is not true, and the article does not say that. What the article says is that Aurora and Sharp tried to be abusive, but others retaliated. The reason for their failure? They are not Linus (doh!), who is widely recognized and respected for his technical achivements. They, not so much.

What the New Yorker does is (insidiously) insituate what you read, by placing this phrase just before, in the very same paragraph:

> Valerie Aurora, a former Linux-kernel contributor, told me that a decade of working in the Linux community convinced her that she could not rise in its hierarchy as a woman.

See that? You just made the connection, not them.

But. In any case, bad behavior is bad no mather who is doing it, isn't it? Then why did they try to be abusive? Would there be a case if they had been allowed to be abusive? Questions, questions...

Thank you.

In what way does the introduction water the article down? There's only one sentence which makes a passing reference to the content, and then a short quote from the article. This falls in the category of notification rather than in-depth article. That the article has some points that long-term followers of Linux might agree and disagree with seems to me to be a true and unobjectionable statement. Do you think there are no Linux followers who found some content in the article "over the top"?

For heaven's sake, one of the nasty e-mails quoted was 1) not an e-mail (it was on a social media account), 2) not directed at anyone in particular, and especially not to a Linux code contributor, and 3) clearly a joke. I don't want to put myself in the position of defending Linus callous communication style, because I haven't liked it either. But I dislike mischaracterizations and distortions.

Let me lead with a highlight of the problem, before providing a stack of other relevant quotes.

> There is, among some people, a sort of hostility toward the kernel community that brings a special toxicity of its own; there is little to be done about that.

The article reserves its strongest language to describe people who complain about the problems in the kernel community. Linus is "a harsh critic", messages on LKML are "inflammatory", but outsiders have "hostility toward the kernel community" and "a special toxicity", and people are afraid of "a threatening change pushed by people with a hostile agenda". (Yes, I'm aware that last one came right before trying to reassure people; you should know full well that saying "you might fear X but don't worry" still highlights X and frames the way X is described.) Defensive, us-vs-them, othering. You're demonizing those who complain or call attention to problems, while feeding into exactly the fears you're theoretically trying to quell.

For the record, I'm not suggesting that it was a bad idea to help reassure people who might well be unfamiliar with this kind of thing, especially those who are coming from a position of privilege and who haven't had to deal with as much of the kind of behavior that this change aims to address, and who wonder what they in practice need to *do*. Parts of that were well said, and needed saying. I'm not suggesting that you *should* "just roll over the very real fears that some people have". Explaining "what this likely means for you in practice" is *useful*. But consider the above, and consider the following:

> in the end, less may change than many expect or fear.

> Torvalds has always been known as a harsh critic; what perhaps fewer people have seen is that he has always been willing to be just as harsh toward himself when (in his opinion) the situation warranted it.

Literally the first sentence after the fold starts out on the defensive.

> Critics of the kernel community have spent years calling for the establishment of a proper code of conduct.

This creates an "us vs them" tone, without observing that many of those "critics" are *part* of the kernel community, not outsiders.

> The adoption of a "code of conflict" in 2015 did little to mollify those critics

Right, because the problem is mollifying critics, rather than solving the problem?

> On the other hand, the kernel community continues to grow,

This is the kind of statement that has made the Linux kernel such a damaging bad example: it keeps growing, so why is there a problem?

> my beliefs that (1) the situation is not as bad as many like to portray it, and (2) things are getting better anyway.

> It was not the code I would have written

And by saying this you immediately start undermining it.

Perhaps you'd like to expand on what code you would have adopted? You have a golden opportunity to do so, since LWN could use one. Personally, I'd suggest using the same one the kernel does, for uniformity, for the same reason that projects should adopt GPL-compatible licenses rather than incompatible and/or custom ones. At the moment, LWN's approach towards toxic behavior in its own community comes across roughly like https://www.youtube.com/watch?v=W9ZD3_ppcPE .

> Such laundry lists of misbehavior can leave a bad taste in the mouth; that can be especially true if, like me, you are an older, run-of-the-mill white male; it is easy to look at a list like that and say "everybody is protected except me".

For the record, lest you think I'm criticizing everything here: this was well said, and needed saying. This *is* how some people feel the first time they encounter a code of conduct; among other things, they have to think about the discomfort that others regularly experience. That *can* be very uncomfortable to think about, but then, it's even more uncomfortable for those who go through it.

> Many bad experiences reported by developers are associated with crossing into a new neighborhood and unwittingly violating one of the norms in place there.

This is an interesting choice of phrasing. I would imagine you might have meant it as, for instance, coming into the kernel community and not following the conventions there (a variation on the old "lurk more"). But given that "norms" (especially in the context of communities or neighborhoods) tends to be most strongly associated with behavioral norms, this also comes across as highlighting developers who find themselves "unwittingly violating" the standards of behavior of one of the more friendly communities.

> Hazards

> Some, seemingly including the author of the Community Covenant, think that the kernel community may be a lost cause.

The tweet you linked to said nothing of the sort. The damage that *has already been done* isn't necessarily reversible. But the very same tweet says "That being said, let’s see if things get better from this point on."

> There is, among some people, a sort of hostility toward the kernel community that brings a special toxicity of its own; there is little to be done about that.

Already commented on above.

> On the other hand, plenty of people — generally not those with a lot of contributions to their credit — claim that the adoption of a code of conduct will force developers out and be the beginning of the end of the kernel project. Now that the "social justice warriors" have taken over, the real developers will flee and it will all collapse into a heap.
> That outcome seems unlikely

For the record: reporting on *these* "predictions of doom" seems perfectly reasonable. (Though I would have also quoted "real developers", since the false dichotomy there is very much an invention of those making such dire predictions.)

> The community as a whole will have to find a way to implement the code that handles the community's inherent conflicts without collateral damage.

This feeds directly into the fears of people who equate "don't be awful to people" with "you can't criticize or reject bad code".

> It is thus unhelpful that it was adopted so abruptly,

More undermining.

(And, of course, we now know why it was adopted abruptly.)

> It would have almost certainly been better to go more slowly

This has been needed for years. Cries to "go more slowly" ring rather hollow when absolutely *nothing* was done before. How many of the people popping up to complain now did anything, at all, to help? How many of those people did anything to improve the hostile environment that made people lose hope anything would *ever* change?

(And again, we now know why it wasn't done more slowly. Which is almost certainly going to lead to more people thinking "it would have been better if we weren't forced from the outside to improve". Personally, I wonder if without that outside impetus, anything would ever have changed.)

> on the other hand, any public discussion would have had a high probability of being an unpleasant experience for everybody involved.

Thank you for observing this.

More importantly, perhaps: such discussion would have had a high probability of either not actually getting anything done (design-by-incremental-evolution-with-small-patches does not work here), or of ending with something very much like the "Code of Conflict".

> in an attempt to show which effects it will (or will not) have

> the code of conduct might be seen as a collection of heuristics that never quite converge on what is really wanted.

Even more undermining.

> I think that some people wanted a piling on, another hit piece; I wasn't going to do that.

I certainly was *not* looking for anything of the sort. Complaining about an article being slanted in one direction does not mean I expected it to be slanted in the other direction. And I *do* think that a careful discussion of "what this means in practice" would have been a valuable contribution. You weren't wrong about there being people who don't fully know what this means yet, and who wonder how it affects them and what they might need to do or to change.

As an example, it might have helped to take some time to talk in more detail about how this doesn't mean accepting bad code, and doesn't prevent people from being direct and clear in their responses and criticism. Or talking about how the first response to problematic behavior in the community would typically be a reply, not a ban; the word of a kernel maintainer carries a lot of weight, and may hopefully be enough to help people rethink their behavior. So many things the article *could* have done to help people...rather than highlighting fears and hazards, undermining, and otherwise coming across as reactionary and defensive.

One point stands out for its simplicity. You quote the individual apologizing, and then talk about them "confronting some internal pain". But you don't have any such words about pain or damage that was inflicted. This is a textbook pattern applied in abuses of power. It is used to shift the readers sympathy to the powerful. Please don't do it.

I think I can summarize my problems with the article by saying that it is defensive.

It quotes from Linus' apology and admission of doing harm to others, and then spends a lot of words to provide mitigations.

This is a bad way to report an apology for harm done. It weakens the apology. Political strategy people could probably explain this better than me, but this can be used as an intentional bad-faith tactic. It is best avoided.

Secondly you had the option to describe several significant reactions of some who were harmed; who were "possibly [sic] drove away from kernel development entirely". Or provide information about the harm. You decided not to.

You *could* have cut down the defensiveness, and be left with a newsworthy update. But the defensiveness and omission together make a terrible combination.

(I think I see defensiveness also in the section where you are most re-assuring about change. As you say, this is not the change you would have written, so you give justifications why you did not protest it at the time. As a reader, I had the impression you felt strongly about writing up the personal justification. My impression is those personal feelings did not help in writing a good article).

You relied on the apology and discounted "critics of the kernel community". But they correctly predicted there was an omission that was significant, and in fact what that omission was. A wisdom gained from harsh experience.

I'd like to read more from the Corbet who wrote: "I don't think we'll improve things by questioning the experiences that people are reporting. I'd much rather focus on how we can make things better without making them worse in other ways."

It is generally the LWN policy to refer to people by surname.

The TAB was also referred to in the previous "code of conflict".

Maybe the TAB election system needs a rule change, call it the diversity rule: two out of ten positions should be reserved for female candidates. I know, you can poke all kinds of holes in that, but it would be better than the current disastrous situation, effectively indistinguishable from a boy's club.

biased or - if we can be a little more paranoid - paid by someone with heavy interest in discrediting Linus and/or Linux. I'm not aware of the kind of journalism this "New Yorker" medium provides (i.e. if they are more credible than what this "Huffington Post"-style article implies) but in this age we live in i'd rather read anything with a truck-load of salt

> Took me about ten seconds to find that

It took a search engine about ten seconds to lift that out of the context of over twenty years' worth of LKML posts. It would be more relevant to state what proportion of Linus' posts such comments represent.

Just for fun, I decided to read the thread to figure out why it ultimately came to such insults. I have to say that while Linus (nor anyone else) should have responded that way, I can certainly see myself *thinking* exactly that (before taking a short break to allow for a toned-down reply). It got me thinking (in general) about what's the best way to deal with behaviour that despite being polite (words used) is also annoying (e.g. badgering).

Linux is community effort, let's drink tea and stop worrying bad code.?

Linus is the reason why ugly code are keep out of the tree.
Look at all those crap ISO accepts because they don't want to hurt people.

> > His critics don't seem to amount to much as programming professionals

> It is unfortunate that LWN user names are pseudonymous so it's not easy to get a sense of who Linus' critics here are. FWIW my name is Robert O'Callahan.

If you want to know what my real name is, I gather it's easy to discover. FWIW, on the net I *always* use the nic "Wol" (which is actually based on my real name).

What I would be concerned about is if the critics are using throw-away nics. If they're using a nic with a history, who cares what their real name is ...

Cheers,
Wol

Well, without Sage Sharp we possibly wouldn't have a USB3 stack, and certainly wouldn't have had one so early or of such high quality. But Sage is clearly, y'know, just some nobody. (Or not.)

> I suspect that such 'coaching' would be key to the future demise of Linux.

Why?

The vast majority of the community ought to be able to improve without psychotherapy. I do not doubt the value, but do we really want to see a lot of "I didn't get the therapy therefore it's business as usual".

What they are doing is spinning a story according to what sells, and to what little they understand of the details.
Accusing Linus of having built a cult of personality is uninformed, rude, and plain clueless. Granted, there _is_ a cult of personality of sorts, because many people love using Linux, trust Linus' technical judgement, find his manners amusing for the most part, and are not easily offended. I'm one of those "cultists".
I still cherish the clash of civilisations between Linus and Andrew S. Tanenbaum (whose opus magnum sits on my bookshelf and is cherished as well). Even then, Linus could go ballistic in a highly entertaining way, and apologise when it was pointed out to him that he was drifting off-target.
But to suggest he _built_ a cult around himself without any proof is just an intellectual short circuit, to put it very mildly.
That's when "I don't like how the world is" equals "something sinister must be at play here".

The title of the piece is disqualifying already. "After years of abusive emails", in this time and day, alludes to all kinds of things that are really orders of magnitude worse than Linus' lack of civility. That's what I hate about PC shitstorms - all of a sudden, raping a person is the same as grooming and blackmailing teenagers is the same as wanking in the presence of others is the same as having an affair is the same as yelling at people.

How about "after years of managing a huge project by email (the first few entirely for free), and occasionally hitting a wrong note and hurting feelings, but also preventing many ill-advised kernel misfeatures"? But no, that's not going generate those clicks...

Touché.

But what is the alternative to quiet reflection that's on the table now? It would appear to be character assassination by tweet in response to a carefully reasoned and worded discussion contribution, even if it's controversial. That Sharp vs. T'so clusterfuck is just plain sad. I don't like to watch the fireworks, and I don't feel like having popcorn.

I was sad to learn of Sharp leave kernel work behind because of unpleasant experiences, and yes, they should be addressed. But eradicating all traces of civility and clearly overstepping all bounds with baseless ad-hominem attacks in the fight for a better world is, well, counterproductive.

So yes, the whole process has been slow in coming and that's nothing to be proud of, but hooray for shutting up in public and quietly reflect.

"I invented the code" is nonsense as code isn't invented and most of the code in Linux wasn't written by Torvalds, anyway. Hence, I don't believe that this is a quote until the person who was supposedly quoted confirms that he actually said that in this way.

Whatever this "suggests" to you is entirely your concern.

The New Yorker reports that Linus used the terms "slut" and "bitch" much less frequently than "crap" (they explicitly mention those terms). But looking at the study itself, it explicitly states that they do not have data on the use of those particular profanities and others, because there is too little data on those words for analysis. So the study does not say that Linus used those terms, and yet the New Yorker's "rigorous fact-checking" cites the study to claim that he does (though infrequently). That stood out to me as an egregious factual misrepresentation in the article, and quite possibly a smear.

I'm sure I saw something a few years ago about how US copyright law allows a licensor to unilaterally withdraw from a licensing agreement, but I'm not sure how to find it now. I'm pretty sure the purpose was to allow an author who entered into an unfavourable licensing agreement before they became famous to cancel that agreement so that they can earn more money later on.

This is great. I hope many others feel the same. Thank you!

Describing personal attacks and personal denigration as "abusive" is a normal use of the term in English. In everyday use, it does not require any specific intent by the person being abusive to gain anything from the abuse.

For example, the definition in Merriam-Webster (one of the standard dictionaries of US English) is:

1 a : using harsh, insulting language
b : harsh and insulting
c : using or involving physical violence or emotional cruelty

Whether Linus's behavior rises to emotional cruelty in definition (c) is part of the hotly-fought ongoing argument over what impact this sort of language from an authority figure has on others, but regardless of one's opinion on that argument, his behavior fits the normal and typical use of definitions (a) and (b) in US English.

The language (newspeak) is part of the problem and an essential attack weapon in these shitstorms. If the language makes it convenient to magnify an everyday rudeness into a sexual offence, of course some people are going to make use (or abuse) of that. I understand the word "abuse" per se doesn't refer to anything particularly violent, but when you hyperbolize that into "years of abusive emails" it sounds pretty bad.

> English is blessed with so many words, we don't have to put antisocial nerds and violent sex offenders and rape murderers in the same bag.

I'm also not a native English speaker. When I was a student in Britain at the turn of the millennium, "antisocial" seemed to be a common word for describing any kind of behavior you didn't like. The young people who used this word were generally very social among themselves, but antisocial to everyone else. For instance, walking together in a tight group that takes the entire width of the walkway.

I'm not against other people in general, so I prefer to be called "asocial" rather than "antisocial". I think it's a common mistake, and not simply at the level of vocabulary. Extroverts expect everyone else to be extroverted too, and if you don't get all social with them, they might think it's because you have something against them. This also shows in the different styles of communication, which is probably one of the root issues here -- coders like to get to the point rather than spending precious baudrate on words like "please".

That is a great explanation.

Okay, so this is in fact about snowflakes, who incorrectly interpret "$open_source_project is for boys" incorrectly as "$open_source_project is _only_ for boys" instead of "$open_source_project is _also_ for boys"? And who you expect to be predominantly female? Fair enough.

I say the world nowadays is: either you agree with this "everyone gets offended" culture, or people will hunt you professionally, socially, personally in order to end your career. So Linus is joining the new worldwide order of folding into the offended to survive.
Just check in the article: Linus discouraged women to joining Linux kernel? When did that happen? Follow the "rumors" and you'll see people who make money out of victimizing others are the ones behind this affirmative.

> Linux development is a work environment.

For many of us it's a free time activity.

And the crux of the matter is this quote from that article "The Contributor Covenant is a code of conduct which transparently attempts to push certain norms of certain parts of the US political spectrum on the rest of the world"

==========

It's not clear to me what "norms" the author is complaining about. Would that be the "norms" of treating people with respect? Is that a deal-breaker for some people?

Here is the video moment of Jim Zemlin's TEDx talk,
https://www.youtube.com/watch?v=7XTHdcmjenI#t=12m50s

He admits that, although the boss of Linus, he was sort of a bystander to the development process of Linux.
In the video he is rationalizing what has been going on on LKML.

It looks like this follows the patterns with some hedge funds and other companies that have star employees.
Those bosses overlook or hide any problems with those star employees in order not to disturb the phenomenal success.

The Linux kernel is extremely popular because it ended up being used in so many places.
Companies have to put employees to work on the Linux kernel and there is a steady stream of new developers even if many of them are turned off by the toxic environment. Currently, the Linux kernel development affords to lose developers and does lose them.