Linux kernel's Torvalds: 'I am truly sorry' for my 'unprofessional' rants, I need a break to get help Linux kernel firebrand Linus Torvalds has apologized for his explosive rants, and vowed to take a break from the open-source project and seek help. In a mailing list message on Sunday, Torvalds admitted his "flippant attacks in emails" to fellow Linux programmers and project contributors "have been both unprofessional and …
Absolutely.
If there's one thing that History shows us, it's that focussed leaders who will not compromise on quality or security and who have headed revolutionary products for decades create bad products compared to teams of management graduates who understand team-building because they went on a course about it once.
So yeah, Elon. You'd better stop giving your all to working towards your dreams, or else!
" it's that focussed leaders who will not compromise on quality or security and who have headed revolutionary products for decades create bad products compared to teams of management graduates who understand team-building because they went on a course about it once."
If you look at really big projects like CERN, or mainstream car builders, Boeing or Airbus, that's pretty much correct (except that European managers tend to be much,much better trained than Anglophone ones.) Then there's the armed forces, also highly trained in management and co-operation. Part of Eisenhower's genius was applying management techniques to invading Europe.
Every project tends to need a Type A person or three to kick it off and establish the outlines. But when it becomes an essential part of a lot of people's work, it needs the skilled administrators to keep the edifice functioning. The difficulty is usually for the founder to let go, while experiencing growing frustration because there are things outside his (or her, in some cases) control. One startup I worked for, the wife of the founder demanded he take a month off in the West Indies away from the phone. It probably saved his sanity.
Unfortunately there are always people like you about, willing to worship the Type As and validate their view of the world when you should be whispering in their ear "Remember, O Caesar, that thou art mortal".
Yet it had to remove Patton because his habit of yelling at people. Just, he was one of his most skilled generals.
Maybe Clark yelled less to subordinates, but he was an arrogant too - he lost the opportunity to cut the German withdrawal because he wanted to enter Rome first, and be crowned "Caesar" there.... just his glory dreams were cut short because of D-Day two days after...
Eisenhower also made big mistakes once in France - as attacking on a too broad front - which probably prolonged the war, albeit a lot was also due to British and US generals unable to look beyond their petty career...
Elon Musk's problem is not that he won't compromise. It's that he won't delegate. Maybe he sees the two things as the same, which is a serious problem. You just can't run a company the size of his companies while micro-managing every detail on the ground; you have to learn to hire smart people and trust them to make the right calls. Otherwise you become a choke point for the whole operation, as Elon is now. While a lot of people saw him pulling all-nighters at the factory as heroic, I saw it as a serious warning sign.
"I hope this is a success for Linus Torvalds and linux."
I also hope normal service is resumed as soon as someone tries to slip some nonsense into the kernel and won't take a polite "no" for an answer. The Code of Conduct referenced includes something about accepting constructive criticism. AFAICS the outbursts have followed failures of that..
The man has taken a good hard look at the toxic way he sometimes communicates and plans on taking a break so that he can come back with a more positive and productive attitude.
I agree with him on his technical points but I'm sure over time he can learn to communicate those in a respectful and professional way.
I sort of know this behaviour - not as strong, but I get really p'd off at people who are plain and provably wrong, and really have the urge to tell them they are morons (they are, to me, at that point). Time taught me not to do this. And the whole social interaction / emotion stuff is... tough. I have some colleagues that are really good at that (and still great in what they are doing), and that sort of showed me where my skills are lacking. Self awareness is the first step to improving things. I hope.
And to those who now go "boo-hoo, he is now all PC and bowing to the pressure of $(group)":
He is not (and your ID should be 10-t for starting this "discussion"). Telling people publicly they are idiots and insulting them and their family tree is not "free speech", it is not "talking true". Your freedom stops where mine starts - and vice versa. Plus you just cannot have a civilised discussion based on facts in an environment like this.
Linus Torvalds is not going to stop complaining about bad code or admitting bad code into the kernel. He is going to be (hopefully) a bit more civilised and no longer doing an ad hominem when pointing out mistakes.
The man has taken a good hard look at the toxic way he sometimes communicates and plans on taking a break so that he can come back with a more positive and productive attitude.
I can't help but think that this may be partly why he decided to holiday in Edinburgh while the Linux kernel dev summit was happening in Canada.
Unfortunately the organisers deciding to move that to Edinburgh may have thwarted his plans for getting the break away from it all he desired and deserved.
[because fork may be taken as offensive over a noisy line] will it be the Code of Conduct Kernel, or HEBIRD for short?
FWIW, I tried to look at the link, and the meta frills were bigger than the entire document should be. Has anybody made it to the end of the thing and can tell me if it has the "assume good faith" it desperately needs?
What's wrong again with "Don't be a dick. Don't be a pussy.", other than the wording?
From my perspective, Linus only gets bent out of shape after all other methods bounce off cloth ears. Contributors get plenty of warning when they are messing up. Really. I've been contributing to the kernel for a quarter century now on and off, and have made a couple very stupid mistakes along the way. Yet I've never even been growled at. Because I figure out where I went wrong, acknowledge the issue, fix it, and move on. The only people who get yelled at NEED yelling at, IMO.
If Linus gives in to the holier-than-thou "I'm offended on the behalf of others" set (see the master/slave bullshit elsewhere), it'll be the beginning of a long, drawn-out slide into mediocrity. It is precisely his abrasive style WHERE NEEDED that has kept the kernel on track all these years.
On the other hand, devs could get "time outs" instead of being yelled at. They could have "do overs" when they break something. Every dev should get a trophy. Yeah, that's it. Treat the developers like a bunch of spoiled rotten children, I'm sure that'll make things all better.
I don't think the argument is that Linux needs to stop telling people off: It's that he needs to do it in a less aggressive/abusive/bullying manner. I can't think of any environment where being so abusive to your subordinates is allowed.
One wonders how many folks who consider Linus a "bully" have actually worked alongside him on the kernel for any length of time ... and how many are simply reacting to what they think they know as "fact" based on a couple of cases reported here at ElReg and other places, primarily as click-bait.
Speaking from experience, he starts with private email, asking something along the lines of "Are you sure you want to do ::that::? Because ::this:: will result, which isn't good". It's only after the idiot ignores the hint, takes it public, and then continues after getting a gentle smack or five, that the full blown bollocking commences.
And it only happens a couple times per year. You lot seem to think it's a daily (or hourly) occurrence. It ain't. Because most folks on the KML are both smart and reasonable, and NOT prima-donnas. The Poetterings and Sievers of the world are (thankfully) few and far between.
Frankly, I think Linus has shown incredible restraint, given the extremely large herd of cats that he is trying to keep in the worm can beyond the open stable door ...
See, that seems like a much more reasonable approach, and if he was doing that when he moved to a more public forum as well I don't think anyone would have an issue with it. I think I'd agree he isn't actually a bully, but he does seem to have a temper which gives him a bad name and discourages people from contributing to the kernel in case they set him off
+1, but, IMHO, you can achieve the same results more dispassionately.
"Your code is inadequate, badly designed, has repeatedly caused problems and we will not accept your contributions anymore"
(feel feel to prefix with, "Dear Mr. Poettering,".)
If you add a few F-bombs, even if they are well-justified, that detracts from the message. "Oh, yeah, Joe, he's always a bit colorful, ain't he?". The polite version is a more effective professional-level torpedo, because it is all about the code and the person delivering it, not about the gatekeeper's conduct. Note that it is not even "only about the code", but also about the person's work. As it should in egregious cases.
In the case of Poettering, Linus has never hidden his dislike of P's contributions but has still not managed to keep it from creeping into, and coupling (tech-wise) heavily with much of Linux. Perhaps a more stick-to-the-point approach, less liable to be brushed off as temperamental, might have kept systemd in its place (i.e. limited to being a _humble_ service manager). I don't know. I do know even Linus' uncompromising position did not avoid this regrettable outcome.
The new CoC is pretty short and quite fluff-free. It would be refreshing to see something like it replace many other projects' tedious CoC.
However, I do wonder one particular line of the CoC, under the examples section:
"Trolling, insulting/derogatory comments, and personal or political attacks"
"insulting/deregatory" seem well-aimed at Linus' occasional outbursts and liable to whack him if he steps out of line again. I wonder if we, as Kremlin-watchers of old, can infer a palace coup.
Linus is probably among the top dozen or so people that have achieved major technical changes in the last 20 years. He's shown, with git, that he isn't a one trick pony and can truly innovate. I hope that he can take this gentle remonstration, remain intransigent when needed while remaining polite and continue his excellent work.
If he gets sidelined, then I fear we've lost more than we've gained, despite the kernel contributors' right to be treated in a civil manner.
"Even if he does that, he still needs to do it in a less aggressive manner."
If I made a stupid mistake and someone punched me in the face I'd consider it aggressive. A bit of swearing? Not so much.
It sounds to me like Linus is feeling burned out.
"I can't think of any environment where being so abusive to your subordinates is allowed."
But can you also think of any other product on such a scale developed in that way? From TFA "10,000 developers from more than 1,200 organizations worldwide" all of whom are self-selected and over whom Linus has none of the conventional managerial options: he doesn't hire or fire (although he does have the option of refusing patches which isn't quite the same thing). He doesn't write annual reviews, recommend or withhold raises or promotions or award bonuses.
One has to conclude that having held that process together over so many years to produce a successful product he might be doing something right.
One has to conclude that having held that process together over so many years to produce a successful product he might be doing something right.
One can to the right thing the wrong way. I'm not sure the ends justify the means, even in Linux kernel-land. I mean, why not simply state, "Your fix is not going into the kernel, period.", as opposed to a 15-paragraph rant with f-bombs and s-bombs and a paragraph on the back of each one to be used against them in a court of law?
Yup. Telling someone "no" is what matters. Unloading all his personal angst onto them in the form of profanity may make him feel better, but it's not going to result in better code. "The beatings will continue until morale improves" rarely works as a management strategy.
"One can to the right thing the wrong way. I'm not sure the ends justify the means, even in Linux kernel-land. I mean, why not simply state, "Your fix is not going into the kernel, period.", as opposed to a 15-paragraph rant with f-bombs and s-bombs and a paragraph on the back of each one to be used against them in a court of law?"
AFAIK, there's only ONE justification for going into a tirade: because they just won't take NO for an answer. And even then there's the risk of getting into a shouting match. At least with online there's no chance of it immediately escalating to fisticuffs which is what usually happens in a shouting match between two parties who each believe he/she is in the right.
>I don't think the argument is that Linux needs to stop telling people off: It's that he needs to do it in a less aggressive/abusive/bullying manner.
That's my take on it too.
> I can't think of any environment where being so abusive to your subordinates is allowed.
Oooohhh, I know lots of places where that has been the case. My last two employers for a start. It does not end well.
Jake, I think I can identify with what you're saying.
The other side you need to look at is how you feel and act in your own niche. Where you're in charge and taking contributions from others. Some good, some dodgy, and some bloomin' impossible to evaluate. And differing attitudes from the contributors, with attitudes having no correlation with quality. When someone's quality is suspect and their attitude insistent, you might feel the need to put them down, right?
Now for most of us, our own God niche is small, and the amount of contributions coming our way is manageable. Linus, by contrast, is God in a very big world. That's a lot of pressure to deal with. Perhaps he does sometimes deliver put-downs beyond where they're well-deserved?
@jake: This. A thousand times this. The last thing we need is some touchy-feely commune with yurts, natural fibres and tofu-based nutritional snacks where nothing gets done but ever so politely and ecologically.
While I do feel that profanity should be more seasoning than sauce, the result of cooking without seasoning isn't exactly something you'd want to eat.
Re your last paragraph. Garrett himself is not immune to some pretty strong rants, including passive-aggressive (that I wouldn't like to be on the wrong side of) and in-your-face SJW. But following his blog (which I do via Planet Debian) is worth it for the good stuff that, for me, clearly outweighs the bad.
Birds of a feather there?
Is that before or after GIMP gets a rename?
(Though I'm still waiting on an open source JS project for applying filters to social network based in Shoreditch called wankr or some such...)
As for swearing... Linus is on the same level as my manager. You very quickly learn to be good enough not to induce his ire. So I'm farely nonplussed by it.
Still, it's worth remembering that Git was absolutely written in anger. Linux was developed using BitKeeper as the version control system until the sudden announcement that all versions of BitKeeper would now have to be paid for. Days later Linus released the first version of Git.
He has later claimed that it was a light-hearted bit of self-deprecation and, like Linux, it is named after him. I don't believe him.
He, he, it's a bit more involved than that. I don't recall seeing whether Linus thought the kernel community had acted in good faith or not, but he's the one who had advocated Bitkeeper so he worked out an alternative.
https://www.infoworld.com/article/2670360/operating-systems/linus-torvalds--bitkeeper-blunder.html
and, yes, this article was written before git came out.
As I recall, the main target of Linus's anger at the time was Andrew Tridgell, for coming out with the reverse-engineered BK client that provoked BK's owner to withdraw the free-beer client license for kernel devs. If git was named for any one person at that time it was probably Tridge.
However, the very swift appearance of git in the wake of BK's move did make me wonder whether Linus had already been working towards dumping BK for some time, having tacitly acknowledged the complaints from many around him about the risks of adopting it in the first place.
