I wholly agree. Having worked for a Linus-type, I've seen both sides of the coin. Yes, you need tangible requirements, project goals, assigned responsibility and real consequence for harmful actions in a large-scale project, but you don't need to be a dick about it. Of course, if that's your nature, then that's who you are. But people who aren't dicks don't need to be to get the same results. Much love to Linus, but for me, FLOSS isn't about that. It's about providing a positive service to humanity that allows freedom, education, communication and a whole host of socially beneficial functions to the world. How is being a dick conducive to that beautiful product of our FLOSS efforts? Cheers :)
Hold on a second: ... "how Greg KH should be tougher on people who contribute patches to -stable" ...
I don't know about you but bar for patches going into stable is that much higher. If people futzing about making crap code they can use it on their own servers, but when patching stable - they should fall in line or submit their code to development branches.
Adding code to the kernel might be fine if you futzing about with your own server, but I sure as hell don't want Linus & Co. letting their guard down because they might hurt someone's feelings.
I've worked with kind of Linus and I have learned from them far more than from most of considerate and gentle minded people(there are exceptions though).
"Now, I understand that Ingo Molnar and Linus Torvalds have a lot more experience in open source development than I do, but we can learn about this from scientific studies and statistics, rather than depending on unreliable personal experience."
So we should rely on your personal experiences instead of scientific studies?
Well that appears to be the exact opposite of what he said. "How odd," I thought. "I know; I'll read the blog post, and if there are no links to scientific studies, it's clear that sdse903 is making a valid point. If there are links to scientific studies, I'll know that sdse903 is either fantastically stupid, or some kind of dickhead."
Many of us have... "survived" significant periods and... "mantras" of ostensibly objective "best practices", in our careers.
Much of what is labelled as "science" turns out to be of a similar character.
Whatever its label, we tire of the bullsh-t.
I'll add my initial thought upon running across this HN title/topic: Well, if Linus is a jerk, at least he's an equal opportunity jerk.
P.S. I think it worthy of further noting that in my perception, Linus may well not be "implementing" a "reward system" and "motivation". He may be implementing a filter.
Discouraging crap, as opposed to encouraging contribution. Two different functions.
I might do things somewhat differently than Linus. But then, I'm my own, different little snowflake.
No one is suggesting they let their guard down. You can keep your guard up and still maintain quality.
> I've worked with kind of Linus and I have learned from them far more than from most of considerate and gentle minded people(there are exceptions though).
That's because it's easier to act that way for most individuals. It takes real talent to balance the two.
People really get their pants in a bunch about Linus' abrasive personality. I wouldnt want to work for him, but on an open source project that he has lead for over a decade I think its a pretty clear cut case of 'vote with your feet'.
Somehow with this management style he has managed to produce the largest open source OS project in the wild... I agree with everyone doesnt need to be a dick just because linus is but we definitely cant claim this kind of approach cant work
Linus supporters' arguments are always I'd rather have it like this than the opposite side of the coin. The idea that there is a decent alternative that is not the 180° of Linus confounds you.
Personally, I have more respect for BDFL's who can represent their community positively. Linus doesn't seem to care, so it's not like my opinion will change anything, and it shouldn't... Nobody should tell him what to do. He's an adult and should find a happy medium on his own that doesn't alienate his invested developers.
I don't know. By swearing, it tells me that he spends more time caring about their feelings. If he could just say "No" instead of "Fuck no! I'm not accept that shit!" he'd save a lot of time and energy dealing with people's feelings.
I review code frequently. When some of the code is bad, I manage to impress upon the writer that it's not up to scratch and why, and they produce better code to replace it and continue to produce better code in the future. Sometimes it's really fucked up code. Really bad.
Yet, incredibly, I don't have to use the word "fuck" at all. Amazing, I know. It's as if it's possible for someone to listen to criticism and improve their work without having to be screamed at. I have funny memories in which this happened to me as well; I wrote bad code, and then someone explained to me what was wrong with it, and then I wrote better code. Again, nobody said "fuck". Truly amazing.
>Yet, incredibly, I don't have to use the word "fuck" at all. Amazing, I know. It's as if it's possible for someone to listen to criticism and improve their work without having to be screamed at.
If it were me and he said "no", my reaction would be, "whoops, my bad". If he said "fuck no, etc", I would be like, "shit, I better not do that again".
I think the point is the next time someone goes to commit a change which fixes nothing and won't even compile, they will think twice and actually test it first due to the chastising they received the previous time they did it. Then there does not need to be a second "No". The quality of the software goes up.
If that were the case, people would know about it before hand. Linus' rants are well-known and well-published. So, pretty sure it's not having it's intended effect. Couple that with the fact that he'll rant and people who have been long-time contributors.
So, what is it really solving? Nothing for new contributors. Nothing for old contributors.
In my opinion it comes down to this, Do you want to play the game, or do you want to change the game.
If you want to play the game, Sure, be a jerk, be the biggest sociopathic jerk you can be, stamp your feet when your timelines get pushed back, force your subordinates to work extra hours, yell at anyone who breaks the build. It's the way the game has always been played.
Or you could change the game. Be the humble, helpful leader, be steadfast and nurturing, look at your employees as investments, because thats what they truly are, you are investing your product, your time, even your value into them, so they better damn well be worth your emotional efforts. Be the leaders that you see that are Charismatic and optimistic. But don't expect it to be easy.
FWIW I have absolutely NO interest in contributing to any project Linus is managing - I will not do it. The only reason is his behavior which is unnecessary and of little honor.
You're missing the point: I'm not the only one. It is obvious that people who don't mind his management style will work together with him.
To put it another way: There's absolutely NO chance that I would work for a boss with Linus' poor manners and rudeness for money. No way in hell would I do it for free. I don't think I'm the only one who feels it would go against myself-respect to work together with a boor of Linus' caliber.
The dissemination of information is a completely different thing from the respect and honor you show the person you're talking to, which is again completely different from your selection of strategy.
For instance, it is possible to be steadfast without lowering yourself and the communication to the level that Linus will do.
(And in stating 'FWIW', I know what it's worth. No need to be rude, is there? In fact, there's no need to be anything but polite. In my opinion. For what that's worth.)
I like the photo of Linus looking like a minister of religion in that suit and those glasses.
But I tend to worship at the church of statistics so right now I believe that it is all about regression to the mean until someone comes up with some hard data to show otherwise. In fact, there may be no causal relationship at all!
