Linus does not care about you
-
he has the final say on what goes into the tree and what doesn't
He has the final say on what goes into his tree and what doesn't.
Kernel development is distributed. You're completely free to adopt anybody else's tree as your definitive kernel source, if it suits your purposes better than Linus's does.
-
He has the final say on what goes into his tree and what doesn't.
His tree is still the basis for everything linux related. Very few people use the deblobbed-FSF blessed-GNUKERNEL (and that's just the most popular one).
Kernel development is distributed. You're completely free to adopt anybody else's tree as your definitive kernel source, if it suits your purposes better than Linus's does.
I'm not entirely sure what your point is. I was speaking about Torvalds' attitude, not the usability of linux distros (by the way Hardened Linux still breaks a lot of packages).
-
You can tear apart people's code and not integrating it into the source tree without using the mailing list as an outlet for your passive-aggresive tendencies.
...Then you watch teh drama of endless whys unfold and people get offended for being ignored, and Linus made into a passive aggressive dick again. Et voilà, instead of doing actual coding, he needs to constantly address the fucking drama queens and other easily microaggressable Untermenschen who, unlike him, have lots of time to spare.
I prefer a good dose of management by perkele instead.
-
That's because it's much harder to get in the news with a kernel feature, for example.
-
Says pretty much about journalism, doesn't it? (shrugs)
-
...Then you watch teh drama of endless whys unfold and people get offended for being ignored, and Linus made into a passive aggressive dick again.
That's a false dicothomy. Sayig "your code is in poor shape because of X,Y and Z and I will not merge it" its different from letting the project go unmannaged. In any case insulting people takes him as much time and is much less constructive. Alternatively he could just keep ignoring them.
Et voilà, instead of doing actual coding,
Except that his day consists mostly of replying to emails and merging requests, you know, instead of doing actual coding...
he needs to constantly address the fucking drama queens and other easily microaggressable Untermenschen who, unlike him, have lots of time to spare.
That's part of the job. And It's not unpaid labor either. He gets handsomely rewarded for doing what he does.
-
Well, those who pay him are satisfied with his management style enough, then. Fair?
-
The Linux Foundation? They don't seem exactly the most impartial and unbiased source in the matter. And the companies that fund Linux development do so because it's profitable for them on some level. Whether or not they care about Torvalds management they can't do much about it.
-
Okay, you're not paying Linus to do his job and yet you critique how he's doing his job.
If I were Linus, I'd tell you to go fuck yourself.
-
Yet they post on Discourse, which we all know can't just let a shade of blue be whatever color it wants to be.
-
Okay, you're not paying Linus to do his job and yet you critique how he's doing his job.
In keeping with that logic, you're not the person being criticised (Linus) and yet you're criticising my critique of him.
If I were Linus, I'd tell you to go fuck yourself.
Well, that escalated quickly.
-
Okay, you're not paying Linus to do his job and yet you critique how he's doing his job.
Ummm.... Hitler? Are you suggesting that people are only allowed to criticize their employees? Who criticizes the leader of an organization or the leader of a country? If a child gets "touched inappropriately" at school, does he/she have no ground for complaint?
-
"touched inappropriately" at school
Right or wrong, I think we can all agree that Linus overstepped on that one.
-
In keeping with that logic, you're not the person being criticised (Linus) and yet you're criticising my critique of him.
You don't have any project of a similar scale to prove that other styles of management do better. Back your critique up with some proof, or go fuck yourself.
-
You don't have any project of a similar scale to prove that other styles of management do better.
And you know this because...? Oh, right you went through all the big software projects in the world and concluded that every leader/BDFL/whatever has to insult everyone all the time.
Far more likely: you only know one example and have to restort to name calling because you're running out of arguments.
Anyway, here are a few examples: Python, FreeBSD, NetBSD, LLVM/Clang, PostgreSQL, MySQL, KDE...
-
Anyway, here are a few examples: Python, FreeBSD, NetBSD, LLVM/Clang, PostgreSQL, MySQL, KDE...
You are better off telling @wtf that has request is pointless. Now he's going to ask you to prove that these projects have been more successful than the kernel and that the difference in success was due to management style.
-
There are people running Python, PostgreSQL and MySQL on Windows and Mac, but not the Linux kernel.
Clearly they are more successful
-
Nope. The number of contributors to Python, MySQL, etc is a tiny fraction of the number of Linux kernel contributor. I asked for a project of similar scale of contributors.
-
You are better off telling @wtf that has request is pointless. Now he's going to ask you to prove that these projects have been more successful than the kernel and that the difference in success was due to management style.
It looks like I got on a no-win situation. He can allways pull more arguments out of his magic hat if I prove him wrong, though.
Nope. The number of contributors to Python, MySQL, etc is a tiny fraction of the number of Linux kernel contributor. I asked for a project of similar scale of contributors.
I doubt it amounts to a tiny fraction but it looks like you intentionally ignored my other examples. But hey, fine, lets go with FreeBSD which is a complete operating system (as in it has way more LoC than a kernel) and never had this sort of problems.
-
47 posts were merged into an existing topic: The "Wario Belgiums a Librarian with Guacamole" Topic
-
Right or wrong, I think we can all agree that Linus overstepped on that one.
That was Jared from Subway, but I can see how you could be confused. They look sort of similar.
-
I doubt it amounts to a tiny fraction but it looks like you intentionally ignored my other examples. But hey, fine, lets go with FreeBSD which is a complete operating system (as in it has way more LoC than a kernel) and never had this sort of problems.
Perhaps (SURPRISE!!!!) more than one methodology can be correct? :O
-
Speaking of Jared...
-
-
The Linux Foundation? They don't seem exactly the most impartial and unbiased source in the matter. And the companies that fund Linux development do so because it's profitable for them on some level. Whether or not they care about Torvalds management they can't do much about it.
Yes, Linux Torvalds is a giant asshole. So, if he's so terrible why not fork the Linux Kernel and develop it under some nice polite management? Because ultimately, nobody cares that Linux Torvalds is a big mean poopyhead.Does the product work? Does it do what is needed without exploding, burning down your house or killing your dog? That's the only thing people care about. All the pissing and moaning about Linux Torvalds' management style is just pointless drama.
-
-
Not entirely true. It's worth knowing whether Linux was successful due to Linus' style, or despite it, if you are going to try something similar.
Also, the "did it work" test results in some bizarre conclusions like:
- Video tapes with better sound are not preferred by consumers
- Shoppers prefer to shop at WalMart
- Everyone likes DRM
- Nokia made bad phones
-
Regardless of whether Linus is nice or an asshole, I seriously doubt it has anything to do with whether or not Linux is successful. Your own examples demonstrate that people will buy shitty products from shitty companies.
-
Perhaps (SURPRISE!!!!) more than one methodology can be correct?
It is enterily possible. In this particular case, I don't consider Torvald's methodology to be correct.
Yes, Linux Torvalds is a giant asshole. So, if he's so terrible why not fork the Linux Kernel and develop it under some nice polite management?
Because it takes manpower and nobody wants to have 20 versions of the same thing. Forking is a non-option and I don't see why people bring it up. Yes, you can fork the Linux kernel but you're not the original author, so it's unlikely that people (especially non developers that don't work with Torvalds) will get behind your version. Intel supports (as in develops drivers for) Linux because they have a potential customer market there. If very few people are going to use your own version of the kernel then there is not a huge market share there.
As an aside, people have forked it. As I mentioned before, the FSF mantains its own deblobbed fork (Libre Linux or something like that) but they still have to rebase torvalds changes now and then.
Because ultimately, nobody cares that Linux Torvalds is a big mean poopyhead.
Nobody being...? Those who are on the recieving end of his rage usually do. Some developers even left the project.
Does the product work? Does it do what is needed without exploding, burning down your house or killing your dog? That's the only thing people care about.
I'm sory, but I fail to see your argument. You're essentially saying that we shouldn't criticize Torvalds because Linux "works"? So do a lot of other products (in fact nearly all of the products that work), its not really far fetched to ask for a product to actually work the way it was intended.
All the pissing and moaning about Linux Torvalds' management style is just pointless drama.
And all this "you can't even discuss about Torvalds management style" is tiresome. Apparently you are taking it as some sort of personal criticism (which it isn't) or trying to end the discussion by repeating that "it works". We get it, it works. It boots on a variety of hardware and is able to execute applications that work on top of it and consume the services it provides. Like literally every other operating system in the world.
-
I doubt it amounts to a tiny fraction but it looks like you intentionally ignored my other examples. But hey, fine, lets go with FreeBSD which is a complete operating system (as in it has way more LoC than a kernel) and never had this sort of problems.
What are we comparing?
The BSDs are weird as they started out as a single product (BSD up to 4.4) and forked from there.
Some of these branches were created due to differences between the project leaders as to how the project should be run (See: OpenBSD's split from NetBSD).
-
Speaking of which, Theo de Raadt is at least as much of a dick as Linus is and no-one cares.
-
trying to end the discussion by repeating that "it works". We get it, it works.
And perhaps if Linus were not such a flaming arsehole, his pet project might attract skilled developers from the set who refuse to work with flaming arseholes on principle, and some of those developers might contribute code that makes it work better.
Or perhaps the willingness to be and/or work with a flaming arsehole is actually positively correlated with the ability to design and implement optimal kernel code. Seems unlikely to me, but weirder things are true.
But as long as "meritocracy" continues to mean "everybody agrees to pretend that interpersonal relationships couldn't conceivably have any effect on code quality" in the minds of most developers, we'll probably never bother to find out.
-
Theo de Raadt is at least as much of a dick as Linus
There needs to be some kind of flame war / UFC cage fight mashup where propositions like this can be put to a public test.
-
It is enterily possible. In this particular case, I don't consider Torvald's methodology to be correct.
Just asking... How many people have you ever managed? Do you speak from experience or do you just pull your arguments from thin air? If you have different actual experience managing a team, please tell a story.You know, you are asserting something, and if you assert something, the burden of proof is on you.
I stand by my point that when a product is being developed healthily, there are two things to care about:
- The product gets developed
- The discussions revolve around code and code-related stuff
Note how unmentioned goes pointless bickering about "the leader said 'fuck' again! How dare he even!"
-
But as long as "meritocracy" continues to mean "everybody agrees to pretend that interpersonal relationships couldn't conceivably have any effect on code quality" in the minds of most developers, we'll probably never bother to find out.
The linux kernal community has more thank likely scared off a lot of talent but the thing is is that this is Linus's baby. Again and again he has shown that he doesn't care much for the opinions of people who don't agree with or conform to his special brand of shoulder alien logic but he has surrounded himself with people who do. That is the linux kernal community and if you want to be part of it you either conform to the community or fuck off. Does anyone really have the right to go into a community and demand that things change to suit them. Especially when the option exists to create your own community with your own brand of shoulder alien logic even with the same codebase.
-
Does anyone really have the right to go into a community and demand that things change to suit them.
Sure. There's no reasonable basis upon which to deny such a right.
Of course, the community has an equally valid right to deal with those demands in any way that suits the community. It might even be to the community's advantage to think them over seriously before summarily rejecting them as irrelevant, especially if the changes demanded are reflective of norms in the wider community.
That said, I do think the only workable methods for bringing about a cultural change in the core Linux development community would need to be built on contributing and/or curating bug-free code at a higher rate than Linus does.
-
especially if the changes demanded are reflective of norms in the wider community.
You had me until this. wider community is quite a nebulous term and is undermined by the fact that Linus doesn't appear to care all that much about the linux world outside the kernal. He's really only interested in his baby and the community he's directly involved in. Wider community could extend from different distros to me complaining about linux. Also why does the wider community get to enforce it's will on his community. If his community is so toxic then have the courage of your convictions and exile it.That said, I do think the only workable methods for bringing about a cultural change in the core Linux development community would need to be built on contributing and/or curating bug-free code at a higher rate than Linus does.
Which implies that the best talent is staying and thriving in his community.
-
why does the wider community get to enforce it's will on his community
Never suggested it should. But if a subcommunity has norms and values distinctly at odds with those of the wider community of which it is a part, it risks becoming a decreasingly relevant little cult; this is something that any subcommunity, it seems to me, would benefit from thinking over from time to time.
-
Which implies that the best talent is staying and thriving in his community.
Not necessarily. It might well be that the best talent considers Linus's little community to be somewhat irrelevant.
The higher the wall that a subcommunity chooses to build around itself, the harder it's going to be to work out which of these propositions is closer to the truth.
-
Never suggested it should. But if a subcommunity has norms and values distinctly at odds with those of the wider community of which it is a part, it risks becoming a decreasingly relevant little cult; this is something that any subcommunity, it seems to me, would benefit from thinking over from time to time.
If they want to continue in their ways let them be. Linus has carved out his own little niche and they deserve to have their community with their own norms without outside interference. This nebulous wider community can just let this decreasingly relevant little cult just die off.
-
If they want to continue in their ways let them be.
As the kind of arsehole who is always amused by having the opportunity to call another arsehole an arsehole, I'm glad Linus exists.
Now where did I put that video I've been meaning to inflict on Blakey? Ah yes. Here it is.
https://www.youtube.com/watch?v=4XpnKHJAok8
-
You're an inhumane monster only one step higher then people who made that video. Fucking hell an entire hour on git. Surely we're looking at a Geneva convention violation.
-
What are we comparing?
How do projects with a large amount of code and contributors deal with bad code.
The BSDs are weird as they started out as a single product (BSD up to 4.4) and forked from there.
I'm not sure what you mean by "weird". Forks happen everyday in the software industry. Those in particular happened 20 years ago.
Do you speak from experience or do you just pull your arguments from thin air?
My personal experiencie is irrelevant as you have no way of confirming if what I'm saying is true or not.
If you have different actual experience managing a team, please tell a story.
We delivered a project on a deadline. Whithout insulting each other. It didn't suck. The end.
You know, you are asserting something, and if you assert something, the burden of proof is on you.
I showed you projects that are actually well managed without needing to shout at their contributors. In other words I presented the evidence you asked for, you just keep moving the goalposts so that it's impossible to have a discussion.
I stand by my point that when a product is being developed healthily, there are two things to care about:1. The product gets developed2. The discussions revolve around code and code-related stuff
That's not what "developed healthily" means. That just means it gets developed which, again, is not very impressive.
Speaking of which, Theo de Raadt is at least as much of a dick as Linus is and no-one cares.
Does he shout at contributors? Because so far I haven't seen him do that. Feel free to point me to the OpenBSD mailing lists though, I might be wrong (which is why I intentionally left OpenBSD out from my list).
-
-
Also why does the wider community get to enforce it's will on his community.
Because they're the Good Guys™.
-
That's funny because it's the wrong question.
What do we want?
TIME TRAVEL
Where do we want it?
HERE
-
It bothers me that most people seem to assume that if time travel exists /will exist /has existed it is /will be /was without limitations.
"The proof time travel isn't possible is that we don't see any time travellers visiting us."
No, you idiot. Just because you have a car doesn't mean you can drive to the middle of the Pacific.
-
Just because you have a car doesn't mean you can drive to the middle of the Pacific.
I rest my case.
-
@Zecc said:
That you're the kind of moron who would attempt to drive to the middle of the pacific in an amphibious car?Just because you have a car doesn't mean you can drive to the middle of the Pacific.
I rest my case.
-
Amphibious vehicle - Wikipedia
