Slashdot Mirror
Torvalds Tells All
Eugenia writes: "Linus Torvalds gives an interesting interview to OSNews.com, talking about everything people are wondering about his personal opinions on several matters: on the GNU/Linux naming, the GUIs currently offered for Linux, the kernel 2.6, his plans for hot-plugged devices & drivers, Microsoft, FreeBSD and the future in general."
(emphasis added)
Now that has GOT to hurt. The guy that tons of geeks look up to (rightly or wrongly), has just said that he doesn't really give a rats ass about what one of the Big Names keeps going on about.... Definately not what anyone in a philosophical debate wants to hear - people loving your idea is great, people loathing your idea is still something you can work with, but disregard? Ouch.
It seems to me that the more interviews I see from Linus, the more tired he sounds, or is exasperated a better word, anyone else noticing this? I think everyone in the UNIX community would like to see real answers to the questions in regards to .NET, and "competing" softwares. He even dodged the "Where do you see Linux in 5 to 10 years" question. Maybe he took some advise from Steve Jobs and decided not to be a preacher. :)
Linus Torvalds: "I'm a big non-believer in manual driver and kernel configuration, be it visual or not. Most of the stuff happens automatically, and we're going to make that more and more common. Things like hot-plugging a device and the driver automatically getting loaded is how things are supposed to work, none of this "device manager" stuff."
That is very good news for the eventual acceptance of linux on the desktop. Allowing users the ability to hot swap devices and not have to reconfigure the kernel for new devices will be a huge step towards mainstream acceptance, and it's good to see Torvalds is looking that way.
Linus Torvalds: See my answer about not caring what the competition does, but doing my own thing as well as I can..
Ellsworth Toohey: "Why don't you tell me what you think of me?......."
Howard Roark: "But I don't think of you."
Somewhere, something incredible is waiting to be known. -- Carl Sagan
Meanwhile, in an RMS office somewhere:
"Okay, so Windows 2001 it is then."
Asikaa
Come in, twenty-seventy-seventy, your time is up.
I get a picture of Linus taking his cue from Chunk, and telling everything.
"And then, when I was in 4th grade, I pushed my sister down the stairs and blamed it on the dog.
But this, this was the worst. I mixed up a batch of fake puke, and snuck it into the movies. I went up into the balcony...."
Now that would be a great interview.
--
"Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next.
Then along comes a guy who doesn't care if anybody adopts his pet project -- which is now the used by a huge number of people all across the planet, and the basis of billions of dollars in development and sales efforts. Hardly fair, is it? ;)
Don't want to sound like a jerk, but:
:-)
>They have binary modules, and I don't have the source to them
Who's mistake is that? Live by the sword, die by the sword, so to speak. Living the proprietary life can be more expensive than most would care to admit...
I personally wish Linus would break binary modules much more often. That way companies would have two choices: Support Linux properly (via Open Source) or go away. I'm more than tired of seeing "Linux Support" on a box when they include nothing more than a crappy binary module. That's almost false advertising: It should say "Supports RedHat running Linux kernel 2.x.y-preZ only" on the box instead.
I no longer buy hardware for my Linux box unless the support is via source code. Even then, I try to avoid certain hardware unless its built into the kernel source tree (I'm not fond of patching the source for storage controllers and the like...). Doing that has kept me sane, and reasonably happy.
I guess that opinions' a little too hardcore. See you in -1 land!
Yeah this one caught my eye too ... he slagged a whole bunch of OSes after saying he didn't follow them. Its pretty tough to know if something is techinically interesting if you don't follow any of the developments on it. If you don't know anything about the OS, just say that.
I forget...are we at war with Eurasia or East Asia?
What if there is no "competition"?
Competition between MS and Linux is an invention of the markets, not a feature of the kernel's existence. At least, I think Linus is right if he thinks as much.
~Tim
--
Rushing on down to the circle of the turn
Back in '92 when I first started working with Linux it was kind of cool. You could do things on your home computer that before were not very possible, or very expensive.
It was just kind of cool, and fun.
Then sometime in '97, shortly after the OS/2 regime was destroyed, Linux took on this holy jihad. Now it was a battle, it wasn't just good enough to create something kind of fun and geeky, the goal was to destroy all the infidels from Microsoft.
It was at that point that Linux became no fun to use, and it was no longer fun to be around the Linux geeks.
Linus has the right attitude. There is no enemy.
[yawn] I'm so sick of people quoting "The Art of War" and "On War" and "The Book of Five Rings" and other military classics in reference to software development. First of all, as several other posters have pointed out, L.T. sees himself primarily as a programmer, not a businessman -- he doesn't define other OS'es as "the enemy" and therefore doesn't worry about ancient military wisdom. Second, and perhaps more important, even more business-oriented programmers are fools if they think military advice translates to any business, especially software. No matter what the Japanese say, business _isn't_ war.
Whatever happend to that fabled Japanese "business is war" economy, anyway? Oh, that's right -- all those warrior businessmen had a couple of decades of success with their slash'n'burn tactics, then kept going with it and drove one of the world's largest economies straight into the toilet.
There's a lesson here, one which Microsoft and Oracle and Sun should learn really fast: war is about killing people and breaking things, and business (ideally) is about empowering people and building a stable, lasting structure to create good products. These are not only different goals, they're opposite and mutually incompatible goals, and techniques that work for one simply _do not work_ for the other.
I've seen this from both sides, by the way -- I was in the Air Force when A.F. leadership went through a "TQM" craze. It didn't work worth a damn then, and "Sun Tzu's Guide To Crushing The Competition In The Global Marketplace" doesn't work now.
The correlation between ignorance of statistics and using "correlation is not causation" as an argument is close to 1.
It's not great unless you're willing to assume that every great idea and innovative feature that will ever be invented in the future will only come from your own developers.
While I think that ryanf's comments and quotes from Sun Tsu elsewhere in this thread completely miss the point, Linus' head-in-the-sand attitude to the many great and exciting developments that are taking place in other operating systems is a shortcoming not a feature.
There are people in this world who are at least as clever as Linus and they will continue to come up with new and useful ideas and refinements to operating system design. Pretending that these innovations aren't useful to you is not the best way to improve your kernel.
I have to admit that I haven't been following Linus's interviews too closely as of late, but I do remember reading in 1996 or 1997 (when I first tried to install Linux) about why he created it; he did it for himself.
He wanted UNIX for his PC because he thought DOS was crappy.
He had a lot of people appreciate his idea and even make him a Geek Icon. Hey that's pretty exciting stuff for a young geek to have lots of other geeks look at you in awe.
Eventually the reality of what you are doing sets in. It's not a hobby anymore and you are not doing it for yourself anymore. People depend on you to run their businesses, they want you to lead an OS holy war, so to speak.
Eventually you either let the crowd push you to insanity, or you have to decide not to care what everyone is screaming at you, and you have to remember why you started all of it in the first place.
Linus is right, though, he shouldn't really be caring much what everyone else is doing. Linux should be it's own product and not the "me too" product that it has become.
Not since Marie-Antoinette played milkmaid has looking simple and honest been so fake and complicated.
So he annoys the faithful by proving himself a mortal?
I never had the misconception he was out to slay the evil Microsoft or other such competitors. He has always been "the author of Linux", nothing more, and certainly nothing less.
This interview simply confirms it, he really is just trying to make it better. He isn't at WAR with anyone, he isn't into that grandstanding.
Maybe a few people here could take a lesson from his interview. Then, maybe you might know what it is all about.
* Winners compare their achievements to their goals, losers compare theirs to that of others.
You gotta love the focus.
.Net, Hailstorm and other M$ attempts at digital domination:
Linus on the competition:
I don't actually follow other operating systems much.
Linus on
See my answer about not caring what the competition does
Linus on Linux vs GNU/Linux:
I don't mind what rms calls the system...I really couldn't care less.
Linus on the marketing of Linux in the years to come:
I don't use a marketing eye, I simply don't care.
Linus rules the kernel, and the kernel is good. His ability to avoid distraction, rhetoric and bullshit is highly commendable.
Of course if we all had that kind of focus then slashdot wouldn't have any comments, now would it?
m00.
I worry about Linus and also Linux. I feel like Linus is trying to disassociate himself from Linux because he has two dynamics at work inside him.
1. Linus realizes that he really is the leader of a large and idealistic movement, and would like to see the Good Things(tm) keep rolling.
2. Linus either feels that he is not the man to lead, or he realizes that he cannot be the leader forever.
His reaction is unfortunate. If he really does want things to keep rolling, he needs to provide for a sustainable method of succession of power.
Linux is a religion these days. Really. It may not have gods, but it has a fiercly defended ideology that really does border on the metaphysical.
Human knowledge is libre is not so much a radical notion, but its particular application to technology is very radical--bordering on the spiritual.
Look at all the major world religions. They have all suffered at some point due to the schisms created by lack of smooth power succession. These problems are inherent to systems where there is one guru.
I hope I'm not decending into troll territory here, but the FreeBSD core team idea is a very good one. There are no succession problems, and the team seems to deal well with changes in staff despite the smaller numbers of people working on the project.
-Peter
. Penguins Surely Ca
Plato: Linus, do you agree there is a human nature?
Torvalds: You know, I could care less. I don't think anything is going to really change if we discover there is indeed a human nature.
Hobbes: But surely you must account that people do what they do to serve their own ends?
Torvalds: Again, see my answer to human nature. It just doesn't matter to me.
[end philosopher round robin]
The thing is that there is an incredible difference between Torvalds and Stallman. Torvalds told us he isn't a big thinker. Stallman is. Insert Stallman in the above conversation he would definitely give the big thinkers something to argue about.
The reason there so much more contraversy over Stallman than Torvalds is because Stallman allows us to disagree with him. You can't disagree with Torvalds point of view because he doesn't have one. Stallman's view of human nature is directly involved in what we consider today free software. Just like the US Fathers of Constitution view of democracy is directly involved in what is today the United States.
I argue that those of you tuned to your computing terminals without thinking of the big picture--the so called pragmatists--that you have no way of arguing against those who do. And I plead you to not argue when you really don't know what you are talking about.
Quite frankly, Linus is writing code. He is contributing to Free Software.
The vast majority of Free Software advocates are exactly that - advocates. They aren't developing code, they aren't reading the source to make improvements.
Free Software isn't about anything philosophical. It's about software and being able to share it to build upon itself. After that, it doesn't matter.
Aside from some work in the HURD, RMS isn't a software developer anymore. He has become a philosopher, trying like Socrates to convert others to his way of thinking.
Linus is coding, creating usable technology. It's HIS technology that acted as the catalyst in the Free Software world. Without Linux, GNU would still be a rather obscure name that many computer scientists don't even recognize. Sure, the GNU tools allowed Linux to start off sooner, but there was nothing special about the GNU tools at the time Linux was created -- save that it was free (gratuis), and our beloved Finn could afford them on a student's budget.
The coders have the right to make the names and use them however they please. The philosophers are only being hypocritical by making any attempt at changing that.
Linux doesn't owe GNU anything. The GNU project gave Linux a tiny stepladder. But Linux gave GNU a Saturn V Rocket.
-- Sometimes you have to turn the lights off in order to see.
On the other hand, if you aren't strictly dependent on specialized hardware, then scrap your old stuff and spend a couple of bucks to buy modern hardware that IS supported.
Your Servant, B. Baggins
Changing the API often will result in companies that *do* opensource drivers to drop support as well. If there's a valid technical reason to change API's, great - go for it. Otherwise, you're pulling the exact same kind of crap that Microsoft does to try to force people to use their platform and that people on this forum regularly complain about. If I were a product manager and got told "you're going to have to have a developer spend 2 weeks rewriting some code because some guy decided to break the API for no good reason", I would *not* be a happy camper.
Go Badgers! -- #include "std/disclaimer.h"
While the BSD people seem to have this massive inferiority complex and make a lot of noise about how great BSD is and how shitty Linux is, most Linux people, save the trolls on Slashdot, really do not give a shit. They are mostly agnostic. They don't feel the need to say that Linux is superior. It works for them, it gives them something to tinker with and the atmosphere around the Linux crowd is generally more relaxed -- thus fostering creativity.
I work at a company where we use a bunch of OSes. Windows, Linux, FreeBSD, Solaris, NetBSD, OpenBSD you name it. Usually the OS for solving a given task is chosen depending on what the people who develop, deploy and run the service are more comfortable with. Sometimes it comes down to particular things that one OS does better than the rest.
For instance we use a lot of Linux machines for development work. Why? Because there are more tools available under Linux. Commercial software vendors create things that just aren't available under, say, FreeBSD. Sure you can run them in Linux emulation, but why would you if there os no reason to do it?
Other than being a good BSD zealot and not soil your soul with the unclean and ungodly Linux.
We deploy a lot of solutions that were developed under Linux on FreeBSD machines. Large scale stuff. A lot larger than anything you are likely to see during your entire carreer in UNIX. Using FreeBSD during deployment is a cost issue for us. If you save a few million dollars using FreeBSD for deploying a solution because some aspect of the OS would require more hardware if you ran Linux, then you do that. I can't remember any of the Linux users in the company bemoan this fact.
Likewise, if you can cut development time in half because you have better tools under Linux it would be stupid to use FreeBSD; just because it is the Right OS.
This sounds pretty obvious, right? Apparently it isn't. The last year I've seen two individuals apply for jobs here who wanted a clause in their contract that they wanted to ONLY deal with one OS. (I'll leave it as an exercise to the reader to figure out which OS they preferred exclusively). Given that the BSD crowd is more prone to childish zealotry it didn't really surprise me.
It told me something important though: you really do not want zealots working for you, whether they are one denomination or another. If you can reduce the number of obvious shitheads in your company that is a good thing.
I think the best thing that could happen to the BSD world would be if someone well respected within the communities would step up and tell people to quit being such whiners. It is embarassing to see grown people who are supposedly intelligent act in a way that makes them look like ignorant bigots. That's not to say that Linux doesn't have the same problem; sure it has, but to a lesser degree. Most chest-thumping Linux users are just that "users". Clueless losers akin to the Amiga losers who claimed the OS of the Amiga was the best OS in the world -- but couldn't really tell you what made it so much better except the usual drivel that set it out from MS-DOS 3.2.
I've met Linus once. A few years ago we had dinner and an evening of talking about this and that. What strikes me about Linus is that this guy is probably the best leader you can get for any software project. He is focused and rational.
He is focused on what he wants to accomplish in a forseeable future and isn't easily led astray by fads or hype (unlike most people).
He is rational in the sense that he doesn't give in to emotional pressure but bases his decisions on what he thinks is right. This is important. I have managed open source projects and one of the things that I find very hard is rejecting bad ideas, bad code and bad people when the intentions are good. It is really hard to do. (Tridge [of Samba] said the same thing in an interview not too long ago). This is one of the things Linus does well. His level-headedness and his apparent lack of passion (apparent being the key word) when he reaches a decision is really something that other people could learn a lot from.
Now instead of flaming me: if you are a "guilty as charged BSD chest-thumper", ease up a little. If you are basically a loser who never wrote any significant piece of software or even tried to contribute with some actuall skills that you have: grow a brain or at least try to put some work into maturing your intellect and keeping your passion restrained long enough to stop bullshit seeping out of your face.
Thank you for reading.
Softupdates? KSE? SMPng? KQueues? They're all worthy of discussion, and not only that but Linus has discusses kqueues on linux-kernel in the past, and while putting down the BSD interface as being over complex, he hasn't managed to get any similar into Linux last I checked. I'd really like to hear what Linus has to say about KSEs vs. clone() as well. And SMPng is doing some very interesting things with giving interrupts a context so that you can use adaptive mutex locks in them to increase scalability -- I'd appreciate hearing Linus' opinion on those as well.
I have a bad feeling, though, that Linus would take his usual tack of being casually dismissive of what other OSes do, while not really adding anything useful to the larger ongoing discusssion. And I'm sorry if people feel that statement is flamebait, but I've read linux-kernel and seen Linus behave this way. He needs to mature a bit and give credit to other people's work, even though he might disagree with it.
kernels are essentially a solved problem, and future interesting stuff will be going on above the kernel level, not in it.
That's incredibly naive. There's a lot of interesting stuff still to do in kernel development. If you think that kernels are "finished" maybe that is because you're spending your time in the Linux world too much?
You know i have read the article and many of the comments here and it brings to my mind a valid point i think is worth sharing. I use linux at home and work, the machine beside me here is Slackware 8.0 Running apache and my notebook at home is Redhat 7.1, i love linux for many reasons, the power it has, the sheer amount of software available, the open and intelligent way so many developers act.
But yet is still use MS for my main system at home (At work i have to and i dont have a problem with that). Why ? I will tell you.
I for one am sick and tired of attempting to install applications etc and finding out that this version of XXX is wrong and you need XXX (GCC 2.69 is a good example) you download a package and try to install it only to find out that you dont have this library etc etc, so its download, configure, make in an endless circle.
Last night i wanted to simply install a Div-x Player for Redhat so i can take my home notebook away on holidays, so i downloaded Mplayer and XMPS, whoops dont have this library, go get it, then install, ok, no didnt find the codec, reintsall, nope still an error and i dont have the patience to find it right now, so i try Mplayer - only it wont work with GCC 2,96 which it says is redhats version so go off and get 2.95, copnfigure, make, nope wont work, so i need to read the documentation - only its no help (not im not an amateur here - i have worked with Unix, Macs, Novell and MS products for almost 10 years)
So i thought damn it i dont have time for this, thus i pulled out my Linux HDD and bunged in my Win2k one, installed the codec and im off an running. Easy - took 3 minutes
Why is it so hard to get this level of funcionality in Linux ? The OS has been around for longer and with open source it should be easier ? The thing is Redhat and Mandrake with Gnome and KDE are getting closer and closer to the point where they can be a desktop replcament - Koffice is very good - but as long as installing ANY software requires arcane knowledge of terminal windows and make files and lib dependencies it will never get there - the average home user (ie 90% of them) wont do this.
The problem is this whole MS V Linux crap detracts from the real issue, making linux the best OS for all users and that means functional and simple - this is what is missing here and it breaks my heart.
To Quote: Linus Torvalds: I'm a big non-believer in manual driver and kernel configuration, be it visual or not. Most of the stuff happens automatically, and we're going to make that more and more common. Things like hot-plugging a device and the driver automatically getting loaded is how things are supposed to work, none of this "device manager" stuff
This is what we need - not another (im gonna get flamed for this but - MS Sucks Linux Rules argument)
A few things i think might be self evident now but i will say them anyway (im asking for it here i think but prove me wrong instead of flaming me)
1. There will always be an MS or MS like desktop- in a world without MS then how do we introduce new people to systems (give me a break i could not give my mother linux - i dont have unlimited support time) THERE IS A THING CALLED CHOICE
2. linux will never destroy MS - this is as stupid as the 'war on terrorism' setting out to destroy puts an enemy on the defensive AND people will continue to use MS products
3. The only way to Win market share is to produce a friendly and superior product - i think i have already commented on this.
I Love Linux - i think its an incredible OS and is getting better - but that does not mean i have to hate MS (its an operating system - get over it)
I refuse to argue with Anonymous Cowards - if you want a discussion get an account....