Slashdot Mirror
Linus Torvalds Speaks Out on Future of Linux
SlinkySausage writes "Linus Torvalds has laid out his plans for the future of Linux, including the 3.0 kernel [there probably won't be one], problems with the Linux release cycles and which distro he personally runs on his home PC. '"Compile everything by hand" ones simply weren't interesting to me,' Torvalds says."
But really at this point, even if he stops developing the kernel, someone else will just pick up where he left off. I don't think we can ever really expect to keep one final generation of the kernel. It'll always be changing and morphing to new cpu's, hardware, etc...
cb_is_cool knows where his towel is.
The one thing I really like about Linux is that it adheres more to a Unix tradition of doing things continually and incrementally. Like, it drives me nuts that on Windows, to talk to SQL Server in C++, one has had to go from db-library to odbc to either OLE-DB or ADO... whereas, in a Unixy type mindset, one might ask, what really needed to change about db-library that required a whole new way of talking to databases? And, the answer is, not a lot. It is absolutely wonderful that in Linux there is a core set of APIs that always work, aren't suddenly abandoned to make a new feature that frankly, most people don't need.
So, in my mind, to say that there won't be a Linux Kernel 3.0 or a Linux 4.0 or something like that, is actually a GOOD THING. If you want dramatic, shocking, breaking releases that require you to rewrite 95% of your code to do the same thing, if you want to find that what you used suddenly can't work largely because it isn't supported any more, then Microsoft has plenty of that.
So three cheers for point releases, and here's to the death of "major" releases.
This is my sig.
Not to be too pedantic (yeah right), but you end up with the same product, so that analogy is kind of flawed. More like the waiter asking you 40-50 questions about how you want each part of your meal prepared, to the point where you get really exasperated and say "Just give me the damn surf and turf and don't burn it please."
This is a sig. It is like every other sig in the world, except that it is mine, and it is different.
So you're saying AMD/ATI have a worthwhile open source video driver, eh? Oh, they don't? Gee, maybe THAT is why he said that. He may very well be an Intel lover, but that statement you quoted holds no proof of it.
"If you make people think they're thinking, they'll love you; But if you really make them think, they'll hate you." - DM
It sounds like you don't grasp the simple brilliance of this. Rather than having the kernel handle these bits, forcing a one-size-fits-all approach, you instead have other teams working on this, developing the GUI, customizing it to the task at hand. Look at Enlightenment, GNOME, KDE, each one fills a need, but none of them are exclusive.
The phrase "My name is Legion for we are many" comes to mind.
Example, at work here, Fedora suits our needs perfectly. While at home, Ubuntu powers my sons desktop and Gentoo is my servers backbone. Yet, when I need to take apps from home, they run with minimal problem. They isolate the desktop from the apps that run on it, giving you infinite flexibility. Yes, it can be overwhelming. Yes, it does not look like a unified front. But by doing this, Linux can be, and is, whatever you need it to be! Hell, my gentoo box doesn't even have a monitor! I ssh in, or when that fails, I have an old teletype in case of emergencys.
Karma Whoring for Fun and Profit.
I managed to read the slashdotted article, and I think you're sorta missing HIS point. Which is that he is a Kernel Developer, not an OS developer. He's not really interested in the other parts of the system, just like you're not really all that interested in the nitty-gritty of kernel mechanics (which is more than fine, btw).
In other words...you need a lot of different kinds of people to build an operating system. Linus never claimed to be the benevolent dictator of an operating system - just of the Linux kernel. There's a difference, and there can't help but be. Related, yes, but the same thing: No, and they can't be.
Thankfully, there are many people out there who want to focus on stuff like UI design and the like. I might even disagree with Linus that such things are "fluffy", but I don't really think his opinions on that have influenced anybody any MORE toward the side of "we just like to code kernels, we don't care about ease-of-use".
So - yes, but no. Ease of use is an obstacle to widespread Linux adoption: Yes, and everyone knows that. But No, this doesn't really have anything to do with Linus.
But if Linus, as an individual, isn't interested in doing things to the GUI, why should he? There are plenty of talented coders fiddling with the GUI and desktop subsystems, why would Linus chipping in make much of a difference? Personally I think he'd be wasted working on something he didn't like doing.
Your claims about the mainstream are sorta valid, but that's not Linus' fight - he doesn't really care about it, and just wants to help make the best kernel he can on techincal merit alone. The interview gives it away - pretty much all he uses a desktop for is a web browser, most of the rest of it is CLI stuff and his usage pattern is completely different to your average desktop PC user.
To analogise, if this was an OS war for the Battle Of The Desktop, Linus would be the dude in charge of making sure the supply lines were always well stocked and that if one supply line stopped, it wouldn't cause the entire army to grind to a halt. The generals on the front line would be people like Mark Shuttleworth and Miguel de Icaza.
(Sorry, couldn't think of a lame car analogy...)
Moderation Total: -1 Troll, +3 Goat
If stability means it's dying. As I see it, the current Linux kernel does all that it needs to, and does it quite well. There is no need to upgrade it, because to do so would be an upgrade for upgrade's sake. Anyone in IT will tell you that to upgrade simply for the sake of upgrading is stupid, and will lead to a multitude of problems. The only reason that the Linux kernel would need a version 3.x is because of a fundamentally new hardware technology. Currently, software is driving hardware development; games are written requiring advanced graphics cards. In the 90's, hardware drove software development; chip makers like Intel put out a new processor, and then software was written to take advantage of the advancements of the new chip.
Even advancements in multi-core technology would not require a 3.x series kernel (unless I'm mistaken in my belief that the 2.6.x series supports multi-core CPU's), simply because once you cam make a dual-core CPU functional with the kernel, expanding that functionality to 4, 8, or even 64 cores is simply an expanding of the current code. And even if the current kernel does not support multi-core CPU's, that would be more of a 2.8.x series, rather than an entirely new kernel version.
Anybody want my mod points?
For years, people have been saying that Linux needs to focus-in on one particular distro, to make it less confusing for new users. I would argue that day has come: you can confidently recommend Ubuntu. (And, once they overcome their initial trepidation about using a new OS, they will be able to migrate to any other flavor of Linux without much issue.) I agree that Linux should keep improving. I am, however, always a little confused by the repeated calls for "uniform packaging in Linux" considering that the software installation methodology in Linux is, in my opinion, light-years ahead of Windows. With a single application (GUI or commandline) you can install any of thousands of tested, malware-free software. It's such an efficient system, that switching to the method of searching the net for a "setup.exe" of questionable origin seems like a huge step backwards.
Like many Linux users, when I first starting using it, I was annoyed at the differences and cried in frustration: "Why can't they just make it simple like what I'm used to?" With a little more experience, I discovered that there are very good reasons for doing things "the Linux way"... now that I'm used to it, I wouldn't want to go back.
like going to a restaurant, ordering your dinner, and having the chef take you back into the kitchen and put you to work making your own meal.
keep giving those million dollar ideas away and you'll never get rich....
As a Gentoo user since 1.4, I have to say that the common image is incorrect, though I know it's the common Gentoo joke.
I generally don't "compile by hand," I generally "emerge -atv (packageName)" to install or "emerge -atuvDN world" to update. Nor am I a ricer with my CFLAGS settings. It just plain works smoother than the other distributions I've used. Harp all you want to about "waiting for compiles," but I'm out doing other things while that happens. It's not as if you have to sit and watch the compiler activity scroll past. It's the computer's time being used, not mine.
Back when I was on RedHat, I'd see "package X" that wasn't part of the official distribution. So I'd find an rpm and try to install it. Then I'd find that I needed another library, and go searching for that rpm. Sometimes then things would work. But sometimes I'd find that some package was looking for things in SuSE layout instead of RedHat, or I was grabbing an rpm from somewhere that didn't play well with RedHat for some other reason. There was a non-trivial set of packages that I never could get installed and running.
On Gentoo I've had far fewer problems getting things to run. In fact, I've only had 1 intractable problem compiling from source, and that's been Doomsday, which isn't released for amd64. I've had a few transient problems with source-based packages that have soon gotten fixed. But by and large, my biggest problems have been related to binary-only packages.
Oh, and there's nothing about the usually-disruptive "upgrade to the next release." My system is just up-to-date. A few times a year they issue a new profile, but that's generally about as disruptive as upgrading any other package. The only really disruptive upgrades have been things like udev, gcc and xorg, but even with those it's better to take them one-at-a-time and cope, instead of the usual "practically everything has undergone significant changes" of a distribution upgrade.
The living have better things to do than to continue hating the dead.
> Why doesn't FreeBSD have the stigma Gentoo does? /usr/ports insipred Portage...
For starters, FreeBSD did not go around making fantastic claims about the efficiency of ports-compiled code, nor was recompiling the base system some sort of rite of passage into enlightenment about "how the system works". But mostly it was the fact that it was BSD, which just doesn't have as big of an advocacy culture.
Done with slashdot, done with nerds, getting a life.
Actually, I'm not a veteran Linux user, I've used it enough to know I don't like it. But I've had rogue application installs break a system - on any OS with any install type. Windows, Linux/apt, Linux/up2date, Linux/ubuntus-up2date-frontend, FreeBSD/ports.
It's happened on all of them to me, and you know what? The best way to get it fixed quickly is to use a distro with a FRIENDLY community which is willing to HELP and ADVISE without being condescending and pedantic. That and google is nice too...
34486853790
Connection too slow for X forwarding? Try "ssh -CX user@host"
Because the FreeBSD community did not propel itself to fame on the back of a bunch of "CFLAGS just kick in, yo" kids, maybe?
Because the FreeBSD community as well as the FreeBSD developers [i]generally[/i] tend to have outgrown the fanboyism displayed by most Gentoo followers? (I'm not knocking the gentoo devs here, btw -far from it- I'm going out of my way to exclude them from the 'fanboy' label.)
Oh, and also because FreeBSD doesn't base it's core OS on a fluxating set of packages that can -and do- hose your system on a regular basis if you try to keep it up to date (meaning FreeBSD has a binary patch mechanism instead of "make 'fuck up my system with the latest packages from sourceforge -k?' ".
Mind you, I don't run FreeBSD (haven't since 4.6), but there's a reason why people who have used Unix for a while look down on Linux in general ("it's friday -time to gratuitously change the scheduler again!") and Gentoo in particular ("more CFLAGS means more vroom!").
Nice troll, Intel graphics are fine for desktop use and don't taint the kernel.
/ATI are okay until you hit a problem like this
NVidia
This is true, it's technically illegal.
/. community a bit of a service even still, because it provides an archived copy of the article for later perusal. The need for this may be less than it was in the past, but as someone who occasionally goes back and reads "the best of Slashdot" from past years, it's frustrating to have a whole discussion archived but with a dead link to TFA, because somebody didn't think anyone would care about their blog in three years time / had a server crash / changed their URLs.
... like it should always be posted as a first-level comment, not into a thread, it should always have the subject "Article Text", should always be posted AC, etc. If you do that, it generally gets buried -- so it doesn't divert traffic from the real site (unless the site gets Slashdotted) -- but it's there for anyone who's really looking, later on.)
However, I've always felt that it does the
Slashdot is a better, more stable archive site than most places on the internet (its track record is basically as good as the Internet Archive's), so it does make a certain amount of sense to keep a copy of the article text around. It's one of those areas where copyright law just fails miserably to encourage an outcome that's useful in the long term.
(Personally I've always felt that there should be a de facto standard for posting article text
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
Yes, because the machine doesn't know that it isn't a build server that should be pushing generic x86 packages to every server on your network and therefore shouldn't optimze for quad-core Opterons.
Dewey, what part of this looks like authorities should be involved?
"It's the difference between 1/2 second to open OpenOffice 2.1 vs. 2+ seconds on one of my systems."
And that's really worth it? This sounds more like a pissing competition. I'll stick with my pre-compiled binaries thanks.
Wouldn't a load time advantage have a LOT more to do with USE flags reducing binary size on Gentoo, not with your march?
march has been shown to make (as already stated) VERY little difference on modern processors, but having USE flags that chop out large chunks of a binary that is fully loaded off disk into memory (the process you're using as your yardstick) would accomplish that.
Yet another clueless Gentoo user, I know, I used to be one. Then I realized that the 1.5 seconds I was saving getting into OO.o was far offset by trying to figure out why the latest portage tree broke wget and didn't complete.
Can't spend all my time making my system work.. rather work with my system.
My Babylon
Linux is the name for both the kernel and the OS. Richard Stallman tried to get people to call it "GNU/Linux", because he felt his contribution was more than that of everyone else that's contributed to the Linux OS. But in the end, almost everyone calls it by the name it was originally given by its creator: Linux.
Linux's UI is based on MIT's X-Windows (why not "MIT/Linux"?) and either FSF's Gnome or KDE, though, so yes, it's not Linus' purview to worry about Linux's UI, so his geekiness on that matter is not why Linux's UI is considered inadequate for desktop use by many.
E pluribus unum
Ok, so my question is:
Why, if this information is available, doesn't the "emerge" pull that info in instead of forcing the user to do it?
Seriously, this is Why Linux is not liked by many, many people. (I really want to get off Windows!)
In the above scenario, a person will be asked a lot of questions unless they put in a "magic" configuration that is readily available by parsing the output of a simple command. Why even ask the user the questions unless the parse of said command fails to return the proper CPU? Am I the only one thinking this?
Every time I start to have faith in humanity, I ruin it by driving to work between 7 and 8 am.
Well, there's a good reason not to automatically set it in this case. Maybe you don't want your binaries fine-tuned for your CPU type. For example, which I was in college, I set up a system for my room mate. It was older, and so I didn't really want to compile everything on it. Instead, I had my Gentoo computer set a CPU type slightly less than it actually was, recompiled everything, and then had his system just download the packages from me.
Anyway, Gentoo isn't the reason Windows users don't switch. I would *never* recommend a new user use Gentoo, and even though I've been using Linux for longer than every other OS combined at this point, once I started working I switched away from Gentoo since I just didn't have the time. (Yes, as easy as Gentoo is once set up, it's still not as easy as Ubuntu. Sorry guys.)
Thank God for evolution.
I believe auto detecting the CPU type is already in place on FreeBSD, but this variable is handy for compiling software for different machines using one "central" powerful machine.
If you can't mod them join them.
bork bork bork!
It's not really news. I don't think that Linus doesn't "like" Debian because of the install though... it's the whole "Debian GNU/Linux" that's probably the showstoper for him, i.e. the fact that Debian was (and is, in many regards) more directly linked with the FSF line.
I mean, real collapsable threads. Because one asshole says "gentoo" in a FP and 99% of what was supposed to be about a Linus interview goes to Hell in a handbasket. So I wish it was possible to click on a "[-]" button the second I saw "gentoo" and be spared of all this.
Nothing against Gentoo, but this was a horrible example when this would've been a really useful Slashdot feature.
i ate crayons when i was a kid and now i have two braincells and the blue ones taste nicer
When I ran Gentoo (several years ago) it was on a laptop. For 99% of my binaries, there was no perceptible difference between -O2, -O3, generic i686 or compiled for my architecture (or any combination thereof.) There was, however, a noticeable difference with -Os (compiling for size) and with not compiling against every possible library that I might want to link in. Size was noticeable (though I never ran out of disk space, so it might not have saved me much) and load times were very noticeable. But once you start using the computer, most of the time, you'll be I/O bound or waiting on user input, so the speed optimizations really don't gain you much.
The way people and organizations select version numbers has always annoyed me and Linus is spot on on this topic. To me, version numbers are stupid. The only number that really matters is the revision number, all other numbers either encapsulate too much non-sense or require too much "thinking" and "creativity" based on the developer's part. At the end of the day it's just a label to say that this version was created after all of those other versions, nothing more nothing less.
In these days, versions and releases are becoming more and more of a marketing strategy or even a get out of jail card (see Google and "beta"). The real answer is, from the first day you decide to even write a document, a version number exists and should keep ticking with every change contributed to the project. That's probably too much information that the user doesn't care about so let's simplify it to only new builds of a binary. But the build of the binary could be from various sources with various optimizations and features. Blah blah blah. So the end result is someone gets a grand idea of "NOW let's give it a version number." BS, finish it and release it or don't. End of story.
At the end of the day, the user probably only cares about a couple of things: is the newer version better than the older version (what are the new features, fixes), is it compatible with my current platform software and hardware, and finally will it break anything or do something to make me very frustrated (dependencies, deprecated features). Version numbers should only be made to address these issues, anymore than that is just marketing. I don't care if it is major or minor in the developer's minds. That tells me nothing. Make the version number useful to the users, not another confusing marketing term.
People who claim that Torvalds should be doing this or saying that should examine their own positions and honestly consider if they're not simply trying to use him and his position within the community to try and further a particular POV.
This view shows that he has not thought enough about the issues. Linux views people who have a different view as "pushing". If I were to use the same mindset, I'd say he was pushing a commercial agenda that threatens real software freedom. Because many more people look up to and will listen to Linus, whatever he advocates gets much more "push" than anything I say.
Leaving aside that a lot of what you say is unfounded crap, and so hardly likely to have any "push", Linus hardly pushes any views. He expresses them, sure, oh boy does he express them, but he doesn't enforce them on anyone or attempt to proselytise. Instead, he's talking about people precisely like you who like to froth at the mouth and produce absolutely nothing except a lot of sound and fury signifying nohing. And hell, he might well agree with you on certain points, but most likely wouldn't take much pride in the "use free software or die in fire and brimstone" nature of most of what you say.
His aversion to politics has cost him - and that's the sign of a real idealog. (sic)
Um, he avoids politics, so he's an ideologue. What are you on about?
Debian is not hard to use, even for a non technical user like myself. I'd say it was easier than Fedora in all things but adding non free software.
Are you joking? Debian is far easier to add non-free software to. I mean, they've only got a whole friggin APT repository for it hosted on the Debian servers.
Mepis and Ubuntu have excellent compatibilty with the rest of the Debian tree, so you don't lose much more than a little stability and trust for the non free inclusions.
Um, the founder of Debian, Ian Murdock, is on record saying that Ubuntu has diverged too far from Debian to remain compatible. And Mepis only recently switched back to using Debian packages from Ubuntu. So... nice try.
The thing that he should realize is that technical excellence happens when you have software freedom.
It can do, but you can't pretend that there's no technical excellence in closed software either. Both sides have merits.
By summer it was all gone...now shesmovedon. --
This "you have no business compiling..." nonsense is just baseless elitism. Your list has a lot of subtle variants, and the differences between some of them are inconsequential for most users. I've got a couple of compile jobs running at the moment (regression testing patches before I commit them to a subversion repository), and I don't know the CPUTYPE on all of the machines. The finest distinction I usually make is x86-compatible vs. PPC vs. MIPS/SPARC/Alpha/etc. and 32 vs. 64-bits. If I were compiling a high-performance numerical app, it might be worth tracking down more information (e.g. what sort of vector-processing unit is available), but it usually doesn't matter. In a given day, I might be testing my code on Linux on an x86, Opteron, or Itanium processor; on Darwin on a G4, G5, x86-compatible, or x86_64 compatible processor; on AIX/Power5, or perhaps something more obscure. I'm certainly not going to waste a lot brain cycles figuring out the difference between the itanium or itanium2 CPUTYPEs when I've got real work to do.
I'm curious (because currently there is no file system reason to purchase a Sun/Windows solution over Linux),
I've had drives formatted ext2/ext3/ntfs/reiser/hfs/hpfs/ProDOS/xfs/Fat etc. Why is ZFS any better or worse?
Do you have any benchmarks to share for ZFS or are we just supposed to assume your word is final. Unlike Windows, with Linux/BSD/FreeDOS etc. you can post your ZFS personal benchmarks and opinions and not get sued.
I'm not trying to start a fight, thanks.
Enjoy,
It's just the normal noises in here.
guessing he's not a gentoo user :)
People who compile stuff by hand when using Gentoo are using it wrong. The package manager handles all that crud transparently.May contain traces of nut.
Made from the freshest electrons.