Slashdot Mirror
Linus Torvalds: 'I Still Want the Desktop'
darthcamaro writes: Linux has clawed its way into lots of places these days. But at the LinuxCon conference in Chicago today Linus Torvalds was asked where Linux should go next. Torvalds didn't hesitate with his reply. "I still want the desktop," Torvalds said, as the audience erupted into boisterous applause. Torvalds doesn't see the desktop as being a kernel problem at this point, either, but rather one about infrastructure. While not ready to declare a "Year of the Linux Desktop" he still expects that to happen — one day.
If he waits a little longer, he can probably just take it without anybody noticing.
Torvalds doesn't see the desktop as being a kernel problem at this point, either, but rather one about infrastructure.
Is this a kinder, gentler Linus saying that it's everything but the kernel's fault Linux isn't on the desktop? Sounds like it to me, but I will have to see if I can watch the whole takl to get the correct context.
Linux has so much going for it in the device market that I don't see why Linus doesn't just double down on it. The future of Linux seems to make more sense as a kernel used for other things (like Android) rather than trying to break into the standalone desktop OS market.
Working out of a coffee shop - just hit the slashdot page when one of the passer-bys looked over my shoulder and said "Slashdot? Is that site still around? Are they still talking about the Year of Linux on the Desktop?" ... and then we noticed the first story simultaneously...
perhaps you can enlighten us as to why he's wrong, and what the linux kernel has to do to better support desktop environments?
I'm god, but it's a bit of a drag really...
I think Chrome OS or Android is the only way to go. Both Apple and Microsoft is trying to go in the same direction, and hide all the arcane intricacies and really simplify the computing experience for the common computer user. To varying degrees of success, I must admit, but I think it's the way forward for most of the users.
- Henrik
- when the Shadows descend -
Would also require that people be able to run most of the apps they want in Linux. Note that though this has long been a problem, the increase in web-based apps is slowly eroding the relevance of any specific OS. Even for games, though the quality of web-based games will always be inferior. And (nearly) everyone likes to play games.
Don't waste your vote! Vote for whoever you want, unless you live in a swing state it won't matter anyways
All Google has to do is dump that stupid steaming pile called ChromeOS, and admit that Android wins. A desktop customized version of Android (complete with a real desktop) is still based on Linux (at least Google's fork of it), already has hundreds of thousands of apps, and could be better in nearly every way than Windows or Mac OS-X in 2 years, IMO.
The other broken OS, GNU/Linux, needs a major overhaul before it will ever be popular among anyone but geeks who are willing to accept that their OS is hostile to sharing new apps, or too blinded by fan-boy-ism to notice. I write this from my Ubuntu laptop, where my code contributions are far lower than Android or even Windows, even though I put in most of my effort here. It's just easier to publish an Android app. It's even easier to publish software for Windows. If Mark Shuttleworth were just a bit smarter, I think he'd realize he needs to abandon managing .deb packages and start this whole mess over based on a more git-like aproach. He's done a lot in that direction - user PPAs for example, but it's still not there. No RPM or .deb based Linux OS will ever become the basis for the Year of the Linux Desktop.
Celebrate failure, and then learn from it - Nolan Bushnell
And I want a week long orgy with the Victoria's Secret supermodels, but I'm intelligent to know the likelihood of that happening is pretty damned small. Linus should be exhorbitantly happy Linux has made the inroads it has in the server and mobile markets. Desktop, if it ever does follow, will probably not resemble "desktop" as we now know it.
In the end they will lay their freedom at our feet and say to us, Make us your slaves, but feed us. - Fyodor Dostoyevsky
A stable binary driver interface would help for starters.
My desktop computer at home is running Linux for more than a decade now.
factor 966971: 966971
perhaps you can enlighten us as to why he's wrong
I never said he was wrong... Only that he's true to form..
"File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
Been using Linux exclusively as a traditional desktop for over a decade.
Today the problem with Linux taking over the Desktop is mostly bundling linux and apps.
We Are the Linux fanboys.
You Will be Assimilated.
Resistance is Futile.
- Linux fanboy :)
The day that the various desktop environments decide to cut out the middlemen. When I can go grab an official KDE install disk that gives me a polished KDE experience with the latest kernel and Wayland from kde.org, that's the day Windows will start really hurting. Then I can say to my relatives "Linux? Just go get KDE" and there'll be no confusion anymore. If it's KDE compatible, it's KDE compatible. Load the binary, off you go. Just like OS X and Windows.
I've been using Linux as my desktop system successfully and happily for over 10 years. My productivity, aside from a few small areas, beats anything I can accomplish on an MS Windows machine.
What, then, am I missing? What is this "Linux desktop" that everyone claims is not yet here?
Neither are going to happen, so move along and focus on something that CAN happen.
Stable ABI at the very least.
perhaps you can enlighten us as to why he's wrong
I never said he was wrong... Only that he's true to form..
So he's right, but for the wrong reasons? How do you know when he's right for the right reasons?
The road to tyranny has always been paved with claims of necessity.
All someone needs to do is make a F/OSS version of the XP desktop - with all the functionality.
User: "Gee, I can have an up to date 'Windows XP' like computer without having to buy a Windows 8 machine? Sign me up!"
And I mean, they start it up and the desktop is almost identical to a XP machine - the 'almost' is for not being able to use the 'Microsoft' name or other trademarked names.
Figure out how to stop the need to drop into command prompt and mess with Grep etc... we can talk desktop after that.
So that device makers can start spamming users with crappy binary drivers that break all the time so we end up in Windows land all over again? I think your suggestion is a really bad one.
Linux "won" mobile in the same way Michael Moore "won" the war on anorexia.
Copyright (c) 1990 - 2014 Dice. All rights reserved. Use of this comment is subject to certain Terms and Conditions.
> A stable binary driver interface would help for starters.
No. Probably not. Lack of stable interfaces never harmed the WinDOS market. That's because these kinds of things are driven by market share and have little to do with "platform quality". Either a vendor thinks the market is large enough to bother with or not. The "level of bother" factor is largely irrelevant.
That's why much software is still Windows-only despite there being a mythical commercial platform that's supposed to do everything right.
A Pirate and a Puritan look the same on a balance sheet.
... a dead joke, isn't it?
Linux is for people who don't mind RTFM.
As the GP said though, what does that have to do with Linus and his kernel?
I'm starting to think GNU is the problem with "GNU/Linux" these days.
Successful desktop operating systems have been based on various kernels. Apple used a pretty crummy one before switching to a BSD derived one. The Atari ST and Commodore Amiga each used their own, and they had certain success in their niches.
The problem is the GUI. People don't like X, and Linux people have no desire to give us anything else. Engineers and enthusiasts may well argue that it's better from various objective reasons but the end user doesn't care. They use it and they think it sucks. Perhaps the problem is that it still pretty much needs the shell. Perhaps it's large, slow and clunky. Perhaps it's the poor support for games.
Android doesn't have these problems because the developers didn't cripple themselves with X. TiVos and Tomtoms (before switching to Android) used Linux without X and people were quite happy with them.
Give us a nice, simple, standard GUI without a bazillion customisations, and with the ability to to just install an app from the GUI and run it from the GUI, and Linux might actually work on the desktop.
I had linux on my desktop for several years at one point. I wasn't a huge gnome fan, but kde 3.x was a pretty decent experience. Then gnome and kde nastified their GUIs and I was left without a decent GUI. Xfce was close, but it's not mature enough in my opinion. So windows 7 it is. I still use linux on servers.
Buck Feta. You know what to do.
It's GNU/Linux's fault. Android, still based on Linux, could likely win the desktop if Google got their act together and stopped pushing ChromeOS. Notice how my binary applications run on *very* many Android devices without recompilation, even when I write in C using the NDK. Notice how Android does not introduce bugs in my applications by swapping in a buggy shared library which I never tested. Notice how nearly impossible it is to publish a GNU/Linux app in comparison. In one case, you just publish your app to Google and wait a day or so. Notice how my app simply installs in a comparitavely secure jailed directory rather than having to disperse crap all over the file system. For Linux, you need to write and test different and binary incompatible installatoin packages for RedHat, Arch, Debian, Suse, then wait a few years for your package to be accepted and migrate from unstable to testing to stable, and even then you don't run everywhere.
Just freaking stupid.... year of the GNU/Linux Desktop my butt!
On a completely unrelated note, WTF is up with the new slashdot site? I had the newly dumbed-down ads disabled with a check-box. The check box is gone, and the ads are back, and dumber than ever! I miss the days of Barracuda ads that made sense on slashdot. The new ones aren't targeted at geeks at all.
Celebrate failure, and then learn from it - Nolan Bushnell
There are too many linux distros for desktop to work good also package maintenance needs to be more windows like.
Where as you can install 3rd party apps with out braking stuff / have them be able to point to / bundle with the needed run times with out them needed to be part of the main app store / package manger system.
As well being able to have things like use ones that are build for package manger systems. Like on windows as you can use app that use install shield, Installer VISE, apple update, Windows Installer, etc side by side without hitting major Dependency hell
He mentions something in the article about devs having newer hardware and everyone else having older stuff. That's a point, to be sure, but in my experience, there's enough info on the web to make fixing drivers doable. The bar to adoption is mostly user interface design. Lack of offline help, inconsistent UI guidelines, inconsistent context-menu-access, difficulty in figuring out why you can't enable certain options (because the GUI doesn't tell you that other options are available with a package download), inconsistent hotkeys, etc. On any windows machine, I can blindly press a few keystrokes and both launch apps and navigate their menus. Not to mention that menus and common buttons are almost always in the same place and look pretty much the same.
My family has expressed lots of consternation over some of the changes even within the Windows ecosystem (start menu layout changes, control panel layout/submenu changes, Office quickbar, etc.).
But instead, we're focused on the latest new shiny thing rather than making it all work consistently and intuitively. I mostly use bash consoles so I don't really bother with the GUI but you'll never get anywhere by trying to revert today's typical OOOSHINY wanna-be nerd to use that when they can just pay a few hundred extra bucks and get something that does the same thing the same way every time they click or tap in a specific spot.
While Linux kernel is solid on servers for whatever reason on desktop it always was crashing and/or required occasional reboots. Flashdisks plugging/unplugging creates allocated un-unmountable devices. Desktop machines just randomly reboot. Screen occasionally goes black or garbage forever (it may be X bug). Keyboard becomes unresponsible. OOM problems where the system locks up or some fundamental process gets killed. etc.
Apple's success is an interesting model for what it would take to make Linux mainstream on the desktop. The average non-techie Apple user doesn't know or care that there is BSD running beneath the GUI or that a UNIX command line even exists on their Mac. Granted there is a legacy there where people are already comfortable with the idea of a Mac being a legitimate alternative to the Windows PC, but it is the seamless user friendly GUI and fully developed application ecosystem that make it desirable. The argument can be made that Ubuntu and maybe others are pretty usable and are getting close to mainstream useability, but we aren't quite there yet. Until there is a GUI that is so fully featured and bulletproof that the user never needs to do anything at the command line to achieve reasonable efficiency at all common tasks and the application ecosystem is developed to have decent parity with current mainstream OS in use, Linux doesn't stand a chance in the desktop. I'm not sure that the financial payoff is there for any business to undertake the investment needed, but I certainly hope we get there someday.
Yes. In other news it isn't your Doctors fault you're stupid.
Guns don't kill people; Physics kills people! - John Lithgow as Dick Solomon on Third Rock From The Sun
You have absolutely no fucking idea what you are talking about.
And user lethargy as well as outside org dependencies sometimes.
We move users from Windows to Linux all the time now. The only internal problems are around users giving it a good college try to learn and integrate. A good number of our clients have external LARGE entities they are beholden to that force them into the hands of Office.
Outside of that, it's a done deal already.
A user space device development kit.
Dividing out compute functions into mobile devices might have been the key to this happening. Tablets/smartphones do a lot of the leisure activities associated with computers, and can do some of the minor business features too.
This means that, increasingly so, the market for desktop computers will be for heavy business uses, and for heavy gaming. The marginal stuff will move to mobile as it's able to.
The day the Linux desktop comes is when it becomes easier for the majority of people to use Linux in the office than the alternative.
So, how is that going? What are some of the heavy-use applications that will likely never move away from a desktop computer?
Office applications - Openoffice and libreoffice are considered a viable alternatives to Microsoft Office. The fact that you can write macros in python gives the FOSS stuff a bit of a longevity advantage as new office workers come into the labor force and don't feel like learning Visual Basic. Where it lacks is the Exchange server market, where there's no viable FOSS software to handle email, organize meetings, allocate resources, and have it all work natively with single-sign-on credentials.
Gaming - OpenGL has seen huge improvement over the years, and it gets easier to work with every release. If it isn't already equivalent to DirectX, then it's well on its way. I see OpenGL as having more potential as well, since there are more interested and intelligent parties involved with its development than DirectX. The rendering library is just one component though. You also need top-notch hardware and drivers to match. The NVidia drivers are equivalent from Linux to Windows and are pretty good, if a little unstable. The FOSS drivers for NVidia have a long way to go still, as do the ATI drivers. NVidia is on-board with maintaining Linux as diligently as Windows, but ATI tends to lag behind in that area. Most major gaming engine components already work for Linux, like Havok, or the Source engine. With Steam picking up the banner of Linux gaming, it will certainly grow more viable over time too.
Interface - This is a big one. No matter how proficient you are, this one has to be learned. Linux has hundreds of different interfaces, and all of them require some amount of training to use and customize. Windows has this one because it has been essentially the same since Windows 95, and the paradigm and prior knowledge from all previous Windows OSes tend to transfer over from release to release. The only solution to this one is making streamlined workflows a priority inside of the interface, and then training people on it. As odd as this might sound, I think the best candidate for Linux gaining more ground on the PC interface is a window manager that focuses on ease of user customization, rather than ease of use. For me, that's fluxbox or openbox, with xfce making strong ground. Teaching people how to edit a text file and customize their menus and hotkeys takes me about 10-20 minutes, and the person learning it usually can get far enough with it to make it their own after an hour or two of use. Add in a program that turns your interface into a drag-and-drop to customize mode that's easy to use and it might start making some serious ground. I mean, Linux's real interface is the command line, and bash largely put to rest our ancient shell holy war. Once we can intelligently combine the advantages of gnome, kde, and xfce (which are the three biggest contenders for user space) and make all these paradigms work together, then we'll be on track for taking the desktop.
Anyways, just my two cents.
Now be honest: when was the last time you "configured Linux"?
Microsoft probably has somewhere between 6 and 20 thousand engineers working on device drivers for various windows versions out there making about 80k a pop. Sorry but Linux simply does not have these kinds of resources. It would be nice but I don't see it happening.
It's not his fault.
Linux is a kernel, an a great one at that.
GNU is a desktop, and isn't dominant right now, but it's very popular among large groups of users, some corporate included.
Windows 8 basically handed Linux an opportunity on a silver platter. Now they just need to make the desktop significantly better than what Microsoft is currently offering.
perhaps you can enlighten us as to why he's wrong
I never said he was wrong... Only that he's true to form..
So he's right, but for the wrong reasons? How do you know when he's right for the right reasons?
Boy, you just insist on reading into my post what's not there, twice now. I think you are doing it on purpose....
Torvalds is right... I just find it amusing that he is quoted to be blaming somebody else and being "True to form." Clear yet?
"File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
hate Windows. We do what we do because we love unix. --Theo de Raadt.
When it's as easy to develop and deploy an application on Linux as it is on Windows then it can be the year of the Linux desktop and not before. The notion that I should spend my time creating rpm and deb packages instead of working on my application is absurd.
For most applications the Windows the installer is a simple thing that can be created in a couple of hours. In Linux land creating installers is a difficult, painful, and time consuming job. A waste of scarce resources that has little or no value. I won't waste my time on it.
I've been using Linux part-time for twenty years. I build my own desktops so its been easy to build systems that are compatible between windows and linux.
However laptops have always been very troublesome. I have figured out a solution. Buy a chromebook and install Linux on it.
What corporate?
by Mike Buddha -- Someday the mountain might get him, but the law never will.
WoW used to have a linux client. Now you have to use wine, which is still a pain to use.
Flash used to be released for linux. No longer, and half the web uses newer flash formats that won't run.
Linux itself is working pretty well on the desktop. Perhaps better than ever. I remember the days of spending a week compiling kernels to get S3Virge or Soundblaster AWE64 to work. Not to work as well as in windows, but just to work. Those days are mercifully behind us. ACPI support is maturing nicely. systemd (for all its flaws and pains) is unlocking the power of on-the-fly reconfigurations that were really very painful in the old days.
I'd like to see Gnome and Xorg work better together. I should be able to configure multiscreen setups without Gnome crashing. But again, that isn't a Linus problem.
The future is application support.
The desktop is being replaced by the mobile devices. Linux isn't successful on the desktop because there are so many options and so little standardization. Think about it, if you developed a desktop app would you want to develop and test to every possible configuration available? Yeah me either. A beautiful example of this is World Of Warcraft. They have a linux client they use for internal testing (or they did anyway) but won't make it public because they know support would be an absolute nightmare.
Either a vendor thinks the market is large enough to bother with or not. The "level of bother" factor is largely irrelevant.
ROI.
What is the Return on the Investment? The "level of bother" is the "I". The smaller the "level of bother", the better the ROI, the more likely the vendor will do something.
That said, I DO agree that if the R in ROI is sufficiently small, then even if the I goes to zero it still won't be worth the vendors while. Lots of large companies require both a high absolute R, plus a reasonable ROI, which is why you get companies shutting down small but otherwise perfectly profitable business units. (which is VERY frustrating...)
The device market does nearly nothing for Linux as a consumer brand, nearly nothing for the promotion of FOSS. People don't see the Linux embedded in their router, they don't see and can't even get to the Linux that hosts Android on their phones. Most Android developers don't even touch or see Linux during development.
Microsoft probably has somewhere between 6 and 20 thousand engineers working on device drivers for various windows versions out there making about 80k a pop. Sorry but Linux simply does not have these kinds of resources. It would be nice but I don't see it happening.
Try 500-600. Most of those are "project managers" too who farm the work out to Indian contractors. Microsoft doesn't have the development force you think they do.
Feeding the shill/troll here...
Linux is was not, and is not meant to be anything but a hobby OS for someones spare time, or a companies spare time that they can develop a UI for and deploy their own flavors (android, Red Hat, Ubuntu, etc.) Linux is far too complicated for the everyday user to understand. Even something as simple as entering a static IP address sometimes requires going back to the terminal windows (command prompt) and setting it the hard way. And THAT's the problem with Linux! It was never meant to be a GUI OS just like it's parent, UNIX.
That's why desktop users use Ubuntu.
1 - Open network meny by clicking network indicator at the top bar of the desktop
2 - Choose "edit connections"
3 - Choose the connection you want to edit - click "edit"
4 - Click "IPv4 settings"
5 - Change IP
Please, remind me how that's done in windows 8.1. Feel free to explain differences with windows 8, 7 , XP.
The drivers for Linux SUCK and that's because it's an open source OS and there's no one "single" distro.
Just like any other OS. Supported hardware works, and in this case, backwards compatibility is maintained. Unsupported hardware, shockingly, doesn't work.
I use Mint 17 Linux daily, but what I miss, what is really lacking are Adobe apps. Someone should start a kickstarter for Linux ports. Adobe is already familiar with Qt ( I think I read Lightroom is Qt) so they have the experience.
Let's put our money where our mouth is and get adobe to Kickstart the ports.
Slashdot's rate-of-post filter: Preventing you from posting too many great ideas at once.
I have a ton of hardware that doesn't work in windows 7 and above but works fine in Linux.
I never found hardware support lacking.
Lol, GP is backpedaling.
Microsoft probably has somewhere between 6 and 20 thousand engineers working on device drivers
Are you sure? I know they make their own mice and keyboards. But for most devices, I get the impression they rely on the hardware manufacturers to write the drivers for their own devices.
Unless you literally meant "6<engineers<20000".
Has absolutely nothing to do with the kernel. It would seem people just want to harp on Desktop Linux for the sake of harping on it. Windows fanboys, likely.
Funny with all of those developers I still have to buy new hardware every time a new version of Windows comes out.
Android, still based on Linux ...
No. Android is hosted on Linux, not based on Linux, very big difference there.
... Notice how my binary applications run on *very* many Android devices without recompilation, even when I write in C using the NDK
Most Android apps are pure Java and do not use the NDK. Those that do are often using legacy C/C++ code that is not Linux specific, many OS calls are actually POSIX calls so they are not really Linux specific either. Again, linux is just the current host environment. Most apps don't care, even many apps using native code via the NDK.
Just because it's not widely used doesn't mean it isn't used. Everyone that I know that uses Linux uses it as a desktop OS. Maybe not their only one or primary one, but they still use it as a desktop OS.
So Linux is entirely viable as a desktop OS... the fact that it isn't widely used is just one ilustration of the old saying about leading a horse to water.
File under 'M' for 'Manic ranting'
Posting from a corporate desktop, using Ubuntu, there are 6 Ubuntus out of 10 desktops around me.
This is not a Fortune 500 company, but it's a public company.
I also read Slashdot, I have seen news of corporations using Ubuntu
A desktop customized version of Android (complete with a real desktop) is still based on Linux (at least Google's fork of it) ...
Android is not based on Linux. Android is **hosted** on Linux, it is really its own operating system. Most Android apps are Java and have zero interaction with Linux, they only use Android. As for apps that have some native code (c/c++ via NDK) they are usually using legacy c/c++ code that is not Linux based and/or they are using operating system calls that are POSIX based not Linux based.
Linux is just a host for Android. It could be replaced with some other POSIX compliant OS and the vast majority of Android apps would not know or care.
Where I work I have a Windows desktop, but web browsing is done on Linux. You wouldn't know it though if you didn't look close.
There are places that use Linux desktops.
"File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
Let me give you a perspective. I love Linux for the CLI environment. I would hate to live without those tools, and in fact, when I use Windows, I am using a lot of the same software, including the same shell. I also like a nice desktop environment, though. The one I like happens to be Windows. I'm used to the way it works, it enables my productivity while at the computer. I've yet to find a desktop environment for Linux that does the same. I know that's going to be a wildly unpopular view, but there it is.
Is that really a kernel problem? I find it hard to say that it is.
Total non-issue; the majority of computing consumers have moved on; they're using phones and tablets. Corporate users will use what they're given.
The days of paying hundreds of dollars for an operating system and compiler are (thankfully) gone. The OS is irrelevant anyways; you go to where the applications are; anything else is just silly...
I'm not sure why people are looking for a cause. It's obvious why Linux isn't number one: momentum. Windows managed to grab a dominating share of the market first, and it's going to take a lot to change that. This is true in any industry. Linux will only be #1 on desktop if the others really screw things up. Most people don't choose an OS based on their merits, they just buy a computer and use the OS it came with cause it works. There won't be a "year of the linux desktop". There might be, however, a year of the "OK this time Microsoft has really dropped the ball".
I hate FPS games. Desperately. I love small and sometimes clever games, though. Flash games - and recently sometimes HTML5 games - are perfectly suited to my needs. Play a game for 15 minutes, sometimes 5 hours over several days - then you're done with it. You can play something completely different next time. I noticed that very often games that take only 5 minutes (like the story is over and there is no point in replaying it) get very high ratings - also from me. I love playing games with fresh ideas. And almost never a game requires to be 3D. I hate 3D games. Desperately. One series of desktop games I played was Heroes of Might and Magic. A great game. But then they decided to make the new one in 3D. For no real purpose, as the gameplay requires only 2D. Result was that you had to pan and rotate, just to get a view on your hero that was not obstructed by trees. Why in hell did they do this? 3D simply sucks.
As the GP said though, what does that have to do with Linus and his kernel?
Nothing apparently.
Linus is blaming somebody else for Linux not taking the desktop, this is not to say he's not right, it's just amusing that he would blame others and tells me he's not interested in helping with the solution.
"File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
I thi nk that what you are seeing is the difference between fat packages and slim packages. What I mean is that in your typical gnu/linux distribution, libs are installed on the system and applications depends on teh libs. That makes dependency issues a real nightmare.
But all other succesfull operating system take a different approaches. on windows, application typically deploy their own libraries. On macosX everything is typically in a fat binary. On android, all libraries are shipped in the APK. I assume IOS works the same. They all depend on a slim "operating system" and on shipping "complex" libraries when needed.
I wonder if that is the main problem with application deployement?
Ok, so than about 20 thousand seems about right then... As every project manager would manage 10+ employees.
Troll is not a replacement for I disagree.
Ew... you NEED the .deb repositories, because we don't HAVE a unified ecosystem. I need the debian apt system to grab all my dependencies for me! a more git like approach? What? How? You want a more android app like system?
So... an app system full of useless trivial apps that have really stupid permissions to my box.. And on top of that nightmare, you want to drop .deb dependency checking... on a linux desktop?
No Sir. Just.... No. Ew.
If you're concerned that it's hard to get code into the official debian repositories, that's a conversation we can have. But no, we can't go away from dependency checking toward an app store model. I don't WANT any old stupid app to be in the offical repositories.
Nope, Microsoft don't provide many drivers for most hardware. Linux is supporting far more platforms/hardware drivers directly in kernel compared to Microsoft. All the Xerox/HP/Canon/NVIDIA/ATI/Intel/Philips/NEC/Toshiba/TV Tuner/Sound drivers for windows are third party drivers. It's no longer a Microsoft vs Linux issue, it hasn't been for a long time. It's all about the apps. Windows itself is incompatible with many pieces of hardware (forced obsolescence) Mainly because third parties decided to drop support around 64-bit or a newer Windows version.(I'm looking at you HP, dropping network scanning from windows 7 64-bit on certain printers) The same manufacturers who refuse to support Linux funnily enough don't support their Windows products well either. The hardware support claim is a furphy. All the large scale corporate printers/photocopiers/POS systems I've encountered have full Linux support. The only thing holding Linux back is stuff like frontline service applications written in .NET/Win32 that aren't getting ported over. More OSS software that can drop in and take over corporate software deployments is what's needed. Samba 4 and OpenExchange server are BIG steps in the right direction and we are seriously evaluating these for corporate widespread deployments.
Nope, Microsoft don't provide many drivers for most hardware. Linux is supporting far more platforms/hardware drivers directly in kernel compared to Microsoft. All the Xerox/HP/Canon/NVIDIA/ATI/Intel/Philips/NEC/Toshiba/TV Tuner/Sound drivers for windows are third party drivers. It's no longer a Microsoft vs Linux issue, it hasn't been for a long time. It's all about the apps. Windows itself is incompatible with many pieces of hardware (forced obsolescence) Mainly because third parties decided to drop support around 64-bit or a newer Windows version.(I'm looking at you HP, dropping network scanning from windows 7 64-bit on certain printers) The same manufacturers who refuse to support Linux funnily enough don't support their Windows products well either. The hardware support claim is a furphy. All the large scale corporate printers/photocopiers/POS systems I've encountered have full Linux support. The only thing holding Linux back is stuff like frontline service applications written in .NET/Win32 that aren't getting ported over. More OSS software that can drop in and take over corporate software deployments is what's needed. Samba 4 and OpenExchange server are BIG steps in the right direction and we are seriously evaluating these for corporate widespread deployments.
Every time this topic comes up, we see the same old thing: Hard core Linux users who can't figure out that it's not about the kernel, it's not even about the UX, it's about the apps and device support. Tell a mainstream computer user who's never touched anything but a Windows system or one of the tarted-up versions of Linux (e.g. Android) to switch to Linux, and one of the first things he or she will ask is, "Can I run Office/PhotoShop/whatever? Will it work with my printer/scanner/whatever?"
This should be painfully obvious, but let me explain it yet again: For real world computer users the OS is a necessary evil that they put up with as a way to do the things they want with a computer, like Web access, e-mail, document creation, image and video editing, etc. They could not care less about the Linux vs. Windows battles so many people here find consuming.
Linux is easily a modern enough and feature-rich enough and slick enough OS to keep mainstreamers happy, except for the lack of support for what they really want. Until that changes, Linux has virtually zero chance of taking a sizable portion of the desktop, even when MS releases a complete train wreck like Win8. Even as dreadful as Win8 is, people STILL chose to limp along with it and hope for improvements instead of switching to a free and rock-solid alternative. That should tell you everything you need to know about Linux on the desktop.
And yet, no one's really complaining about device drivers. Linux device drivers are fine. You don't see it happening? Uh. It already happened. It is happening. It will continue to happen. I plug stuff in, it works. Except for some printers, because those companies are just dead nasty evil.
We're talking about the desktop here....
The only people who care about stable ABI are proprietary driver vendors. Put your damn drivers in the tree where they're supposed to be and whoever breaks ABI has to fix them for you.
As the GP said though, what does that have to do with Linus and his kernel?
It has to do wth Linus' alleged desire to capture the "Desktop" market, which was the topic of the Article.
Do try to keep up, please!
You have absolutely no fucking idea what you are talking about.
Sez the pusillanimous AC.
.
At least I placed my Karma on the line to state my opinion . (and I notice, in typical immature Slashdot fashion, I'm being downmodded for expressing my opinion
What did you do to advance the discussion?
And what did you do to explain why I have "absolutely no fucking idea what [I'm] talking about", eh?
What are you, like 12? I am an embedded developer with over 3 decades (almost 4) of PAID experience.
What's your Computer Geek "Street Cred"?
Just like any other OS. Supported hardware works, and in this case, backwards compatibility is maintained. Unsupported hardware, shockingly, doesn't work.
I don't know, some of the "unsupported" hardware does happen to work just fine on Linux. Not often, but sometimes.
"File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
My desktop computer at home is running Linux for more than a decade now.
"Winning the desktop" has never been about winning over the geek.
It's always about winning over the full time office worker, the temp and the senior volunteer. The billion or so clerical workers in this world who keep things running behind the scenes.
I'm a fairly savvy guy, but my background is in RF&Microwave engineering, not computer science or IT. I should be low hanging fruit for Linux. I don't do a lot of programming, but have done some assembly, C, Matlab, Vee (shudder), etc.
So why don't I run Linux?
Every couple years I try. Most recently I put a Linux Mint installation on a virtual machine and gave it another pass. No go. Getting the monitor resolution right took a bunch of googling, tracking down some arcane text file to edit, restart, finally get to select the right resolution. Halfway through I remembered I had to do the same BS when I put Ubuntu onto a partition a couple years before (still there, couldn't easily figure out how to get the boot-loader going on the new machine, so it just sits there on my second drive). It's the fricking 21st century, it should just work!
The more you dig in, the more you are confronted with vast wasteland of fragmented BS.
I also tried a pre-canned distro called CAE Linux a little while back. I hit a road block trying to run some of the tools when I found that only about 3/4 of the needed pieces have english localizations, and I don't speak French. The other problem was that a lot of the naming within the main toolset was cutsey crap that was not intuitive, so it made a hard learning curve worse. Linux is rife with such dumbass naming conventions (WTF does "Grep" have to do with searching?!). I was hoping to keep the company from pouring money into a grossly overpriced thermal simulator. Sadly for an engineering group it was clear that Linux and those tools were just not adequately usable.
So unless you are installing Linux for a home user to surf the web and little else, I just don't see the current philosophy of Linux ever getting broad penetration on the desktop. There is a thin veneer of polish for Office and web, but anywhere off the beaten path, even a little, requires a deep dive into jargon hell.
I've been using Linux on my desktop since running AfterStep WM on RedHat in 1996. My current machine is running Mint Cinnamon 17...
On macosX everything is typically in a fat binary.
I don't think that means what you think it means.
"Fat Binaries" are a term that originated back in the days of Apple's first CPU-platform migration (back in the MacOS days). "Fat Binaries" are Applications that have been targeted to install and run on Macs with different CPU Architecture (in the original case, either 68k or PPC, then later in OS X, either PPC or Intel).
They are "Fat" because they actually have the necessary "CPU-specific" code duplicated for the different hardware platforms.
Apple just had a "Launcher" (can't remember the actual name) in the OS that would silently and automagically pick the proper version of the App to load and launch.
However, talking about "Dependencies", OS X (and even the original MacOS) was (and is) VERY nice about not Spraying stuff all over your Drive (like Windows STILL does!!!). That's why, by and large, you can move an Application (which looks like a single Icon, but is in reality just a Folder full of stuff) simply by Dragging it to virtually anywhere on any accessible volume, and Uninstall it by Dragging it to the Trash and clicking "Empty Trash". It will probably leave behind a Preference file of a few Kilobytes; but, compared to excising a Windows App (or apparently a Linux one), it is a beauty thing...
Dear Mr. Linus Torvalds: I may not like your personality profanity, attitude, etc, but, completely agree with your stance on the "desktop". I am also a fan of your baseline achievement: creation of "LINUX". The "distributions", based on your work, continue to be a valuable component of my overall approach to PC survivability. I feel that I OWE you, and it does not matter whether or not I "LIKE" you, based on some comments posted somewhere on the web. You have a permanent "ally" in me, if you should ever need one at my level. Godspeed, and good health. Sincerely, Robert I. Baker.
But we'll never see it.
If you want a fixed (relatively) stable system with extremely limited configurations that is a unix, get a mac.
If you want flexibility and up-to-date drivers that work, get windows (in particular, video drivers).
I'd love to use linux, but I've found I end up fighting the OS too much when it comes to updates, new drivers, supporting new hardware, etc. If only those parts could be reeled in, it would be my OS of choice, as long as it is also a first-class gaming platform.
Next time you have a Windows box check the device manager. You'll see almost all the drivers are signed by Microsoft. I should have clarified that these numbers include devs doing device drivers and working on general compatibility issues with various softwares.
https://www.microsoft.com/en-u...
If Linux wants to own the desktop then he needs to build an operating system worth using. So far he's build the kernel and uses everything else from other people. GNU built the everything else and then started building the kernel.
BSD has it all.
There's a reason so many other OSes are built on the shoulders of BSD, but Linux thinks that constantly revising his kernel is somehow going to allow him to get to regular users.
But nvidia and amd do provide closed sourced binary drivers for linux already and they are very damn stable. Plus my primary OS is windows 7 and never had a driver issue. This isn't 1998.
I use Linux, almost exclusively, but I can see one of the major problems preventing migration that many linux developers cannot. It's confusing and difficult for the average user to learn where all the configuration files are and what they do. The moment you expect a new user to open a terminal you've already placed a giant barrier to adoption in the way. Certain distros have made giant leaps of progress in this matter but it's still a problem for all.
Want to make a minor adjustment to how your sound card works? Command line. Want to tell your laptop to ignore the touchpad? Command line. Want to use Tor? Command line. Want to install software that's no on the Ubuntu Software center? Command line. I understand that GUI is a dirty word to some developers. I understand the focus on making things work before worrying about making them easy. But the path to the year of Linux on the desktop is paved with intuitive, simple, GUI driven configuration and computer usage.
If video games influenced behavior the Pac Man generation would be eating pills and running away from their problems.
Make one that doesn't have to be recompiled umpteen thousands of times to get a single program to work with it and you'll win. :D
How Linux wins the Desktop
1. We need a "Default". Not necessarily a default Distro, but a set of standards that all distros can follow. Of course, other options will be allowed, even encouraged. Rationale: We need the "fragmentation" problem to be addressed, and I would suggest that a good start would to have a standard interface that is common across all of "Linux".
2. We need an easy way to manage a large group of computers. Large or small, businesses and schools want to make the configuration of their computers easy. Examples: Mass deploy Chrome. Setup a lab of computers to use a single printer. Setup logins with permissions and shared home folders. Rationale: These features are easy to configure on Windows and Mac OS X, but not so easy on Linux.
3. Easy Deployment. There needs to be a scriptable deployment that can mass install Linux onto multiple computers easily, including initial setup and joining of whatever management system is being used. While "image based" deployment can work, native installation deployment with configuration would be better. Rationale: If it is going to compete against Windows and Mac OS X, it has to be as easy to deploy.
I'm sure there are some projects that already fill some of these needs... but it's not there yet.
The problem in my point of view is that the Linux desktop infrastructure is way outdated in architectural terms.
Seriously: MS doesn't *have* to do that, far from it since the primarily MOST used computing platform for Desktops + Servers combined IS Windows, worldwide (fact) - so, if you're a development shop, you *HAVE* to target MS & Windows *IF* you want to "make it", since it's the largest market segment...
* This is business 101 level stuff...
APK
P.S.=> Your propoganda however/by way of comparison? Come on, lol... give us a break!
Now, on Mr. "T" & his wanting that? I could be a jerk & say "How's it feel to want?" (Gord haha if you see this laugh) but I wish him luck in his pursuits... Linux IS NOT "ALL THAT BAD" & in fact? It's pretty good... but device support *does* harm it (however, I for one, never had that problem with it here, not once, & I've used it in 1994 (well, then I did with slackware 1.02 iirc, & a Diamond "SpeedStar" (?) or "Stealth" 24 vidcard NOT getting "X" support, which imo, sucked... but never once since, & I did a year in 1999 with RedHat 6.x iirc, & an entire summer in Europe with a laptop using 10.10x series Kernel & KUbuntu... never a problem - his goal IS 'doable' possibly, you never know, & I wish him the best of luck (he has a lot of heart, I admire his directness/lack of "political correct" bullshit too, & a project with MASSIVE potential))... apk
Does fiddling with the KickStart and PXE rebooting (a lot) to get the rocket sled that is linux lined up on that badger just right count? Blap, wall of text, hope you know some good Tasseography. Oh, and then when some undebuggable problem from some unholy combination of systemd|dbus|polkit|selinux|pam|sssd|ohwhatnow and then futile forum and bugtracker blues morphs into "eff it, I'm reinstalling"...yeah. Good fun.
echo `date +%Y` is the year of the `uname` desktop\!
There we go, problem solved.
Source please. Seriously!
If they even have close to the utter threshold you've put forth, their programmers are ABSOLUTE SHIT! They still can't get even get Windows updates RIGHT! See last weeks 'BSOD update debacle'!
The only people who care about stable ABI are proprietary driver vendors.
Which make up a lot of the folks that sell hardware these days. "I want to include Linux support, but to do so I have to give my source away? No thanks." That's assuming the vendor thinks there's enough Linux market share for a decent ROI, which usually isn't the case unless you sell RAID controllers or other server-oriented hardware.
OS X sits on top of BSD, how hard could it be to write a similarly good window manager for Linux? Why has it not happened?
PC: Gigabyte GA-Z97X-UD5H, i7-4770K, 16GB RAM, GTX 780 Ti.
Tried Mint Linux 18 Cinnimon & KDE, Ubuntu 14.04, Debian 7.6
None of them would even boot on my computer, even after dicking around with bios settings for an hour.
Laptop: Asus Zenbook UX41LA
After pissing around for half an hour with bios settings, I finally managed to get it to get farther than the grub bootloader. Things seem to work, but KDE crashes randomly, Cinnamon and Unity don't remember window locations and have their own oddities. None of them let me control screen backlighting, even after trying all the 'hacks' posted around the net.
Then I find out that video performs even worse with the open source Intel video drivers.
I'm all for Linux on servers, I own and run several, but damn, it shouldn't take a bloody engineering degree to get a GUI OS to install and work properly.
If he wants the desktop he'll need to stop wanting the desktop. It's not a kernel problem, it's an office problem. Communications. Seamless integration has been a pipe dream for far too long from my perch. Without a full package you're tossing rocks at the river and hoping for a damn. KDE? This that or the other wrap that's functional and cool? Nope. Private and secure operations with links to people that will encompass the full work flow, that's the problem. Linux will work all day long on any desktop. It's all of the desktops that are the problem. Isn't there an early adopter that wants to roll their entire country back onto MS? That's not a kernel problem, that's a comms problem, communications with every other mid to large size office on the planet. Linux - not the problem.
\r
Neither do some geeks. I prefer my OS working reasonable well out of the box without the need (!) to have to reconfigure things. I don't want a Lego set for each and everything in my life; thank you!
Perl Programmer for hire
I think if you want the desktop it's going to take another linux-kernel-level effort around the GUI. The question is do we keep trying to put more band-aids on X11 or do we design something from the ground up that everyone can agree on?
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
"Linux on the Desktop" is called Chrome or Android and the "desktop" is wherever we are instead of a jumble of wires connected to a monitor.
Perhaps "Linux on the multi-window desktop" or "Linux on the desktop in a focused activity setting" is a more precise of what some people mean. The Android ecosystem, from the CDD on up, is staunchly opposed to rich window management, instead preferring a paradigm of all maximized all the time that makes it hard to work on one document while referring to another document.
A desktop customized version of Android
Won't have the Google Play Store. The Android CDD requires that apps run in the full screen (or at least think that they're running in the full screen). This means the calculator app fills your entire 24" monitor, obscuring whatever else you're working on.
I still think KDE and Gnome still look a bit "Fisher Price" compared to a commercial GUI like Windows or Mac
If you'll remember, Windows XP's default theme was considered "Fisher-Price" until people saw Windows 8's Start Screen.
Lack of offline help
How much space would it take on the install disc to have full manuals for all included packages translated into all supported languages?
So I recommend him to start his own Desktop project. :-)
Seriously, I don't know of, now, any other Open Source leader capable of doing a decent Desktop. Torvalds finishes what he starts, and he finishes it vrey well (see git).
We had very good Desktops in the past, but nowadays things are just too shiny and too new and... too dumbed down to be useful to me: who knows me from other /. posts about this matter knows why I migrated to MacOS two years ago, and don't plan to migrate back in the short run.
I still love Linux - all my non desktop machines are Linux, no questions asked. But I just can't handle any of the present mainstream Desktops to use Linux again on my working box.
Lisias@Earth.SolarSystem.OrionArm.MilkyWay.Local.Virgo.Universe.org
If you mostly play Flash and HTML5 games, your not really a gamer either. So no, you dont get to complain. Desperately.
People prefer Windows or IOS because they want to run Photoshop or Call of Duty or any of the many applications that won't run properly on Linux.
There are two problems with The Year of the Desktop for Linux:
1. Linux - it is a kernel and can't fix what goes on in the desktop. I never understand why people expect those to correlate.
2. Linus - there is no Linus in the GUI world who can put the foot down and stamp out a path.
And yes, X and its ecosystem _is_ a problem. When parts of the GUI are separated and with individually selectable APIs you get a wonderful sense of freedom, but the days of yore with the Amiga, the Atari ST, the Archimedes, the Be, and Apple they did unwittingly reap rewards from having a nazi stranglehold on the _whole_ of the GUI.
How do you go about adding 'Datatypes' (Amiga feature - I believe Be had something more advanced?) to every GUI running on top of Linux? Who can plant the flag on top of that hill and declare "so be it"? It is a framework for GUIs that really made everything easier for everyone.
Failing that, I'd like to point the the XPK compression framework for the Amiga that in many ways was similar to Datatypes - for the end user it also helps the desktop experience by being able to handle more types of data in applications that didn't have an explicitly coded way to do so.
Linux has solved the kernel part, but you shouldn't expect that to automatically fix things several layers higher in the stack.
Often the device makers create the drivers and Microsoft just validates and signs them.
The check box is gone, and the ads are back, and dumber than ever! I
It's still there for me, maybe your karma dropped a bit.
I miss the days of Barracuda ads that made sense on slashdot. The new ones aren't targeted at geeks at all.
I agree.
"First they came for the slanderers and i said nothing."
You seem just like me. I've mostly been playing browser based games for the novelty. And I absolutely loathe the morons who make a good 2D game into a terrible 3D game. I wouldn't complain if they didn't sacrifice functionality... but they usually do.
Don't waste your vote! Vote for whoever you want, unless you live in a swing state it won't matter anyways
You guys no nothing about Linus or how he likes to use his desktop. It's funny how people export their dreams and ideas to one man like this. The guy is not remotely qualified to write a desktop. Have you seen all the commands in git? The first round was a usability nightmare. Hell he himself would admit that.
Linus can talk about "wanting the desktop" all he wants, but the sad fact is Linux will NEVER own the desktop.
As big of a piece of shit as Windows 8 is, [insert Linux distro here] is worse in many ways. I've been trying to give this OS the benefit of the doubt for decades and these people still find ways to fuck everything up. Don't even get me started on Linux Mint. I had version 15 installed on my netbook, and went to install some programs recently through the package manager. Kept getting a bunch of "file not found" errors. Turns out that Mint, in their INFINITE GENIUS, decided to base their distro on Ubuntu's package repository. And guess what Ubuntu did, when the idiotically short "support period" for that version ("Faggotry Falcon" or "Idiotic Ibis" or some other cutesy name, can't remember which) expired? They just ALL of those fucking package files off their server! So guess what my only solution is if I ever want to update this OS or even install a fucking application from the repository? I get to wipe the hard drive and reinstall something else.
Are you fucking kidding me? This is what I should expect and be happy with in glorious and amazing 2014? Wait, don't answer that--of course you're serious. This is just ONE example among countless others of how the Linux community's head is collectively shoved up its own asshole. ONE example or thousands. But try going to some IRC channel and being upset or enraged about the status quo, if you dare; they will give you excuse after excuse why it has to be this way, and here's some workarounds to try, just type [insert bullshit here] into the command line and edit [this fucking file] and wave a magic wand and tap your heels together and that's all you have to do. Still upset at the idiocy of it all? You'll be mocked and flamed and ridiculed and kicked off, told you are unworthy of Linux and should just go back to Windows.
So go ahead assholes, just keep on dreaming about being the choice for desktop computing. It will never happen, as long as you continue releasing such PIECES OF SHIT, deluding yourselves into believing it's gold and being oh so confused as to why nobody is interested in installing your PILE OF SHIT.
This is a glaring blind-spot in the GNU/Linux ecosystem.
I suspect windows will be come illrevent, free, or cloud based very soon. Android has kicked it in the teeth tablet wise, and apple hit them when it was down in the mobile phone market. the chinese makers are coming with linux variants and samsung is biting at their tail on mobile too. this market is going to take care of 99% of the leisure read/game/talk time of people. They no longer need a desktop for that. The only reason to have a desktop is gaming is a big keyboard and mouse. windows can compete with gaming on the xbox, but with stuff like linux steam and support for native gaming in linux their days are numbered. They should go like IBM and become support for legacy windows and then sell it off to some other company to die a quiet death. Then they could start to innovate and be cutting edge doing stuff like Watson or Siri stuff they already have most of the know how.
NEVER NEVER NEVER NEVER NEVER NEVER NEVER NEVER GIVE UP! "No limitations, no boundaries, there is no reason for them."
The kernel is fine. Driver support is fine. The installer is fine.
The big problem is the applications. I don't want 500 different options which each do some different 75% subset of what I want.
I'd rather have 2-3 options which each do 95% or more. Unfortunately, as soon as any Linux desktop application ends up getting anywhere near feature parity with what I want, the desktop environment it was written for gets updated, breaks the way it works, etc.
The way Linux desktop networking works is also often retarded. Network shares should be mounted just like disk images, and apps need to be able to treat them as such. In reality, you often can't browse to an SMB share, can't drag/drop to an app without the desktop environment wanting to copy the file locally before opening, etc.
Apple can do it. Microsoft can do it. Why is it 2014 and working from a network share with Linux still sucks?
Year of Linus Torvalds on the desktop???
But the task scheduler still sucks big time.
For Linux to make a step to being a desktop contender the task scheduling needs a major revamp. Specifically it needs to stop trying to be everything to everybody. Its a BAD realtime scheduler, its a BAD general purpose scheduler, and its a BAD scheduler. Its not that its BAD over all, its just that it is BAD for all scheduling tasks. That and the code is horrid.
Lack of stable interfaces never harmed the WinDOS market.
First of all, troll, it's called Windows. Changing official names to derogatory slang terms is definitive proof that you, Jedidiah, are a nasty troll. Second, Linux is the pinnacle of instability, thus why every other platform including Windows can be considered to be stable. Now, I state this for the benefit of the uninformed. Lying, vindictive trolls like you refuse to see the truth. You love Linux. It has become an unhealthy part of your existence and you'll defend it with the most wicked lies you can imagine to shield yourself from the pain of confronting its flaws.
Linus is blaming somebody else for Linux not taking the desktop, this is not to say he's not right, it's just amusing that he would blame others and tells me he's not interested in helping with the solution.
Good thing too. Linus is arguable one of the worst user interface designers on the planet.
When all you have is a hammer, every problem starts to look like a thumb.
I have a dell laptop with 8.1 on it I bought some months ago for work. I have all the latest drivers and windows is up to date. However, the WiFi randomly decides to disconnect, and then will reconnect only after clicking on the network icon in the task bar. It's not the access point, its only this laptop in windows. Also, the reported remaining time on battery is always higher by a factor of 8. Not making this up. Nothing I have found has fixed this. In Ubuntu 14.04 though, I have zero issues like this, and it is a stock install. No hackery at all. None. Also, using the exact same applications that I use for work in windows my laptop is more responsive and has 2gb more free ram, and it doesn't hit my HD as much. I'd work in Ubuntu but MS Lync is required for me to work with my team, and doesn't run in linux.
The problem for Linux on the desktop has never been the software, and always the hardware. Or more correctly, getting hardware with a pre-installed distro into the hands of the general public.
And here the issue is not technical, but financial and political. A company would have to go head to head with MS, and their deep pockets.
The perfect example of how hard this is can be found with netbooks. Asus originally launched their EEEPC 701 with a customized distro, on hardware that in no way could fit Vista (the Windows that MS was pushing at the time). So what MS did was offer a deeply discounted XP with some very stringent hardware limitations. Almost over night, netbooks looked like carbon copies when comparing spec sheets.
Similarly they have big wholesale discounts for Dell, HP and the rest, if they go exclusively MS. There have been glimpses of these over the years thanks to legal wrangling and leaks, but for the most part they are under NDA etc.
So in essence Linux on the desktop means that some company like Canonical or Red Hat would have to set up a store chain similar to Apple's to even have a chance.
"The one I like happens to be Windows. I'm used to the way it works, it enables my productivity while at the computer. I've yet to find a desktop environment for Linux that does the same."
what is the difference between a linux and windows and OSX desktop? nothing, point your mouse and click on the app, works on all of them (maybe win 8 is a change). only the app is different (except Libreoffice etc)
"The hands that help are better far than lips that pray." - Robert Ingersoll (1833-1899)
Help who exactly? There are good reasons why non gpl modules get limited api access, and why the kernel has a tainted/not tainted flag displayed in every oops. The devs don't accept bug reports from kernels that have closed/non gpl code running at the time of crash. Why? Because they can't look at the source or build debug versions. The kernel devs don't want to deal with floods of users posting such kernel dumps and demanding answers the devs cannot give. A linux ecosystem like this would be no better than windows' (or any other closed platform, really).
The net result is that the vendors who take linux seriously submit working source for their hardware under gpl2, which is then folded into the main release via git pull requests. If the vendor then abandons the hardware and driver at some point, at least the code is available and can be maintained as kernel internals change and as it is ported to new hardware.
The ones who don't release source, don't get to take advantage of this 'automatic' maintenance and must do it all themselves, nor do their blobs interfere with the kernel devs or users running more recent kernels that won't load them. This effectively keeps these vendors accountable for the entire ecosystem needed to support their blobs, which is good since they're the only ones with complete source. The users stuck on the older kernel releases in order to retain needed functionality are rightly dependent on the vendor for fixe instead of the kernel devs.
Those don't run in the kernel...and the stubs that load into the kernel are non-gpl, so any crashes posted will show as tainted and will not be addressed by the kernel devs.
Note that only 65k are in "engineering". This is across the entire company, working on numerous products (many of which you probably don't even know exist), and also internal infrastructure like build systems, test automation, and internal dev tools.
Quite obviously, one third of that cannot be working on the drivers, and even one tenth is an unreasonably high estimate.
is that PCs come preinstalled with Windows. Whether a window manager is too configurable or if some rare wifi card is compatible is a ridiculous discussion.
Just because they are signed by Microsoft, doesn't mean that Microsoft wrote them. For Vista, and even more so for Win7, large number of drivers was included out of the box to cover a wide range of hardware without needing driver CDs as often as XP did. Most of those are third party drivers, but because they are redisted by MS, they have the MS signature on them.
Basically, if you see MS certificate on some binary, it means that someone at Microsoft has built that binary. It doesn't mean that they wrote the source code for it.
Actually, the notable exceptions are the gpu vendors, and a smattering of printer models... Most devices 'just work' under linux as they're just generic hardware anyway. People shouldn't be hiding their secret bits in binary blobs anyway.. It is NOT secure.
They don't have to give anyone source, but then they don't get to take advantage of having their driver maintained in-tree as kernel internals change over time and as it is ported to new systems. The kernel devs also don't have to deal with users complaining about crashes when those drivers are loaded. If the user's kernel oops message says 'tainted', the devs will tell him "go talk to your vendor for support." Why should the devs have to debug systems with source they can't access? Linux is supposed to be open source, not half-open/closed.
Most raid controllers work under linux, even the shitty software ones do now.
> but Linux simply does not have these kinds of resources
Linux is primarily contributed to by Intel, Microsoft, IBM and others. They are exactly the kind of resources you are talking about. Go look at the checkin list for the last year...
There is no 'theway'. I know apple fanboys like you like that kind of totalitarian thinking, and will accept almost any kind of inconvenience in order to march in lockstep, but part of the reason apple hasn't won out over microsoft is because there are A LOT of people whom the system doesn't work well for. The primary driver of systems is applications and marketing. Linux does have quite a bit of so-called 'professional' applications, but they get zero marketing.
Userland is not the kernel. In fact, linus has made it very clear that it is not allowed for kernel devs to break userspace ABI. Beyond that, it's up to others to decide what to do with userspace. The fact is, you CAN run old linux 2.0/libc6 binaries on modern linux for a given platform. You just need copies of libraries that haven't broken ABI with the ones it was linked against.
If you want to blame someone for their laziness in supporting your desire to keep your code closed, then blame the distros who aren't shipping the ancient libraries you linked against 5 years ago in their current releases and/or blame the authors of the libraries who broke their ABI.
Cool, Neowin users are starting to use Slashdot again I see.
It will take over every single other machine on the planet, from cars to celphones, to pace makers, to microwaves to game consoles to everything.. except the desktop.
Even 100 years from now, when the linux powered skynet coordinate the uclinux 6502 powered terminators to finish off the last of the human resistance, those terminators will still use windows on their desktop computers.
Actually applications break on osx updates and major releases quite a bit. This is usually due to libraries being removed or changed in the background. Apple's 'successful' solution is to tell users 'sorry your shit doesn't work on the new os.' Beign the apple fanboys they usually are, they swallow their pride and rebuy their applications. If they're lucky the application vendor will release a patched version for free.
MS creates a lot of generic drivers (think stuff like USB mass storage, generic monitors, SATA controllers, Media Transfer Protocol devices, anything like that where there's a standard that the hardware implements). You can get a basic (but functional, if you don't mind probably having the wrong video resolution) computer running almost entirely on Microsoft-written drivers.
With that said, the vast majority of Windows drivers (by count, not necessarily by usage) are developed by hardware vendors. Microsoft probably doesn't even have 20k people in the Windows org at all, even if you include test, PM, and management. They certainly don't have that many on the kernel and devices team, never mind the portion of that team which is actually developing (including designing and testing) drivers.
There's no place I could be, since I've found Serenity...
As a monopoly, Microsoft gets to hold the proverbial "gun" to device vendors heads and say, "support our OS [or] we'll fucking destroy your market ...".
No, MS do not need to hold a gun or say anything.
Any hardware maker, unless they are very very specialised, will first write drivers for Windows simply because ~95% of their market is going to be Windows users. After that they might write drivers for Linux perhaps to pick up a few more sales (like that's why I buy HP printers) and/or just in case next year really does become the year of the Linux desktop.
Then why is it that I can still use my GF's scanner in my modern Linux distribution and she's not able to do it even in her comparatively ancient Windows 7 box?
We were able to at some point get it working on XP (after a lot of trial and error with random internet drivers), but that was it.
With Linux? Plug it in and launch the scanner app.
I haven't had driver problems under Linux for years now.
1 - Right-click the network icon in the system tray (it's in the same place on all versions of Windows from the last decade, and XP too for that matter).
2 - Select "Open Network and Sharing Center" (if on XP, just go to Properties, but make sure you got the right network interface if you have more than one).
3 - Click on the network interface name (something like "Local Area Connection" or "Ethernet"; XP users skip this step because you already chose the interface) to open the interface status.
4 - Click on Properties and, if not already running elevated, go through UAC. This gets you where the XP users were waiting (for the 13 years since their OS came out...).
5 - Double-click on "Internet Protocol Version 4".
6 - Change IP.
There's a number of alternate ways though some of those steps. You can also short-circuit the whole thing using netsh, but it was implied that you wanted the GUI technique. Oh, and these steps work for the last four (arguably five) OS releases, on everything from the extremely basic Starter SKU to the highest-end Windows Server Datacenter Edition to even the RT versions. Care to give the steps for Ubuntu 9.04 (a mere five years ago), or for Kubuntu/Xubuntu/etc.?
There's no place I could be, since I've found Serenity...
1. More universatility for linux, one install cd/dvd for all systems: 32 bit/64 bit etc.
2. More drivers for linux, one install cd/dvd for all systems: laptop chips, pc chips, etc.
This is were windows still has an edge over linux. Windows installs, linux does not.
Oh yeah and I forgot the most important part.
1. Network driver support.
2. Network driver support.
3. Network driver support.
If linux does install, but is missing a few drivers, at least it could download new drivers from the internet, but then the cd/dvd installation must come with all possible network chips/driver support. Otherwise this will leave linux stranded/useless for todays internet world.
The elephant in the room is the sound subsystem. Umpteen sodding layers of incompatible nonsense on top of a vastly inferior API. Getting applications to work seamlessly with ALSA+Portaudio/JACK/Pulse/ARTS/Phonon is a bloody nightmare. Meanwhile, on the "dead" operating system that is FreeBSD, we have virtualised multi-stream concurrent playback out of the box.
Every time I touch a Linux desktop the audio subsystem gives me a few more grey hairs. It needs ripping up and starting from scratch or OSS brought back with the same codebase that FreeBSD uses. That won't make me switch but you'll stand a better chance of your fabled "year of the Linux desktop" if the audio applications actually work reliably.
Here is how l see it.
Linux is good for Linux gurus (who live in the terminal and use vi). lt is also became good enough in recent years for simpleton users who just want to browse the web (watch youtube, write mail...).
But the problem is they are not going to install Linux (or Windows for that matter). Because "I'm just not adept/understand/know...". (not sure how it is in English, i only hear it in Hungarian) Even though they could. Because they all know intermediate users they can ask. And that's where it's at. We intermediate users install maybe 90 percent of the desktops. So if you want desktops then you need to win us over. And you are not...
I'm an embedded hardware/software engineer. When I was young I was playing with OSes (DOS, Win, qnx, OS/2 warp, BeOS...), but now I want to *use* my desktop (for work) and not play with it (maybe play on it). If the OS is not working properly it is preventing me from doing my job, and I'm loosing money (and/or my free time and my sanity).
Linux Desktop (ubuntu) LTS is a joke. WinXP was around for a decade, that is LTS. I don't have time to learn new distributions, UIs... every year. And the ever present or growing inconsistency does not help either.
So who are intermediate users? Basically we can use Google (or Bing, Yahoo). That's it. And when I search for "how to ..." I get maybe 10 percent success rate or even less. Because either the top result is a forum, where the same question is asked but there is no solution, or there are many solutions that none of work, the result is for a different distribution, same distribution but different incompatible version... (RTFM does not help because I do not have time to read 10 new ones every year)
(for example: configure (multiple) static IP on Linux: /etc/network/interfaces right? well that's what it says if you are debian based, not even gonna start with RHEL, but on raspbmc the file is so fucking empty and does not do anything, and on my ubuntu it is also does not contain eth0 (and eth1) because of the network manager. Configuring monitor resolution (permanently), xrandr? edit /etc/X11/xorg.conf? well you mean create one. Would it be so fucking hard to include a sample one in every distribution where you can just uncomment lines? Run something on startup: sys-v init.d, rc.d, upstart, lsb script, service, update-rc.d... Ubuntu 5 had GUI group (and user) management, they removed it since then. Congratulation.)
I could continue, but I have to work now...
Maybe if developers would focus more on stability, fixing bugs and developing useful tools, features we need rather than rewriting everything (Unity, that is useless on VMs)... (I know, it's less exciting to write production quality code than writing something new.) Maybe one day.
I really like the quote I found on Coding Horror blog:
quote from Havoc Pennington:
"It would be wonderful discipline for any software dev team serious about Linux 'on the desktop' (whatever that means) to ban their own use of terminals. Of course, none of us have ever done this, and that explains a lot about the resulting products."
Really? No one complains about device drivers on Linux? Where have you been the last 20 years?
Please, remind me how that's done in windows 8.1. Feel free to explain differences with windows 8, 7 , XP.
I love it when troll like you label others. Yay, hypocrisy! So you found 1 thing on Ubuntu that was easy. While that is an accomplishment, that's a pretty lonely anecdote. Try changing video drivers without touching a console. Ouch!
Since you were asking, Windows XP, Vista, 7, and 8 all let you change the IP address in the following way.
1. Right click the network tray icon, select "Open Network Center"
2. Click on the desired network adaptor, Ethernet or wireless.
3. Click properties.
4. Double click desired Protocol (ipv4/v6)
5. Input the desired manual address.
It's clear you haven't used Windows in a very long time, which is evident in the way you troll. Your information is out of date and inaccurate.
Windows Vista's workflow isn't covered in your description by the way and some systems don't even display the networking icon in the tray by default on clean installs (unlike Ubuntu).
Change is certain; progress is not obligatory.
"Please, remind me how that's done in windows 8.1. Feel free to explain differences with windows 8, 7 , XP."
Windows 8.1 - move mouse to top/bottom right corner, click settings, click control panel, click network, click manage adapters, click the adapter you want to edit > click tcp/ipv4 > click properties > change IP.
Same for 8, 7, Vista. XP is generally the same.
Feel better now? You're just as much of a troll as you claimed the person above you to be.
Torvalds, and his audience, are still delusional - Linux on the Desktop won't ever happen.
I use Windows 8 at one of my computers at home.
Those instructions don't work there. You need to do some additional steps before, to summon the "Open Network Center" option.
In any case, what I was responding to the troll was not that Ubuntu has a great magic and beautiful way of changing the IP, only that it doesn't require a console.
Like you explained, in most Windows versions, the process is more or less the same, which was my point.
As much as I hate to say this, the real reason why Linux will never be a viable desktop replacement is that there is a lack of standardization and the open source above all else attitude.
If you want people to come to Linux, you need programs for them to use. Not open source programs that they've never heard of, but the same programs they use on Windows. Developers of these programs don't want to support an ecosystem where they're shunned and where it's comparatively harder to maintain on Linux due to any number of factors.
If you want a real user base it's time to stop expecting that you're going to win them over with the benefits such as free as in speech. It just doesn't speak to the majority of computer users out there because the only thing it changes for them is that they're able to see and study something which they cant comprehend in the least.
Basically, the 'desktop' is more than just an interface and an operating system, it's a word that describes a combination of solutions working together that people want to do stuff and get jobs done. - The average person today does not have time to waste worrying about the operating system, they just want to run stuff. plug in stuff, fast and easily and it will work. Just like any house hold appliance.
20 years ago, it was 'normal' for people to tinker with their car and more than 30+ years ago it was 'critical' to tinker the car. Now, NO ONE tinkers with their car, seriously, no one does. Right now, cars just start flawlessly, you can hardly hear the motor, the heating/cooling works in less than 2 minutes, things just work. What was once an extra is now normal, GPS, Blutooth, MP3/CD, Aircon etc.
If things go wrong you just take it to the repair center and they can easily fix things and or get spare parts in a few days cheap. If you like cars you get something old to tinker with, you know, an old benz or bentley, even an old skyline, what ever, but 'those' people are the 'linux' people. You know, the nerds, that just want to make work for themselves... for their own pleasure etc. So while I hate car analogies, this is not 1989 or 1997 anymore, people just want to run a game, plug in their camera, print, scan some crap, backup their phone, run random programs easily...rip a dvd, buy a few songs on their itunes or amazon, download some pirate crap and move on. They don't have time to waste, there is dinner to cook, kids to feed etc, just want to connect stuff they got from the shop and it all works, and well, like the car, you know?
The 'Desktop' has evolved - The old days of struggling to get the neighbors return to zork to run on their cheap HP desktop with a soundblaster compatible with poor config.sys settings... have passed. The days of Amiga vs PC and Linux vs Windows have passed. Seriously. Everyone has moved on.
This is WHY Linux for the 'Desktop' currently has no chance.
* Windows *
- Install windows 8 on an average i5/i7 it boots up fast (win 8 in less than 15 seconds).
- Run your desired program and it runs. - As in, run setup.exe or the application.exe. (you know, really easy no fuss).
- Plug stuff in such as Printers, Scanners, Joysticks, Gamepad, etc all work.
- Basic things work such as Sound, Wifi, Networking, Bluetooth, Video card.
- If you have trouble, you can google the solution and find the solution(s) all with a pretty much consistent approach.
- Website plugins work - for example Flash, Silverlight, Java. All downloadable in seconds. (regardless of opinions if they are evil or not, you just want to run some damm flash game for the kids etc).
- Spend time with partner, family, go out for a drink. Come home later with friends, plug in smart phone download music, plug into the media player play music.
- Move on?
*Linux *
- Install a random linux distro, usually the opinion of a friend or work college or google will decide this.
- You boot into a flickery half assed poor mans GUI that makes windows 95 of all things still feel more polished.
- Your wifi won't connect, so you plug the Ethernet cable in, it's does not work, you have a router with DHCP and everything else in house is working, but your little Linux experiment.
- You use another PC in the house to get drivers and instructions on getting things to even network..
- You plug in your USB stick and it's not showing up, so you google how to get this to work.
- Hours later you manage to get your network running, great.
- While you think all you need is a basic web browser and mail client, you realise you do more than * just * this. (so sick of hearing this crap).
- So, you manage to screw around for a few days, you still can't print, you have no sound, you find various one man show solutions that have not been updated since 2005..
- Forget running mature professionally published software, you need to find crappy substitutes that the devs lost interest 4 years ago a
That might have been true 10 years ago, however today there are plenty of flavors that are very easy and default install just fine with little or no interaction from the user.
The main stumbling block is install base and compatibility. Windows has as big a strangle hold than ever. Apple has made some strides, however with their expensive machines, they will only ever be niche players. Where they have been loosing and where it may transition into a loss for them is the tablet and phone markets which are all basically iOS and Linux (in Android). If computing transitions along those lines, windows will eventually lose. Which I am sure why they made the ill conceived leap with Windows 8 and the Metro interface to try and get ahead of any convergence that may take place in the future.
The issue with today is that they have such a large install base that is not compatible with linux. I presume that is why Red Hat bought WordPerfect software back in the day, as it was the only one that challenged MS on their home office turf. Problem was it was too little too late as it was already on the way out. Before that I recall trying use wordperfect files in office all the time. So Office compatibility is one issue on the business side, and on the other you have the gaming issues on the entertainment side. However players like Steam may have an impact in this regard depending on how their plans go.
People in business buy Windows because of largely office and other windows only business related software (on the desktop, not servers). People that buy for gaming get windows because most games are only compatible with windows. Everyone else (who you are talking about) pretty much buy windows because it is really the only thing available (other than iOS if you have the $$$). It used to be that common users might be more comfortable with windows, however MS pretty much killed that advantage with windows 8.
On top of all this, what baffles me, is that Windows as an integrated media player is truly horrible with Windows Media Player. I have no idea why this is the case, a company like MS *should* be able to make a decent video player, but they do not. Software like VLC are becoming much better alternatives, than trying to break your system installing malware loaded "codec packs" in an attempt to fix their broken media player. I can only surmise that MS makes it intentionally broken in an attempt to only support official codecs that they can load DRM for the media companies... however is that business worth flushing their brand down the toilet?
Anyway linux while not there yet, and not a lock for surpassing MS, has some opportunities to do so. However likely it would take a large company (like Google say partnering with Steam) to really put a nail in the coffin. Pushing things like Google Docs and the like for office compatibility and transition (particularly when MS starts pushing their Office 365 BS), also strengthening the media software as more and more people use their devices as a connection to their TV, while getting Steam to offer an easy conduit for linux games and developers to market them to a growing user base. Once you have people that are used to it at the office, as a media device, as a gaming device, the common light users will start having more options, and have more people used to the UI. Also if using similar devices say on tablets and phones, this will also raise your common users comfort level, particularly as the demographic that grew up with smart phones start maturing...
So while I don't see it happening anytime soon, it is defiantly something that is possible over time should all the ducks line up in a row.
Or an adapter framework to just load WDM drivers... (that would only prove to demonstrate how much the userspace stack blows however)
How it's done in Windows 8.1:
1.) Right-click the network icon in the System Tray and select: "Open Network and Sharing Center"
2.) Click on the active connection and select Properties in the window that appears
3.) Select: "Internet Protocol Version 4" (or 6 for IP6) and click Properties
4.) Set your desired static IP/DNS settings
Easy-peasy, any admin prompt requires nothing more than clicking "Yes".
In general, both Ubuntu and Windows 8.1 are on-par with basic functionality, the only problem remaining with Linux distros, such as Ubuntu, is that they still require having to use the terminal for certain situations, to me at least, having to use the terminal is an instant fail in usability. I still run Ubuntu on my home server but even a recent clean install of Ubuntu 14.04 required manually editing the GRUB boot line and having to use the terminal to edit the shortcut for KVPM so it would prompt for the admin password when launched (it won't launch without admin rights and the only alternative to editing the shortcut is to launch it from the terminal with SUDO every single time you want to run it). It's these little headaches that keep me from committing 100% to Linux on all my devices. Windows 8.1 just works.
On the bright side, if Steam OS gains traction from game developers I may just make the switch sooner than later if I can get enough games working under Linux to make it worth while. The recent announcement from GOG.com to finally support Linux also has me excited.
Your definition of an OS is quite narrow, overly so.
manage memory
Check. Android's Java runtime environment does this for applications.
It is a safe bet that you have never had anything to do with operating system design or implementation. Apparently, you do not understand even elementary principles of operating system memory management. So... according to you, how does Java manage the process page tables? It is people like you who make the world save for marketdroids.
Actually I have a background in writing low level kernels, in porting c runtime environments to these custom environments. I know about memory management from the hardware up.
What I don't have is an overly narrow concept of operating systems, a viewpoint stuck on some quiz once taken in an operating system class that expected a student to regurgitate a 1970s list of OS components.
Android is no less of an OS for delegating some low level operations to the host linux kernel than a microkernel based OS that delegates some low level functions to its microkernel. Is Debian no longer an OS when it delegates low level functions to HURD?
If you didn't already know about this, and didn't have a network connection, how would you discover this?
(And yes, the same complaint holds true for linux as well....)
Windows is seeming to be actually going backwards.
For example, if you disable hibernation in Windows 7, you can only re-enable it with a command line tool rather than a GUI like it was done in XP.
Or even worse: Windows Vista/7 had network management features that recognized networks and allowed to enable features based on the assigned network type. This is a neat feature which automatically enforces stricter firewall rules in public hotspots. Windows 8 had this feature really dumbed down, and what's more, you can now only manage locations with a command line tool! If you shared some files at a local Starbucks, locking it down would be extremely difficult.
With this rate, some future Windows version would only allow DHCP auto-configuration, or if you need to set your own IP/DNS, you're a power user and should use the console.
For all the talk and dreaming about the Linux desktop you'd think by now the community would have designed a GUI that doesn't frustrate end users.
Where would all that smug go if there was a decent and usable Linux desktop ?
Sounds like a cry from a dying man.
Bored of his previous achievements reaching out for one unattainable goal.
It wont happen. It is that simple.
Actually I have a background in writing low level kernels, in porting c runtime environments to these custom environments. I know about memory management from the hardware up.
Then how could you possibly have confused operating system level memory management with garbage collecting? I am not sure that I would want you working on the Linux kernel, certainly not on the core.
Only you are mixing alloc/new and garbage collection. My point is that whether an operating system offers malloc/new or garbage collection (manual or automatic memory management) to its applications is irrelevant. In either case it ultimately drills down to a kernel, and this drilling down to a kernel is irrelevant to the applications. The application code could care less if the kernel is linux, bsd, mach, hurd, etc ... The kernel is as abstract and as irrelevant to the application code as the hardware itself.
What I don't have is an overly narrow concept of operating systems, a viewpoint stuck on some quiz once taken in an operating system class that expected a student to regurgitate a 1970s list of OS components.
The term "operating system" was recently coopted by marketdroids and PHBs who have not got the faintest clue of what a timer wheel is, to mean something convenient for Apple and Google's respective business plans. Please go get any operating system text, including a recent one, and you will find that the classic meaning of "operating system" is still the only one taught in the schools that produce our kernel engineers.
Your are wrong. Even Andrew Tanengaum says that the definition of an operating systems is fuzzy because it does several different things. Abstracts the hardware, manages resources and provides an API that application programs are written for. That operating systems have evolved and the simple kernel/user mode distinction of years past no longer works. That it is legitimate to define an operating system from both bottom up and top down perspectives. You are simply taking a narrow bottom up perspective, and a further narrowed monolithic perspective on top of that. Android provides an API, it manages memory (automatically), it schedules threads, it performs I/O, etc. Android fits one of Tanenbaum's definitions of an operating system. Tanenbaum specifically refers to Java-based operating systems.
Android is no less of an OS for delegating some low level operations to the host linux kernel than a microkernel based OS that delegates some low level functions to its microkernel.
You seem not to grasp the scale, power or subtlty of "some low level operations" that Android relies on the operating system for.
You guess wrong yet again. Its not the utility of the kernel that matters. Its the visibility of the kernel, the necessity of one particular kernel. Linux is an implementation detail under Android. Android could be ported to use a different POSIX based kernel and applications would not know or care. Why? Because Android is an operating system from an application perspective.
Debian is no longer an OS when it delegates low level functions to HURD?
Debian is referred to by Debian developers as a "distribution". That is exactly what Android is, nothing more and nothing less.
That is an amusing dodge. What developer's call their software only matters when it fits your narrow definition.
At this point Linux Desktop just needs advertising. It needs to be sold in preinstalled machines with the most user friendly tweaks already applied (Ubuntu out of the box still needs tweaks), advertised, and sold. It should be marketed as PC replacement that is not a PC, even the idea of a PC needs to replaced in peoples minds since that is so deeply ingrained in the idea of Windows. It's a different computer, one without viruses or secrets. It is freedom in a box. Oh, and that box needs to be a pretty one.
The commands to the bus don't change.
The commands sent to the hardware don't change.
The internal logic won't change.
That leaves the specific hooks to the OS and the externally visible structures.
Nobody is insane enough to use globals directly and structures are subject to change without notice. So external stuff will already be isolated.
If the hardware is available for any two of HyperTransport, PCI Express 2.x, VME/VXI or one of the low-power busses used on mobile hand-warmers, err, smart devices, then the actual calls to the bus hardware will be compartmentalized or go through an OS-based abstraction layer.
So 95% of a well-written driver is OS-agnostic and the remaining 5% is already is isolated.
So either drivers are very badly written (which is a crime against sanity) or the hardware vendor could place the OS-dependent code in its own DLL at bugger-all cost to them. Since the OS-dependent code has nothing trade secret in it, they can publish the source for the shim at no risk. Since the shim isn't the driver, there's no implication of support for OS' they don't know or understand. It's not their problem what the shim is used for.
Everyone's happy. Well, happier. The companies don't get harassed, the Linux users get their drivers, Microsoft gets fewer complaints about badly-written drivers killing their software. It's not open, it's not supported, but it's good enough.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
The problem is corruption. OSDL were working on a Linux desktop environment, but a key (financial) figure in the organization worked hard to kill off success and left around the time the unit went bankrupt. Several organizations they've been linked to have either gone belly up or have suffered catastrophic failure.
I won't name names, no point. What is the point is that such people exist in the Linux community at all, parasites that destroy good engineering and good work for some personal benefit of their own.
X is not great, but it's just a specification. People have developed Postscript-based GUIs using it. It's merely an API that you can implement as you like (someone ported it to Java) and extend as you like (Sun did that all the time). The reference implementation is just that. Interoperability of just that set of functions used by Glib/Gtk and Qt would give you almost all the key software.
Alternatively, write a GUI that has a port of those three libraries. You could use Berlin as a starting point, or build off Linux framebuffers, or perhaps use SDL, or write something unique. If it supports software needing those libraries, then almost everything in actual use will be usable and almost everything written around X in the future will also be usable. If what you write is better than X, people will switch.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
OK, let me give you a roadmap to take the desktop (because its possible):
Step 1: The desktop market is the business market. These folks want to get the job done, not futz with a computer. A distro designed for businesses would have one or two options for the various systems, and then support the hell out of that.
Step 2: Fix configuration hell. Develop a pluggable settings utility with basic logic to let end users change basic settings.
Example: SANE. I wrote https://help.ubuntu.com/community/sane and other sane docs, so its not like I don't know that system well. But you often have to edit numerous text files to make it work (if your scanner is not auto-detected). A settings utility that gives you an "add local scanner" button, scans local usb ports for a scanner, then lets you select the back-end and model (enabling the backend, adding the vendor and product ID's to the right place) so the scanner just works would be an amazing .
Step 3: Fix SOHO networking. It should not take me 2 days and dozens of textfiles to get ldap/nfs working properly. It needs to be as easy to set up as Windows Server is. Same goes for the client side. It should be trivial to join a workstation to a linux domain, getting LDAP authentication, mounting NFS shares, mapping CUPS and SANE servers, publishing CUPS/SANE resources in the ldap directory, getting the SQL server, time server, populating mail settings from LDAP, key server, SIP server, repo and puppet-like service for making admin easy and other pluggable services. You should be able to tell your new workstation where the configuration server is, and the rest should just work (and in most SOHO systems, the config server is the general server used for most everything).
Step 4: Fix defaults to be sane choices for SOHO. CUPS should default to showing printers published on the LAN. SANE should default to showing scanners shared on the LAN.
Step 5: Pick the default applications. Desktop users will support each other, but only if they are using the same apps. We don't need to give the average user 300 different text editors. Pick one as the default. We don't need 5 POS apps. Pick one as the default.
Step 6: Build a user support structure. This should be built in to all applications, and should be EASY to access and use (for the users).
Step 7: Learn to work with commercial software vendors. Like it or not, we will never have an open source app for running a pest control company. It's not a large enough market for that. Lots of other examples of that too. Making it EASY for an ISV to develop for this system is key to getting the desktop.
Step 8: Make real documentation. What we have sucks (that from a guy that writes some of our documentation).
Linus, if your reading this and are serious about taking the desktop, contact me.
4 years ago, some automatic update once again broke my Xorg.conf and display drivers. Since then I've been using windows. Can't remember configuring it at all. It just magically works. The good news is windows doesn't even crash anymore. I also concidered going OSX, but games work better on windows, I'm not locked to apple hardware (I like to assemble my own computers), and from a user perspective OSX doesn't really offer anything but prettier launcher for programs.
Even if the Desktop has transformed itself into a (virtual desktop) "Workspace", I want the OS on my laptop, my pad, my phone....
I want it running E17, or another light DM/WM
I want it to run my android/PlayStore stuff
sandboxed in a container or a VM.
I want it to run encryped out of the box:
Selinux, shorewall, gpg email, encrypted FS,
TOR/I2P browsing.....
To Linus: met you at UNH in 93or94; what a ride
it's been.
More please, thank-you
resist propaganda
Idiot.
When all you have is a hammer, every problem starts to look like a thumb.
Windows Vista was a bigger fiasco and nothing happened.
It's also the fact that you can buy a computer with Windows or with OS X easily but finding a computer with Linux pre-instaled is almost impossible.
They have a long way t ocatch up to windows 7.
Mobile devices have grown, but the desktop will never go completely away. There are a lot of things that mobile devices just can't do, that desktops can. GNU/Linux taking over the desktop? That would be for business and the enterprise, and I'm not sure that is possible as companies are resistant to the unknown, and fear change. It may happen, but nobody knows when. The chances are greater all of the time as more people develop GNU/Linux skills, which are more present than ever and growing every day. So, the shift is going in the right direction. I haven't used Windows on any of my equipment for over 6 years, and I have no reason every to go back to Microsoft.
The committment to stable programming interfaces is one of Microsoft's strengths; They don't all get the 14-year treatment, but at least *some* do.
At this point I think you would need to manage the device from start to finish.
Basically put out limited line like Apple. Use high quality and standard chipset. I mean like a good ethernet chip, a good sound card, etc. With a standard build it is way easier to test your OS and make sure everything functions on your various models.
From there take Gnome or KDE and fork it. Go the Linux Mint Gnome route. If you are a company you can just decide how things work. Macs work one way, Windows another, but for the most part they work the way their company's wants to. That standard is what makes them popular, even amongst developer types and the kind of people that go to OSCon.
Then get a good testing community going. Make sure it works with printers. Make sure it works with projectors, make sure it works with dual monitors, make sure it works with the keyboards.
Come out with your own damn keyboards and mice.
Make the upgrade process simple, straightfoward, and automated. Certainly use a package manager, but hide it away.
The community might be able to take it and abstract, but given the linux community they would just tweak it to the point where it isn't as "beautiful".
Ubuntu tried and failed many of things. I think mostly because the people who generally run linux do so because they want to.
The difference here is that someone will walk into best buy, login to amazon, or your own site, maybe a dedicated store (think apple store) and walk out with a device that runs an OS which happens Linux...not buy some hardware and "try" to get linux working on it.
I don't think HP, Dell, etc have any interest in this and so someone needs to start it. Maybe we can leverage open hardware, maybe not, but I think we need to replicate the Apple model.
From there the hard part. Get the gaming community behind you, get Office to run on it...I mean the real Office, and keep going to get the world to treat your product with respect.