Slashdot Mirror
Why are Free-Desktop Developers Wedded to Linux?
An anonymous reader wonders: "We have been hearing promising predictions like 'This year will be the year of Linux on the desktop' for the last decade. However, the Linux of today seems to be as far away as ever from realizing the expectations of mass adoption we once had for it, without significant growth in home usage since the late 90s. Clearly, if Linux is unable to reproduce a third of Firefox's end user uptake over a much longer time-frame, there are deficiencies with the direction the GNU/Linux/X/Gnome/KDE system has taken. Of course, almost all free software and desktop efforts and development remain unquestioningly oriented around Linux.
Other free-desktop operating system projects which take different and innovative approaches like ReactOS, AROS, Mona and Syllable remain comparatively starved of developers and interest. An often cited reason for using a non-Microsoft OS is to avoid a monoculture, but free-desktop efforts have created a total monoculture around developing and promoting Linux, despite a decade of failure in supplanting Microsoft's proprietorial OSes with it. Why are free-desktop developers neglecting to consider an alternative to the penguin?"
So, what's BSD then, chopped liver?
--- These are not words: wierd, genious, rediculous
This is going to be the year of the penguin
It is very unlikely that developers follow Linux only.
They support some well documented and mature standards like Gnu Libc, X window and POSIX, among others.
Infact, for example, most of the desktop software can be compiled and run under almost all OS that comply to those standards.
Sometimes even under Microsoft's OSs.
Maybe Computers will never be as intelligent as Humans.
For sure they won't ever become so stupid. [VR-1988]
To choose exactly the same thing as they do.
I know this will get troll/flamebait, this community does not like criticism, even though taking it into account could cause improvements.
Seriously though, the thoughts are this:
(1) They are enamored by the GPL license. I'll grant for certain uses and purposes, it's an excellent license, even if I don't agree with it.
(2) Momentum - Linux is the first OSS OS to gain popularity, and it hit it off big for such things. What this means is that it has more support and developers, which provides a more feature-filled system which brings the people and culture more of what they want.
(3) Flexibility - I'm not sure the whole background of it, partially it's the GPL, partially it's the management, but the Linux system is highly flexible in terms of development, allowing people to develop their projects how they want to. Especially at the kernel level. It may not be a coders dream environment, but it's pretty close.
(4) UNIX Like. I know ReactOS isn't Unix like, I don't know about the others. I know BSD, which you didn't mention, but lacks 1-3 is also a Unix OS. Regardless, the Unix methidologies are very comfortable to developers because (a) they are relatively regular in setup. (b) They tend to be highly modular, making things easier to work with and build - lots of re-use of things you made or thigns others made. B can exist in other OSes as well, but it isn't as pervasive as in the UNIX environments.
Note, there's probably a lot more to it than this, but this is what I've gathered from what I have seen and read on the various topics. and discussions.
34486853790
Connection too slow for X forwarding? Try "ssh -CX user@host"
All the interesting stuff in supplanting Windows in the desktop is in, well, the desktop. The underlying operating system is irrelevent so long as it works, and Linux is going to continue doing that far better than upstart efforts.
Linux is an OS. Firefox is a desktop application. An OS differs from an application in many ways, including ease of installation and the impact to the rest of the desktop.
Perhaps this suggests "alternative OSs" should make it even easier to make use of virtualization on "popular OSs" (LiveCDs are popular & this would be the next logical step).
Of course the way to find the adoption of any software is difficult & the ways people look at browser usage compared to OS usage often differ.
Firefox can run on many OSs, including Linux. Unless another browser becomes very dominant on Linux or Firefox becomes unpopular in other OSs, it isn't a good point of comparison.
The fact that a browser was the basis for comparison is telling--server-side apps are becoming more important & many of these do run on Linux.
Some do consider alternatives, and that is why programs like ReactOS exist. Most of the smaller alternatives aren't really designed to be desktop replacements for the world, but rather small niche desktops. Of all the alternatives, Linux is the best candidate to supplant Windows on someone's PC.
Firefox used aggressive marketing in quick blitz. It had a great name. And Firefox had rapid growth because of that.
Linux is associated with geeks and carries plenty of negative baggage with the average person. When Mozilla became Firefox, it was able to be reborn in a marketing perspective, and may someday win the fight that Mozilla never could.
If Linux gets a similiar marketing facelift, you could see similiar adoption rates that Firefox had. It is a much bigger adjustment for people, but in the wake of Vista, more people may be looking for alternatives. However, the majority of the Linux community is quite content to cater to themselves rather than try to cater to the outside market. For mass consumption you would really need:
1 major primary distro for home users.
1 major desktop
Easy conversion wizard to help people convert Microsoft documents, desktops and settings.
1 major form of package management, and thusly one major package repository
Remember the GetFirefox.com campaign? Remember all the CDs thrown around?
Imagine a LiveCD distribution campaign that did the same thing, but also helped you convert/migrate? Give it a snappy name, and a cute mascot and there you go!
http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
Here's the Flamebait/Insightful reason why Linux will never be a desktop OS: 99% of the development is driven by developers. Developers are geeks. Developers have their friends and the rest of the OSS community test their stuff. If they ran it by their grandmothers once in awhile maybe we'd make some headway...
I disagree "free-desktop efforts have created a total monoculture around developing and promoting Linux" because KDE, Blackbox, XFCE, etc, etc.. all compile on pretty much any implementation of Unix, of which Linux is just a clone. Solaris now runs Gnome (branded as Java Desktop System) as the prefered desktop.
Unix is probably popular with developers because it is "open" and standardized in the specifications and widely know and taught in computer science departments.
So the "failure" to catch on is wider than Linux. Solaris/SunOS alone has been deployed in probably every large corporation in America and Western Europe since the '80s, but has never broken out of the specialist server/workstation market and into the general desktop market. And during all that time, SunOS/Solaris has gone from OpenLook, to CDE, to Gnome. The various X-Windows desktops really didn't get off the ground in a meaningful way until the mid-1990's with CDE (which was announced in 1993, I first saw it myself in 1996 on HP-UX), by which time Win3.1 and Win95 were already entrenched. Also, compare Win95 and FVWM circa 1995, and you'll see why Windows was the only desktop game in town at the time.
Windows owes it sucess to the ubiquity of MS-DOS in the 1980s-early 1990s. MS-DOS owed its ubiquity to the "street-credit" granted to it by IBM's endorsement. Had IBM implemented their PC with Xenix or some flavor of Unix capable of running on an 8088, then we would all have unix desktops.
I think the simple answer is critical mass: you need a sufficient number of developers developing not just the platform itself, but applications to run on it. Without a sufficient base of applications you're going to inevitably be perceived as a minority player and fail to attract many users, and hence many extra developers. Past a certain threshold you can be roughly self sustaining - Linux is across it, and so is MacOS X, but I don't think the projects you mention are even close. There is simply too much software built up on the GNU and X11 toolchain (and increasingly on GNOME and KDE) that people would have to leave behind to move to a new open source OS - it just isn't that tempting when the alternatives look so application poor.
To succeed you really need some base to start with (as Apple had when they moved to MacOS X, although even they lean on X11 and apps built on the GNU toolchain to some extent), or you need to support the toolchains of the applications (see OpenSolaris and BSD, which lean on X11, GNOME, KDE, etc.). Depending on what it is you wish to get rid off things can go from easy to very hard. Just ditching the Linux kernel is feasible - see the BSD and OpenSolaris options, among others. If you want to get rid of X11 as well... well that's trickier, but if you have a graphics system that will run GTK+ and or QT you might get by because you'll still have the rich supply of GTK+ apps, and can probably get KDE ported. If you want to ditch everything up to GNOME and KDE... well that means rewriting all your applications from scratch, and really that's a huge and incredibly daunting task. It's not just the big applications like web browsers and email clients, its all the different little niche applications that make the environment so rich. Its that that keeps many people on Windows - the one little application that few other people have any interest in, but happens to be vital to them; because everyone has a slightly different vital niche program it adds up to a lot of applications to reproduce before you can truly draw a large user base.
Linux has crossed the first threshold: it has enough users and application developers working on it that its self sustaining. It has yet to cross the next threshold where it provides a rich enough ecosystem of applications to entice the myriad of home users. It is, however, slowly crawling toward that goal.
Craft Beer Programming T-shirts
You have posted two questions, why are all free software developers headed towards Linux and why Linux has not supplanted Windows as a Desktop OS..
Answers:
1) Most free software developers I know gravitate towards standards, not an OS. Their programs will run well on a GNU BSD system and cygwin. That's their goal. Every developer whose motivation for development is philanthropy or ego will aim to maximize compatibility rather than being exclusive to Linux.
2) Linux cannot take over the desktop for a few simple reasons. First and foremost is the lack of standards. Theres gnome AND kde. And there are several popular distros to develop and test for to make sure installation is smooth and seamless like in Windows. Windows is a single distro and extremely predictable in that regard. Developing and deploying a desktop app for it is much easier.
Secondly there is a lack of opensourced drivers and directx doesnt exist. DirectX makes things much easier than opengl plus other api.
Once a real and effective standard is settled upon in Linux (api, distro, installation and package maintenance mechanism) I suspect Linux would be much more popular on the layman's desktop.
"Give orange me give eat orange me eat orange give me eat orange give me you." -Nim Chimpsky
This is a bit of a red herring. Firefox gained market share for a number of reasons, some that may be applicable to Linux as well. But the single biggest reason for Firefox's market share is that you could install it on Windows.
Have you ever tried installing Windows from scratch? That is like two days effort (by the time you get all your drivers and programs installed, and everything set up as you like). I don't think Linux is worse than Windows, just different. And for certain setups, its better (consider all the good programs that are already available by most distros default install).
So, the main reason Firefox gained so much popularity compared to Linux, was that you could use it on whatever OS you were already using. Possibly this includes it being "so easy to install and use", but that is a misleading statement because you are implying a Linux distro isn't. Firefox installs like any other application on any supported OS, and is as usable as most mature programs. Linux distros are likely the easiest operating systems to install, but that doesn't really matter, because most people will never install an OS. Linux is quite usable, as long as you don't expect it to be the same as Windows or OSX and are willing to get used to it.
Be who you are and say what you feel, because those who mind don't matter and those who matter don't mind. - Dr. Seus
The premise of the question is that Linux' lack of desktop market penetration indicates some failing with Linux. I think that premise is flawed. I think Linux has achieved more desktop market share than could reasonably be expected in the time elapsed, and that all of those who have predicted more widespread use were simply fooling themselves.
See, every bit of desktop market share that Linux achieves must be taken away from the Microsoft desktop monopoly (plus maybe a bit from Apple, but that's a tiny corner of the market and one that is very hard to crack). That means that Linux has to deal with the fact that pretty much all of the desktop software in the world, and all of the PC hardware in the world, is built for and around Microsoft Windows.
Look, for example, at the reasons why people here on /. commonly say that they don't want to (or can't) switch to Linux:
Looking beyond the slashdot crowd to the more general PC user base, Linux has another, even bigger obstacle: Most people don't install their own operating system, ever. They buy a PC with an OS already on it, and that's what they use. What OS comes on every PC on the shelf? The latest version of Microsoft Windows, of course.
Given that these are the real problems holding back widespread desktop adoption of Linux, what is some other OS, that supports less hardware and has less software available, and even less mindshare among PC vendors going to do to fix the problem?
Not a damned thing, obviously.
Desktop Linux will make its breakout, if it does, in exactly the same way that Desktop DOS and Windows achieved theirs -- via the business desktop. In the more-controlled corporate environment, where hardware is less varied, the IT support staff is better educated (i.e., there is an IT support staff), application sets are more limited (e.g. no games), and there is a stronger focus on cost containment and security, Linux is beginning to make some inroads, and will continue to make more. Linux is getting serious attention as a preferred desktop platform by governments, both for reasons of openness and for reasons of cost management.
When a significant percentage of the world's desktop PC users use Linux at work, then you'll start to see significant home market penetration as well. And that business desktop penetration is happening, but it's going to be a long, slow process because it's a fight against a very deeply entrenched and very powerful monopoly.
I think Linux is doing an excellent job of getting there. The Free desktop environments and application suites are in excellent shape, and are continuing to improve rapidly. I think KDE and GNOME are both much *more* usable than MS Windows, each in their own way, and I can cite numerous Free applications that rival or even exceed the best of their commercial competition. Linux is *ready* for the desktop, and has been for quite some time. But being ready isn't enough to displace Windows. There have to be other advantages, to counter the massive juggernaut of Windows inertia. And there *are* other advantages, but even so, it will take time. Lots of time.
People don't focus much on the other a
Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
Don't forget Haiku, the free BeOS reimplementation. What's been done so far is impressive for the number of developers working on it; if a few more developers joined the progress, I (personally, IMHO) think R1 could happen this year.