I found this one via HackerNews – a 2003 article on what Linux needs for “world domination”, written by Hugh Fisher.
If Linux is to achieve world domination, it must have One Frickin’ User Interface (1FUI): a single user experience / interface behaviour and a single underlying UI toolkit API / widget set. World domination means putting Linux into corporations, schools, PDAs, and cell phones. This will only happen with 1FUI, and if this upsets the nerds, too bad. History clearly shows that if a platform/system offers a choice of user interfaces, the potential users will choose a different system.
It’s almost 2015 now, and it turns out he was right. That “1FUI” is called whatever Android has, and it has made Linux the dominant player in the next big computer revolution. Linux does great in servers, embedded stuff, supercomputing, and utterly owns mobile computing (Apple people, the world is bigger than the US, UK, and Australia).
Linux didn’t need a ‘year of desktop Linux’ after all.
I view Android as just one UI among many. The difference between Android and i.e. KDE is that Android doesn’t have any X within it… but in the end it is just a Java based GUI on top of Linux. One out of many.
Linux distributions that made a choice between KDE and GNOME have not conquered the world. They are as much a counter example against the 1FUI theorem as that Android is a confirmation.
The original poster is still correct. Linux will never be a desktop alternative because it is still ruled by geeks. Don’t get me wrong. I’m a geek too. However…I have work to do. I don’t have time anymore nor the need to tweek and phreak my system 18 hours a day. I settled on KDE as my GUI 10 years ago as it is STILL the only Linux GUI that offers a holistic, complete, desktop software stack (ala Windows and Mac) along with a more mature and complete toolkit in the form of QT. Yes….I also use many GTK+ and GNOME flavored apps. But the geeks still rule the GTK+ and GNOME world too much and are too immature to realize they are hurting Linux adoption with their attitudes on radical design changes (hint….you’re not supposed to be Windows 8)….and of course this had led to GNOME fragmentation with MATE and Cinnamon to compete with GNOME 3. New people looking to Linux don’t want this. REAL people don’t want this. EXPERIENCED users such as myself don’t want this. Which is why we keep losing new users to the Mac when they have finally had enough of Windows and are looking for an alternative.
And please…..the self serving triumphalism of the article writer that we don’t need a “year of the Desktop Linux” because Linux rules the server space where desktops are not as important and mobile through Android is silly and juvenile. First of all the mobile space is disposable. Cell phones and tablets get dropped and broken all the time. You’re contract ends and you get a new phone. Etc…etc. The desktop whether it be a true desktop or a laptop ( I know…it’s “mobile” too….but most folks who replaced their Desktop used their laptop as a space saving desktop with the ability to take it to the couch if need be. ) But it’s still a desktop….or in other words….your “WORKSTATION”. The workstation will always be where you get most of your “WORK” done. Your mobile phone or tablet is an APPENDAGE of your workstation, much like your fingers on your hand and your hand to your brain.
And the only way to win the “WORKSTATION” battle is to have One Frickin’ UI. Even in the mobile space it is the same. The only reason Android dominates Apple when it comes to pure market share is the cheapness of most Android phones. But most folks do not like their various Android experience when they finally use an iPhone. Why? Most Android phones do not have a consistent user experience because the various manufacturers of phones and even the phone companies want to put their own crap on top of the Android GUI to fool you into thinking that their is something MUCH MUCH better on THEIR overpriced Android phone as opposed to the low end plain vanilla Android which does the very same thing. Thus leading to “FRAGMENTATION”…..which is the thing that STILL plagues desktop (“workstation”) Linux after all these years.
Oh well…..the geeks will have a field day with this. That’s ok. I have work to do which includes getting my family and friends off Windows and even their 10 year old Macs because they can’t afford to upgrade to the shiniest, newer version of the Mac….and getting them over to Linux. You know how I am doing that? By introducing them to Kubuntu. The best of all Linux worlds. The rich ecosystem of Ubuntu with the standard, consistent UI experience and software stack of KDE. I’ve even converted BACK to Linux some folks who gave Red Hat or some other version of Linux that were put off by the “hacker” vibe they got while trying to get used to it.
Here’s a parting thought. You geeks may actually WANT to never have a “Year of the Desktop Linux” because that will take the “cool” factor away from you if Linux were to ever be as popular as Windows or Apple. But here’s a thought. The geekiest thing a geek can do is to take difficult technology and make it easy for the common person to use and to be productive. It is the height of arrogance to think of the common person as beneath you and not make Linux as easy to use and CONSISTENT of experience as Windows and Apple. The spark of greatness lies in the common man. We need those sparks now more than ever. Make that possible for someone by making a clear…consistent….easy to use GUI and software stack for Linux.
ONE….FRICKIN’…. GUI !!!!!!
There are two problems with this 1FUI arguement.
1. Android has multiple user interfaces. In fact, there is a market for alternative android “desktop/lauchers”. Therefore, if you argue 1FUI is required, then Android proves that wrong. If you argue Android is an example of 1FUI then that shows a bit of ignorance about Android’s audience. Many of the people who use Android use alternaive launchers or customizations to make their Android “desktop” look and act different.
2. A lot of the people I know use Linux on the desktop, for a good chunk of my family and friends it has been Year of the Linux Desktop for several years now. Since most of them use different desktops (KDE and LXDE are popular choices) it’s pretty obvious the diversity in the Linux ecosystem has not stopped them from adopting the platform.
What has kept Linux adoption on the desktop relatively low isn’t diveristy, it is a lack of support from big name OEMs. Dell has relatively few Linux desktop options, likewise HP and Acer haven’t really put any weight into Linux on the desktop. People will run whatever comes pre-installed for them. For most people that is Windows, for a significant percentage it means OS X. Linux just doesn’t have a lot of OEM support and, without it, won’t gain a big desktop market share.
Look at it this way, if almost all phones sold with Windows or BlackBerry OS on them, how many people would install Android or iOS? Probably less than 10%. It’s not the merits of the OS that count for the vast majority of people, it’s not the interface that matters either. It’s what comes pre-installed.
I disagree, I do not think its customizability counters the argument: having a common base that works more-or-less the same on all Android-phones is what enables this behaviour, but if Android-phones were like Linux on the desktop is, ie. one ships with GNOME and another ships with KDE and yet a third one ships with WindowMaker users wouldn’t have that common base, the installation prodecures and whatnot for these launchers would differ from phone to phone.
The argument never was that people would simply use the defaults the OS ships with and therefore your counter-argument that customizability proves it wrong is simply not relevant. Quite to the contrary; a common base enables the ease at which users use the platform, including customizing it.
The problem with saying that Linux has won because most computing devices on earth probably now run some variant of Linux is that it is both true and yet it explains almost nothing of import. It’s like saying Unix has won, true but so what?
Saying Linux has won, because of the number of devices that run a variant of Linux says essentially nothing about the actual dynamics of the computing device tech world. Nothing about the stuff that actually matters.
It explains nothing about the dynamics of money in computing devices, the computing device platform dynamics, the success or failure of the various OEMs and business players, who the winners and the losers are, the roles of computing devices in the enterprise or corporate world, the success or failure of various business models, or how computing devices will evolve in the future. All those very interesting and important questions cannot be answered by looking at the computing device market by looking at the lowest OS abstraction level and counting units.
It’s like an idiot savant who knows the number of matches in a box of matches by just looking, impressive statistical trick but ultimately a useless fact because it explains nothing.
I’m not sure if you replied to a wrong comment or something. I haven’t said anything the sort of Linux having won anything.
Sorry – meant to post as a new comment but posted as reply to you by mistake. Still recovering from Christmas
Oh, okay. I was wondering about that!
That aside, I do agree with you in that that saying “Linux has won” is ignorant, though I base my opinion on Linux technically just being a kernel and the “war” being an on-going process — being on top doesn’t even mean Linux has won a battle, let alone the whole war. The kernel hasn’t even been enthusiast-driven anymore for a long, long time, what with multitudes of companies working on it with profit in mind. At best it’s a great example of what open-source can achieve, but I don’t know if one should be drawing much more from its success.
Nod to PP.
The reason why Linux is now the most dominant OS on the planet is *because* they never settled on “One GUI to rule them all ™”.
Because of the continuous changes made to Linux, the innovation rate of Linux was (and is) *much* higher than any other OS (like OSX or Windows). It’s like a mini Darwinian cesspool. Because anything goes, GUIs for a lot of environments are available and still improved.
Just picture if in 2003 a copy of the Windows GUI was adopted. Linux would have stagnated and Android wouldn’t have happened. Diversity is good!
Edited 2014-12-28 20:42 UTC
Agree, I don’t think it’s a UI problem. The UI problem is just a small part.
If there is one thing the Linux distributions could do is give all ‘newbies’ the same desktop environment by default.
Not that it is likely to happen, but that would be a way to go about it.
This would obviously means: other desktop environments can be installed/should be supported.
Edited 2014-12-28 21:55 UTC
As someone who doesn’t use desktop Linux primarily for this reason, I can assure you that it is mostly the problem. Well, not the UI per se, but more along the lines of the desktop environments, APIs, and UI toolkits that drive the whole thing. This stuff REALLY needs to be standardized. Of course, it should be customizable enough so users can make it look and feel however they want, but the foundation really needs to be the same from distro to distro. Otherwise, desktop Linux will have the same paltry marketshare in 20 years as it does now.
Another BIG thing Android has that Linux really needs is one app store/repository to rule them all. When I’m using Android (and not a bastardized Android hybrid like the Kindle Fire), I know that I’ve got access to the Play Store, which includes the latest version of ALL the apps on the platform. (Except for the ones that have been pulled, of course.) It’s nice to have that same experience no matter what Android phone or tablet I’m using, and not have to go begging to the distro gods when an app I want to install is not present.
When one of my absolutely non-techie friends had a problem with her Windows installation (“problem” as in became completely broken, and she had no way to restore it), I posted her a few Ubuntu Live CDs to try out, plain Ubuntu, Kubuntu, Xubuntu, etc. Now she’s really not the sort to want to customise or play about with computery things (including devices like her smartphone), and she prefers that stuff just works with minimum hassle, in fact she gets annoyed incredibly easily by overcomplications… but she really appreciated being able to try out a few different approaches to the desktop before settling on which version of Ubuntu to install. I believe she went for XFCE for a while, then moved onto either the MATE or Cinnamon version of Linux Mint when she decided after a few days that she didn’t like XFCE as much as she thought.
That is the crux of the matter. But I think the explanation goes deeper… It isn’t just “what comes pre-installed?”, its “whose name is on it?” It’s brand recognition, product positioning, and marketing.
Lack of corporate support isn’t the problem, it is a lack of corporate ownership… Google is by far the dominant force in the mobile space as far as Linux goes. They essentially own it – to the point that the vast majority of people don’t even know and definitely don’t care that Android runs on top of Linux. Android is a Google product, Google is a strong brand that the general public perceives as smart and progressive, the product thus far has been extremely competitive, well supported, and generally liked. That is why it is successful.
Google may in reality owe nearly all of Android’s success to Linux… But in the marketplace, as a product, Linux is not even a worth mentioning as a line-item feature.
Google used Linux as a shortcut to dominance, but by doing so they didn’t enhance the brand of Linux one bit… They developed the right product with the right business model and are fiercely competitive – that is the key to their success. They could have used a BSD, or even written their own kernel, it wouldn’t have made much difference in the marketplace.
Why did they go about things this way? Here is the thing that Google realizes but no one wants to admit: you can’t market Linux to consumers.
Yeah, I said it.
People will never buy Linux. The one and only thing most people know about it is that it is free and no one owns it. How does this reconcile with the notion of a product with a price tag? Sure, in the corporate space you can make it work through support contracts, but that simply doesn’t work with consumers. The want a company to worship when things are great, and someone to blame when things go wrong…
I’m not saying Linux can’t achieve respectable market share in the consumer space, I’m saying for it to do so as a product that is sold to users it has to essentially be wrapped in a corporate blanket. There has to be enough “secret sauce” on top of it to justify a price tag – which means that Gnome, KDE, etc. are non-starters. The more “Linux” shows through the cracks the worse it will look as a product to consumers.
The key to success for Linux on the desktop lies in its exploitation by a company with the brand capitol and financial resources to compete with Microsoft and Apple. In short Linux on the desktop needs a Microsoft, and (thankfully, imo) it doesn’t have one.
Personally, I think Linux is wildly successful as it is now. Keeping it free and open is vastly more important that making it pervasive… I don’t really care if it becomes “dominant” on the desktop, in fact the thought kind of scares me. For that to happen it will end up having to be exploited by yet another company that will go out of their way to minimize it’s relevance to their product.
To be clear, I’m not really opposed to what Google did with Linux as far as Android goes. I look at it more as they did what they had to do to succeed. And there is no argument that all the activity, the coding, the driver support, the platform patches, etc. that resulted from that made Linux stronger technically. But will it be a “win” for Linux in the long run? I’m not so sure…
Great great great, wonderful post, and I agree to 99%
Im certain that Android will be good for linux in the long run though. What can Google do to linux to kill it or make it less strong?
And how would Google win by make linux suffer?
Google bought VP8 and open sourced it, they bought Etherpad, then open sourced it, it part of their business model. They love open source, and open source in general benefits from google.
Im happy for every new chromebook bought. Not because I love chromebooks, nor Google, but because it gives Microsoft a harder time exploiting their huge market share with vendor lock in etc
Microsoft tax Cromebooks, and do not even have the confidence to state what patents are being infringed, and fight their corner in court. Buying a Chromebook just feeds Microsoft and Google.
A Linux laptop is a preferable proposition.
What is underneath the kernel within Android is mostly BSD code.
LOL, whaaaat?
Well, he is halfway-right.
AOSP is licensed under BSD license, (except external projects like kernel or chromium). However, BSD licensed code doesn’t mean that it is BSD code. It was written by either Google or Android Inc.
Subtleties like this are lost on many people, so they share with others their warped understanding of issues.
So the previous poster is not half right, he’s fully wrong then.
BTW, last I checked the license for AOSP is Apache.
No, I’m 100% right. But this is OSNews so I don’t expect anything better from posters.
Your wording was confusing, unless you meant exactly what you said, in which case you are just wrong. Either way…
It’s confusing if you’re an idiot. I meant what I said and I’m 100% right.
No, what I meant was the code was taken from OpenBSD / FreeBSD. Nice try trying to interpret what I said and getting it wrong.
I assume he is talking about the fact that Android’s version of libc (Bionic) is derived from the BSD libc, as are many of the commands in toolbox (NetBSD, if memory serves).
Of course, it is still quite an exaggeration to say that Android is “mostly BSD code” under the kernel – but it does have some BSD-derived bits in it.
No, most people use what comes with their phones. Very few of them change launchers. Very few of them understand what a launcher is. Just about every android phone follows the same UI paradigm.
Yes, that makes sense. I’m not sure how that ties in with your first two points at all, but that seems to support the 1FUI argument. OEM support would be be key, but many would argue ( including myself) that 1FUI is a requirement for greater OEM support.
I agree, and the reason why desktop Linux isn’t supported by large companies are the lack of unified naming and versioning of libraries, and a common way to package and distribute software. Android have all that, and this is one major reason for its success.
To me Android is a Google product, increasinly being used to Google strangle the net (along with Apple the net is being redirected to suit a couple of vested interests). I see Android as holding back better Linux products as Sailfish, Tizen, Ubuntu phone, etc.
As far as desktop Linux is concerned the fact you can do what you want with it is a Linux plus point; imagine if Unity was the standard UI!
Currently using KDE 4.10.5 on FreeBSD.
Steve, I think I speak for many of us here at osnews.com when I say that you appear to be a complete imbecile.
just because it has a Linux-based kernel it doesn’t mean it’s Linux as intended by that 2003 article. It’s a mobile OS that serves specific purposes (Google ad revenues first, and vendor/carrier commodity, giving them maximum freedom to load it with custom user interfaces and mixed crapware), the Linux kernel is there just because it was the easiest to implement and to customize.
Exactly. Linux has never been and never will be an OS. It’s a kernel and that is all. Android is just another OS powered by the Linux kernel.
Yet there is no single mention of Linux in the Android’s marketing materials
You have to use Search option and you will wind Linux in various places of developer documentation – which most phone users don’t care about.
This guy is absolutely right. The previous comments are from Linux boosters who are out of touch with reality. Android dominates because it is standardized. Software suppliers and users have something they can depend on. Legacy Linux is chaos (for layers above the kernel) and will remain a niche market. The substantial talents of Red Hat, Canonical, etc. are being wasted. So sad.
Android is so successful because it is backed by Google and a number of big hardware manufacturers – no other reason.
If Google and the major manufacturers backed a desktop distro (and stopped selling Windows) they would achieve a similar result to Android.
Linux is dominating in different markets, like servers and smartphones, so I will not talk about those. I’m interested in GNU/Linux for the Desktop.
I’ve been using GNU/Linux on the desktop almost every day for the past 15 years. I’m a highly technical user who prefers the command line, but with a strong opinion on usability and its merits. For a long time the GNU/Linux desktop was very immature and, frankly, a mess. Then freedesktop.org came along and defined standards that should be followed by desktop environments. This changed everything. The big names, KDE and GNOME, became fully compatible with each other and choosing one or the other was a breeze. Smaller desktop environments, window managers and desktop applications followed them and the future seemed bright. I was very proud when a computer illiterate cousin turned to me once and said: “Linux is weird… It’s so easy to use!”.
Unfortunately, a couple of years ago, the big desktop names decided unilaterally that the very successful desktop model was dead and tried to came up with something different, with some serious breaking changes. This halted the evolutionary process that was used before, triggered a desktop retrocession and turned us into some sort of desktop environment dark ages. That’s the way I describe it and I think it’s a very accurate description.
I sincerely hope we get back on track and that the big desktop environments start doing some real usability tests again, instead of coming up with highly debatable “desktop is dead” bullshit. Some of us still appreciate using a Free desktop and we want to be back on top again.
Agreed; after KDE 3.5 and GNOME 2 the Linux desktops slid into a dark age. After being a happy KDE 3.5 user, KDE 4 pushed me over to FVWM, Xmonad, a bit of Awesome, and finally i3. Eventually I got tired to having to read strftime(3) to change my clock when I moved time zones, and now I’m on OS X.
I’m not even sure the desktop dark age is over yet. Unity is still a mess in many regards, and my most recent experience with KDE on openSUSE (supposedly a good KDE distro) was a mess of desktop search indexing and heaps of confusing panels of options.
Interestingly, many things changed since 2003. Even then UIs for Windows were nowhere close to being consistent. Eg. WinAMP (which apparently was the audio player back then) did never behave like anything else. Applications with system tray icons worked differently (eg. some minimized to tray on clicking “minimize” button, others – on clicking “close” button, and some even required choosing menu item.) And I recall having issues with columnated views – some of such widgets allowed sorting on multiple criteria while others didn’t; some allowed selecting columns while others didn’t; automatic resize would respect sorting arrows for some of them but not for others. Later Chrome introduced Vista-like window border on XP, Microsoft pushed Ribbon UI, and now we have Metro apps alongside desktop apps.
At the same time, the problems outlined in the 1FUI “essey” are largerly solved: on distros that care about consistency it is not so easy to tell GTK apps from Qt apps, and no interoperability issues like clipboard problem remain. That is: now Linux desktop is more consistent then Windows desktop with third-party apps, but the usage shares remain close to those back in 2003.
P.S.: Android does not have a single UI. In fact, it is much easier to get a GNOME-only desktop then consistent application set of Android software. My ebook reader’s UI is nowhere close to my music player, which does not resamble my video player in any aspect, and my browser behaves yet differently from all three. And I did really try to build consistent “experience” on my phone…
P.P.S.: And don’t forget vendors’ launchers and other “enhensements”. My parents have Android phones from different vendors, and while they are quite fluent with own phones, they have problems with each others’.
Edited 2014-12-29 01:08 UTC
Just LIKE Android…who cares if the UI’s are a little different. As long as all your apps work on all the devices.
That is probably the single biggest issue in the Linux world to me the fact that every flavor has its own package format, its own package manager, and its own package depot.
The real place the Linux world went wrong was in not having a consistent software package system. Its either Debians fault for not staying with the Linux Standard Base and using yum, or its redhats fault for not willing to try something different.
Regardless, that article missed the real problem(s) back then, just like it does today.
Edited 2014-12-29 01:08 UTC
You’re right, but it’s not just the UI and not just the package management system. It’s everything.
It’s KDE vs. Gnome vs. others; and apt-get vs. portage vs. Pacman vs. dpkg vs. others; and systemd vs. init; and X vs. wayland; and bash vs. csh vs. zsh; and slightly different alternatives for CRON or system logger; and many different kernel compile-time options; and plenty more differences.
All of this combined mean that (e.g.) if you have any problem of any kind and phone your local computer shop, they hang up on you as soon as they hear the word “Linux” because nobody is stupid enough to try to support 123 billion different permutations of Linux.
It also means that (e.g.) if you want to buy a modern game (like Crysis or Skyrim or call of duty or…) you’re screwed because no major game developer is stupid enough to try to support 123 billion different permutations of Linux.
It also means that (e.g.) if you want to buy very new hardware (e.g. just released yesterday) you’re screwed because the device itself won’t come with a Linux driver (so you’ll have to wait for ~6 months for a buggy open source driver to make it into the kernel, and then another ~3 months before they fix the bugs, and by that time your “new” hardware is old); because no hardware manufacturer is stupid enough to try to support 123 billion different permutations of Linux.
These things aren’t so much of a problem for “advanced users” and professional server administrators; but it completely destroys any hope for your average user.
It will be interesting to see how SteamOS goes. Either Steam will force “de-facto standards” onto Linux (where other distributions adopt Steam’s choices to be able to take advantage of Steam’s games); or Steam will lose a lot of $$$ until they realise Linux is an unsupportable platform and give up.
– Brendan
Edited 2014-12-29 03:13 UTC
Valve solves this problem with Steam. It’s called the Steam Runtime and is the stable base in which game developers use to port their games to Linux. Valve saves themselves trouble by only officially supporting Steam on Ubuntu and SteamOS but that hasn’t stopped other distributions from adding support for Steam but has allowed AAA games, such as Metro Redux and the upcoming Bioshock Infinite land on Linux.
That said, I don’t really see much of a compatability issue. I mean, I can still run ancient binary-only Linux games from the late 90’s and early 2000’s on bleeding-edge Arch. Even the old Loki games still work on modern distros with the updated game installers. I tested the ancient Mind Rover demo the other day, along with Doom 3 and Quake 4.
Edited 2014-12-29 04:05 UTC
That is only a problem for commercial developers looking to support Linux and usually their solution ends up being to just officially support 1 or 2 distributions with a DEB or RPM package or Installer script. They leave it up to other distros to make their own packages and get support from the community.
I like how Arch’s AUR system does it: downloads the RPM or DEB package from the company website, extracts it with tools, such as bsdtar, then generates a native package on the fly and installs it.
Linux did not fail
It is simply banned from government desktops, despite being more linux friendly, more accessible, ezoom etc, and more secure
Supposedly linux was banned as a security measure
Forcing an operating system on people by force is the problem, and it doesn’t promote security
Android and GNU/Linux are pretty different beasts despite using the same underlining kernel.
That said,the two are becoming more compatible. For example, it’s now possible to run some (not all) x86 Android apps on ChromeOS and regular GNU/Linux using the ARChon runtime (Archlinux even has a script to run Android APKs by double clicking on them) and even more apps with Genymotion plus the seperate ARM translation layer.
Linux has two major failings that I see.
1. it is defaulting to a tablet UI in Gnome. Yes Unity/Gnome3 can be turned into Classic mode, but honestly there should be a panel asking you if you want to run a tablet mode or a desktop mode the first time the OS starts. Apple never fell into this trap.
Microsoft is trying to crawl out of it with Windows 10.
Linux is still stumbling through and fumbling the ball here.
2. Apps Apps Apps. Desktop usage is all about apps and productivity. More apps = more users.
To elaborate on the above second point:
Can’t model in 3D without learning Blender or Maya?:
FAIL, (I’m trying to fix this by cloning two 3d
modeller UIs (basic functions mainly) in a single application, one of which is 3D Studio Max). 3D artists don’t care about your UI/fantastic tool. They learnt their craft on their tools, and they will stop using them when you pry them from their cold dead hands.
Can’t make a level for the game you like? Don’t expect much game development to happen on Linux. See FRED2 for Freespace 2, and UT4/GTKRadiant/OpenCS for UT/Quake/Morrowind for where this is falling over, and also being done right.
I am currently working on this for VegaStrike to try and fix the ten year old problems with making missions/campaigns. It is a big job for someone who has learnt programming purely to fix a bug that actual programmers could have fixed.
Freespace 2 has 0 mods being made on Linux despite being a completely open source game, because the level/ship tools don’t exist on Linux. (there are plans to fix this but they look mostly aborted so far)
OpenMW/OpenCS is rapidly catching up to Windows for tools, and I expect we will see an Elder Scrolls renaissance on Linux once it’s released in full.
KDE/Gnome is irrelevant, everything works on everything unless you are running a strange outlier such as GNUstep and even there things are looking pretty good for standards based interop.
Things Linux needs: a 3D model viewer that reloads all textures on the model every time you switch focus to the model window. Better 3D tools, and probably a hard point editor for FPS/Space games. Better GUI tools for developing code. Why can’t I select 10 buttons in glade and change all their properties at once? Visual Studio does it and it’s a useful/obvious feature. I’m sure there’s more but these are specific to what I’m trying to work on/noticed. Until these issues are overcome, mainly with respect to Productivity tools, noone is going to adopt Linux.
Windows is crap but it has every tool you ever needed, and some you don’t think you need, but will need waiting for you.
I will also note that I am from a game modding/3D art hobbyist background. I have taught myself programming so I can work on passion projects, mainly space game related. I am trying to fix the tools issues on Linux.
Edited 2014-12-29 03:29 UTC
“They learnt their craft on their tools, and they will stop using them when you pry them from their cold dead hands.”
This is an oxymoron, if one learns their craft, one is tool-independent. If one isn’t tool-independent, the one isn’t really a craftsman, just a tool user.
In a programming example: one learns to write code or one, as many do, learns to just use an IDE. They are not the same thing.
“Better GUI tools for developing code.”
what.
There’s another oxymoron here, and this one goes with a flat what. If one is writing code, one doesn’t even need GUI tools, most of the best programmers in the world work just in the command line.
Windows is not for programmers, and when not dealing with C# and its siblings for obvious reasons, most of the tutorials one finds automatically assume one is in an Unix environment. At least from my experience and my education in college.
Really? That’s quite odd given that there are plenty of Windows programs knocking about.
I wonder who writes them if not programmers?
As much as i like Qt and KDE, i think Gnome might succeed there with its desktop, while Qt seems more likely to succeed as a toolkit. Gnome3 feels quite a bit like using a mac – most clutter is well hidden, and what is not will satisfy a lot of users.
In the long run it seems that KDE Frameworks approach might pay off instead, even if Gnome offers more polished and user friendly out-of-the-box experience.
If KF5 matches current gnome 3 in its unified and fairly intuitive design, we might have a clear winner there, especially with portability of Qt.
Desktop Linux was doomed to fail and I don’t think it was just the UI. Being a collaborative effort all the parties involved in developing it used it to scratch their respective itches – independent developers used it to do some fun programming after hours or submit a patch to pad their resume and companies used it to offer a product in their particular niche and didn’t care much for the whole picture.
So the result was a product that was all over the place, with an uncertain direction, unstable APIs, countless repositories, bugs, lack of polish, inconsistent across core applications, with little regard for usability. It couldn’t have been any another way.
This is why a lot of companies didn’t rush to port their applications and unfortunately using only open source software can only get you so far in some fields.
This doesn’t mean that Linux is crap or without merits – it probably works fine for a lot of people actively involved in it. It’s just not very good for the general public (yes, I know your grandma is using Linux and it’s great).
So, here comes Google.
They took the core of Linux and then redid everything that is user facing. I would barely call that Linux. Google could have used any other OS as a foundation and may still do so in the future. In fact look at BBOS which runs Android applications and uses QNX as a foundation.
Google just used Linux like everyone else – they had an itch to scratch. This doesn’t help Linux as a desktop OS very much.
Really, I wouldn’t even call Android still Linux.
1FUI won’t magically solve several dozens crucial problems with Linux/GNU, and probably the most serious one is the fact that there’s no Linux/GNU platform.
Windows is a platform with a stable kernel and stable API/ABI.
MacOS X is a platform with a stable kernel and stable API/ABI.
Android is a platform with a stable kernel and stable API/ABI.
Users and developers love stable things. Linux/GNU, however, is a bunch of loosely tied components which have no stable API/ABI whatsoever. More on this here: http://linuxfonts.narod.ru/why.linux.is.not.ready.for.the.desktop.c…
Throw in one more – most major os’es (maybe except android) have some kind of OLE-like mechanism, so you can drag and drop or copy-paste various things between different programs and have it well handled, as well as system-wide clipboard management.
Seems trivial, but for typical desktop user workflow it’s often critical.
Linux is successfull where it either runs headless (servers and supercomputers), or some company has poured lots of monies in R&D to replace the broken X.org and PulseAudio mess.
Of course, this realization would force open source fans to admit that when it comes to building audio stacks and windowing systems, the community open source developmemt method is broken with no hope of being fixed.
There is a reason the Win7 release candidate surpassed the marketshare of Desktop Linux marketshare despite being preinstalled by no OEM or not marketed in any way. It’s back compat and quality of the audio and windowing stack.
Edited 2014-12-29 10:12 UTC
That article is still wrong today, as it was wrong in 2003. Android is dominating its marked since it provides the apps most people want to run. Linux does not manage that. Yes, a lot of people today spend most of their computing time in a browser window, and this increased the Linux acceptance, but is not enough for world dominance. Is all about the apps, the user interface is not important for “normal” users.
You couldn’t be more wrong about the UI not being important (`for “normal” users` or otherwise). People raised all kinds of hell when Microsoft went screwing around with the UI in Windows 8. As a matter of fact, the UI was so important to users that Microsoft had to backtrack and reinstate their traditional UI into the next iteration of Windows (Windows 10). They’ve even used this fact as a selling point for Windows 10 saying they’ve listened to the users and are giving them what they want.
People raising hell for Windows 8 aren’t “normal” people, they are geeks or “power users”. “Normal” people went “so that’s the new Windows? oh well, if that’s how my computer works I’ll use it”. Normal people don’t go on internet forums venting their frustration with software, they at most may complain to a coworker.
Nope, it was all the negative feedback from normal users that made Microsoft do a 180 on the UI. Normal people don’t typically go on internet forums but they blew twitter, Facebook, and Windows Support up about it. `Geeks and power users`, who are vastly the minority, took one look at Windows 8, said no thanks, and never installed it to begin with.
The GNU/Linux is using Linux Kernel to in essence work as a Unix server. That is why it is really good for a server, and why it never really got on for the Desktop.
Android is using the Linux Kernel, and the rest was designed to be a mobile platform.
The Linux as a Desktop never got the steam it though it would because most were Distributions of GNU/Linux modified to copy a lot of desktop functionality.
What needed to be done is what they did with OS/X used the existing kernel then made the OS with the idea of a desktop in mind.
As mentioned above. I am just a stupid user of linux. I can hardly use CLI. Is always have to copy paste. I know how to install and know how to update. And I know which program is needed for my needs.
I use Crunchbang for old computers and it works.
Linuxmint for the rest. It works.
The GUI is what the computer can handle and what I like. It is like a candy store. Yummy. Choices.
And I am really not interested if somebody else uses what I use. I use it, because it works.
So, this whole “linux for the desktop crap” or “why linux will never be a good desktop thing” is flashing by with a giant Whoosh every time. I don’t care. I call it freedom. And freedom is not for sale. Thats a terrible thing for companies. They can’t sell freedom. But they can sell Chromebooks, Android or whatever. And it helps development of GNU/Linux.
If one interface was settled on in 2003 we might not have Android. We also might not have the features we now have in KDE and GNOME. Competition is a good thing and it’s even better when you have an open system that allows you to put any interface you want on it.
As a side note the ONE INTERFACE TO RULE THEM ALL mentality for Linux was always stupid. You can’t force people to NOT work on the project they want to work on. Because Linux is free and open you will always have alternative applications and interfaces.
The only way you could have ‘1fui’ in the FOSS world is if something came along that was so good that practically all developers adopted it.
That hasn’t happened, and I doubt it will happen, if it does I think it would have to be something new coming along, because I can’t see either GTK or Qt emerging as a winner after all this time.
Also given how users have been able to cope with transitions to mobile UI’s, and also changes to UI’s in Windows and OSX etc, I find the whole notion of users not accepting multiple user interfaces to be nonsense.
Personally I don’t see the problem of running both GTK and Qt based apps on the same system, it’s not as if the interfaces are particularly different useability-wise, and you can use a suitable theme which looks the same across both toolkits if that bothers you.
Its so very close to the actual problem, android, apple, and Microsoft have SDKs that make apps work without any extra work on anyone’s part. Linux, good luck in 6 months after the next release. Without an SDK that makes it so apps just work for over a decade after they were developed, desktop linux simply can’t compete. Sure there’s android, but but it’s not really friendly for productivity and Google seems to frown on OEMs using it on non tablet/phone PCs. To the end user, the sdks just work an linux desktop doesn’t.
“Linux is not an operating system, it’s just a kernel! And yes, Android uses the Linux kernel, but somehow it is magically still not Linux because it doesn’t have any of the traditional crap thrown atop Linux that I just said are not part of Linux but that I will arbitrarily make a part of Linux again just to be able to say Android is not Linux because Android does not conform to my ’90s idea of what Linux should contain, thereby contributing to keeping the desktop side of Linux the same crappy mess it has always been!”
Loony as hell.
I can’t quite tell if you’re playing a bit of devil’s advocate and being provocative or if you believe that.
Sure, Linux by definition perhaps is just the kernel, but the defacto usage is that it’s at least the kernel and common userland tools and libraries (even if we forgot about the different desktops or window managers or whatever for a minute).
And you know this. Yes, we could have an “Android Linux” desktop implementation.. but most of the myriad extant desktop Linux apps wouldn’t run on this as-is. And perhaps that’s your ideal path, maybe google could have a new AOSP project AOSP Desktop to refine that mode and court ‘Linux’ Desktop app developers etc. But maybe that’s not wanted by the free brigade, but all those wanting more choice than this.
I think I agree with the earlier poster espousing the lack of any kind of real Linux Platform (ABI/API/and someone else mentioned OLE and better system wide application integration)
Maybe forget Android and Android wish fulfillment for now. And particularly forget the corporate overlords. With that in mind, surely the major players in Desktop Linux – KDE Gnome, and the lightweight players XFCE LXDE etc deserve a bit more of a chance, as the truly open source (non-corporate ) contenders to get their act together – perhaps once and for all.
I know it unlikely but with stable API/ABI’s etc in mind and creating that elusive desktop linux platform, why couldn’t these project combine forces with each other and create a Unified Desktop Linux Core (or whatever they wish to call it).. which wouldn’t necessarily have to be the end for the subprojects but would require a standardised set of libraries, toolkits and so on to be installed on the ‘full’ install (and choice of running kde/gnome as default desktop)–but really have everything underneath smoothed out much better – and really work well together–by which i mean side by side.
Shit if they really got their act together they could rest a certain amount of control back from the systemd lot – perhaps admit the benefits of that project, say yes, we’ll work with you and fully cooperate but only up to THIS point (have some mutually agreed core services included) but leave a few more open to 3rd parties, but also demand project oversight and direction is divorced from RH and other corporates.
Anyway, as far as Free Software goes, more than just OpenSource.. the Linux Desktop project is much more important than just the linux kernel. (that’s just a hugely important enabler, it’s not the whole). I’m not sure Android can satisfactorily supply the whole.
Maybe this nonsense would make sense if Linus created the userland when he created Linux in the first place and there was a Linux OS but there wasn’t. An OS isn’t a kernel on its own. I find it odd how this concept was clear for every other OS in the history of computing until the Linux kernel came along. It’s such a simple concept. Maybe that’s the issue, it is too simple.
The only thing that is defacto is there are a variety of OS’s using the Linux kernel. That’s it.
..I was kind of agreeing with you 0brad0
Linux being the title of the kernel, GNU/Linux (simplistically) being the OS that i’m suggesting is referred to commonly as simply linux.
Maybe I’m not being pedantic enough for you. but probably more than enough for most.
maybe kde 5.
i already found kde 4 the best desktop manager, can be easily modified if you don’t like it and there are a lot of great application…
i think anyway the future is the web
Linux is now a permanent part of the technology “ecosystem” (or at least until kernels are no longer needed). It has succeeded probably beyond Torvald’s wildest expectations(it’s on the NASA space-station for God’s sake!). (Or Microsoft’s worst nightmares.) It’s used to make Hollywood movies, run supercomputers, astrophysical research, etc… Whole European cities have moved their government systems to Linux desktops.
Are we really concerned that Windows users don’t get Linux’s advantages? I really *like* the dynamism of the “Linux Desktop”. It’s a niche desktop, but it’s as large a niche desktop as Apple (OSX) according to some estimates. I now walk into stores and see “Compatible with Linux” right on the labeling.
If any desktop has the potential to eventually capture marketshare from Apple or MS, it’s Canonical’s Ubuntu. I may prefer KDE, but the hope for the future of the “Linux Desktop” is Unity. It’s too early to count Shuttleworth out. He’s still playing for keeps.
Let’s get Unity 8 out the door (f-i-n-a-l-l-y), along with click to install, and then we’ll talk.
Why would we conclude that he was right when Android still would only represent one “UI”, one that has many variations itself, but all these variations are still just a subset of what Linux is. Linux dominates mobile and leads the way in many server applications precisely because it doesn’t have jus 1FUI. If anything, the premise of the article was totally wrong.
For Linux to have a prayer of making it in the business world, the following requirements must be met:
1. It must come pre-installed.
2. There should be only one version of a ‘business Linux’ that developers can develop for.
3. There should be only one version of a ‘business Linux’ interface.
4. You have to be able to easily authenticate against a Windows domain.
5. There have to be available applications that are 1000% compatible with MS Office and that can handle all MS Office documents flawlessly.
6. There has to be an application that is the duplicate of MS Outlook.
Understand that I’ve been using Linux since 2000 and am even able to use it at work. In most respects, compared to Linux, Windows is a crude little toy that will always be insecure. However, like it or not, it’s a Windows world and until the above requirements (and probably more) are met, Linux will be a non-starter in the business world.
I disagree with the one UI concept, I feel its too limited but there should be some Organization and a little renaming. Mainly there should be a BIG player for each environment, not multiple environments per distro.
Launcers ( for want of a real name ) – Android
Unity – Ubuntu UnityLinux
Cinnamon – Mate CinnamonLinux
Gnome 3 – Fedora GnomeLinux
ChromeUI – ChromeOS ( Even though its basically a browser )
I am going against my preference here, I use Ubuntu Gnome, I prefer a Debian / Ubuntu base, but I think fedora is the best gnome experience.
Obviously, other players should step up for the other enviroments. Everything else which interchanges UI’s should be classed as NicheLinux
1. Android has multiple UIs which are used by variety of people.
2. Having multiple UIs introduces competition to desktop environments forcing them to improve upon themselves at faster rate. They’re in a kind of tug-of-war for the users of the operating sysytem and that keeps the projects not only healthy but also constantly developing. We can argue that neither Gnome or KDE would be the mega projects they are at the moment if not for the competition between them.
4. This competition can also produce surprising results like Gnome 3. When community behind the UI feels that their current approach to take over the world isn’t working. They’re free to start from scratch and operating system community doesn’t have the fear of losing users. Look at the Metro UI in Windows 8. That abomination discouraged people from moving to Windows 8. People dislike Gnome 3 but that didn’t affect the GNU/Linux users much because there is choice.
5. Freedom to innovate. If existing solutions don’t satisfy and the big projects won’t accept your designs then you’re free to start your own, possibly fork an existing project, and do your thing from there. It might grow up to be a large project which takes users from existing old solutions.
6. Other stuff.
It didn’t affect Windows users much either because they had choice as well – to stick with Windows 7, which many people love.
I recently tried out the FEDORA 21 – Workstation ISO. I like it. Wayland – is working very well – even on somewhat low end hardware. I like FEDORA’s approach to offering three “versions” – I have always been in favor of Gnome Classic, MATE or some simplified desktop. I like to tinker too… But, most of us have real work to perform and don’t have the skills to TINKER endlessly to keep tweeking a desktop every day. Thats why Microsoft has gained a foothold and continues its dominance. It looks OK, its tried and tested, and its the same box after box after box.
I recently tried out the FEDORA 21 – Workstation. I like it. Wayland – is working very well – even on somewhat low end hardware. I like FEDORA’s approach to offering three “versions” – I have always been in favor of Gnome Classic, MATE or some simplified desktop. I like to tinker too… But, most of us have real work to perform and don’t have the skills to TINKER endlessly to keep tweeking a desktop every day. Thats why Microsoft has gained a foothold and continues its dominance. It looks OK, its tried and tested, and its the same box after box after box.