Slashdot Mirror
Why Linux Is Not Yet Ready For the Desktop
An anonymous reader writes "Every now and then a new- or old-media journalist tries to explain to everyone why Linux is not yet ready for the desktop. However all those men who graduated from their engineering universities years ago have only superficial knowledge about operating systems and their inner works. An unknown author from Russia has decided to draw up a list of technical reasons and limitations hampering Linux domination on the desktop." Some of the gripes listed here really resonate with me, having just moved to an early version of Ubuntu 9.10 on my main testing-stuff laptop; it's frustrating especially that while many seemingly more esoteric things work perfectly, sound now works only in part, and even that partial success took some fiddling.
Without the big labels like Valve developing their titles on Linux, you aren't going to see Linux widely used in desktop soon.
New Economic Perspectives
The first alpha of 9.10 was released a couple days ago with new kernel, new gcc, lots of new libraries... you should not be surprised things don't work well yet. Jaunty seems pretty stable to me. Minor issues with my intel video card, but works fine for all my daily work.
There is also the fact that web-based is the new way of making money from software. No piracy since its mostly server-side, lace it with ads and nobody complains about adware.
But people do complain about not being able to access web applications from notebook computers while away from Internet access, such as on the road or in a restaurant that does not offer free Wi-Fi.
If you're not using it now, you probably never will. As a long time (and current) Linux user, I have come across all these issues first-hand, as has every other Linux user, developer and advocate out there. That they are still problems even though they've been known for years - sometimes decades shows that they will never be addressed, or fixed.
Linux is a hobby systyem. The code is donated mostly by amateurs (or people working for rewards other than money - for example the recognition of their peers) and is therefore not within the normal disciplines of IT developemt. If you tell a Linux developer their code is crap - or the application they have written is junk, they'll just walk. As they will if you ask them to do things they don't want to: such as write a manual, fix bugs, add (or remove) features.
Basically guys, this is as good as it gets. Live with it or go elsewhere.
politicians are like babies' nappies: they should both be changed regularly and for the same reasons
The TFA is a worthless troll, even more so than usual in these "Linux is not ready for the desktop" Slashdot articles.
It has the usual list of ignorant complaints (oh no, there is a choice of distributions, boo hoo! oh no, there is a choice of GUI toolkits, boo hoo!), but some points stand out in their sheer stupidity.
"Bad security model: there's zero protection against keyboard keyloggers and against running malicious software (Linux is viruses free only due to its extremely low popularity). sudo is very easy to circumvent (social engineering). sudo still requires CLI (see clause 4.)"
Really?
Who admits these articles to the front page anyway?
we discovered a new way to think.
Lucky? I hardly think so. WoW is hardly an obscure game - it is the most popular MMORPG in the world. The idea that wine can run, out of the box, such a high profile game is perfectly incompatible with 'no games, period.'
If we can put a man on the moon, why can't we shoot people for Apollo-related non-sequiturs?
...but insists that reproduction of any kind is prohibited without permission. So I won't quote from the article. I will just refer to it.
You might want to refer him to the concept of "fair use".
Seems like we've had this exact argument a thousand times. This list at least makes mostly good points. But it still misses the mark many times. Particularly annoying is the absolutism in so many statements, like:
This is obviously false. There are games on Linux. Many are open sourced, and some commercials games are available on Linux (e.g. World of Goo). Now I wouldn't have argued if he had said "Very few games." But instead he tried to make his point punchier by being absolute... and this weakens his whole argument by introducing lies.
And as usual the author prefaces by mentioning that this is some sort of relative comparison with Windows, yet points out problems that exist with all operating systems, like "A galore of software bugs across all applications", or "huge shutdown time" (I've timed it on dual-boot systems and for me Kubuntu was faster than Windows XP. YMMV.) and "poor documentation" (does Windows come with an awesome manual I wasn't made aware of? No. For both Win and Linux you end up searching online. Both have tons of 3rd-party documentation.)...
And then there are kind nonsensical complaints like "don't allow you to easily set up a server with e.g. such a configuration: Samba, SMTP/POP3, Apache HTTP Auth and FTP where all users are virtual" Does Windows let you do this easily? The heading said that this was an analysis of whether Linux is ready for the Desktop and instead the author injects one of his pet-peeves about configuring Linux as a server?
And then there are spurious assumptions used to justify complaints, like "Linux is viruses free only due to its extremely low popularity". We've had this argument many times... undoubtedly the low market-share of Linux helps keep viruses off the platform. But there is also plenty of evidence that it is robust security-wise (e.g. infection rates for servers). At a minimum it's not the settled question the author implies.
I could go on and on. No doubt this thread will tear-apart other statements from TFA. It's too bad, because many of the points made are very much correct, and deserve attention. But it seems that whenever someone tries to compile lists such as this, they end up not only making good points about what needs work, but throwing in their own anecdotal annoyances and personal viewpoints, which muddies the whole argument...
The desktop/laptop is NOT dead.
The reason is simple: people don't trust computing "over the cloud," because your device will be essentially useless if you are in an area with little to no Internet connectivity. Besides, you can get a netbook computer for under US$400 nowadays, and with improving technology those netbooks will soon store as much as 250 to 320 GB of data on the hard drive in the machine itself, way more than enough to store local data for business documents, spreadsheets, and smaller presentation files.
The parent poster is not talking about corporate use, or geeks like us, he's talking about the folks at home. You know the other NINETY percent of the market.
Tired of all the isms, don't exploit people as an employer, or a government, mmmmK?
Standard Asound driver.
You miss the point: I did get it to work. That means it could always work, from the beginning, but Ubuntu did not include the correct hardware recognition to set up the system. It also provides no easy method for me to report back those settings for others. If it was truly a manufacturer problem then I would still not be having sound.
That means the moment I got it to work I stopped fiddling- and every time I get the little red upgrade spot I hesitate and think: Is this the reboot that kills it?
Most of the software I work(ed) with is still to get heavy duty tasks done are still very much on my computer. CAD, programming, mathematica-type programs... not that I want them all to be, just how it is with current internet pervasiveness and speed.
I like using Google Apps because I don't have to worry about keeping files updated across multiple computers. I think Google is safer than carrying a tangle of USB sticks about. If the file is that important or secret, I stick it onto a computer that has absolutely no net access, no modem, and no ethernet connected to it, no wireless, etc.
There are more reasons to like net apps than just being clueless. Besides the aforementioned syncing problem with files, services like mint.com provide, say, an iPhone user a convenient look at their finances impossible with a regular desktop/notebook unless you're really regimented.
That's a decent insight. However, I have no problem with people making money on software that way, as long as software patents don't block competition. What's more problematic with me is being at the whim of the software service provide at any moment to hold your data hostage and your account in their hands. I had enough experiences with ebay's arbitrariness to make me wary. That's why I do keep a backup of the google documents (and important emails too, as webmail is the essentially the same thing with the same pitfalls as any web hosted app, although more comfortable to many because it's been around a bit longer)
While the interface is important, I think many like the convenience and lack of carrying files around like I said earlier, and that will be hard to replicate for any desktop app.
You are effectively saying I *should* use Linux
*because* hardware manufacturers have a problem.
Er. That doesn't make sense.
I don't care WHERE the problem is. I am a USER and
I want my computer to WORK. As a user I ought not
to even KNOW the difference between software and
hardware, let alone the feud between the
manufacturing organization and the Free
programmer. Let alone justify my use of the
software with some kind of support-of-the-little-guy argument.
I am thankful to developers, but those same
developers also tried to convince me
that Linux was more useful
than it really was and provided a half-baked
solution. They were not really honest
and you don't sound like you are going to be
honest with the user either.
I'd rather pay poolah and give no thanks
and get something that works.
And don't come with that "Free" is not "free"
kwap. In PRACTICE free and Free have turned
out to be the same thing.
1999 called, they want to know what percentage of desktop users are using Linux.
I think that Linux is more than ready for the desktop but as you say people don't know that it exists. Over the weekend I reformatted my semi-corporate laptop (lenovo thinkpad r61) from Vista Business to Ubuntu 9.04. And I've been totally surprised how well the whole process went. All of the hardware worked OOTB. Today I've been even more surprised when I brought the machine to work. I connected the network cable. Fired up Evolution and connected to the Exchange server. When I wanted to print something Ubuntu had found all of the available printers on the network and let me choose which one I wanted to use. Everything has been flawless the whole way. Not to mention that it's blazing fast compared to Vista. I can even run a Windows XP environment in Virutalbox without any significant slowdown (4GB RAM) so if I need to sync my iphone apps on it I can do that too. But people don't know that the possibility exists, or even if they know the headaches caused by Windows aren't big enough to warrant a change, or a couple other excuses to just stay with what they know.
xb0x
Very little. Not just because TINC, but because those who wold speak up for Linux know better than to equate Linux with Ubuntu.
Then with what distribution of Linux-for-the-desktop should the promoters of Linux-for-the-desktop equate Linux-for-the-desktop? If not Ubuntu, then what?
Is that why Linux Desktop is such a blazing success right now ?
Please stop talking about bad support of devices in linux, but talk about stupid hardware manufacturers.
Regardless of who's to blame, the fact remains that it's "a problem with Linux on the desktop".
I'm pretty much OS-agnostic. I ran a dual-boot Windows XP/Ubuntu 8.04 system for over a year and found both had their strengths and weaknesses. If I had to choose one OS, it would HAVE to be Windows (and in fact, I just scrubbed Ubuntu from my hard disk at the weekend, leaving Windows XP SP3 as the sole OS). All my musical hardware and software will only work under Windows or OS X and I ain't buying a Mac when I've got a perfectly stable and usable Windows PC already.
Bad device support is the single biggest factor working against Linux adoption on home machines. Not everyone with a PC can perform all their computer tasks using an office suite, browser and e-mail client. Lexmark printer ? Cellphone with only a Windows sync client on the supplied disk ? Ditto for a digital camera. TV tuner card ? Webcam ?
Squirrel!
Linux is a hobby system
So wait, what does this mean, exactly? It's a hobby system that's cute to fiddle with then turn it off when I want to do "real" work? Like working with a database system that holds hundreds of millions of rows, used every day? That's in an Oracle database, running on a Linux machine.
Is my Tivo a "hobby" system? Does TomTom only make "hobby" devices ("you didn't get where you're going? Oh well, you know it's just a hobby system, right?"). I guess I shouldn't expect much from the routers, phones, and other devices that have put Linux at the core of their stack. I mean, it's just a hobby, right?
So what is a "professional" system to you? Windows? Sure, it's used a lot of professional capacities, sure there's a lot of software available for it, but are you saying it's somehow more "professional" than Linux? Why is that? Because it's written by Microsoft? Is Microsoft somehow more professional than Oracle or IBM?
Your post is breathtaking in its ignorance, and I know I'm doing myself no favors by feeding the trolls, but *come* *on*...at least a descent job of flame baiting would latch on to some obvious, specific weakness and exploit it, rightly or wrongly. This is post is just raving.
But why are Linux enthusiasts hoping for a future of Linux on the Desktop (TM)?
I mean, I am the one of the mystic, claimed-by-some-to-be-nonexistent "Linux-exclusive" users you've heard of, and I like it with a passion. However I don't understand why people like me are busy trying to push Linux to the Joe Q. Users. Is it because that a Linux future must be better than something else? But how do we know for sure? Even if we were, then why should we be pushing it for some global acceptance?
And yes, I know the technical advantages of Linux that could be beneficial to average users. I know the ideals for which Linux claims to stand and I think they are fine, but on the other hand something being fine doesn't necessarily imply that we should be pushing it everywhere. You may want to share your joyful experience with your new shiny $DISTRO desktop but everyone has his/her own definition of joyfulness.
In other words, I value a future of Everyone Happy with His/Her Own Fucking Favorate Operating System far greater than one of "Linux on the Desktop". It's all about choice, huh? We are supposed to be the more technical-savvy group so we should have understood our own needs (which means I need what I need but I don't necessarily need what $BIG_GREED_CORPORATION tells me to need), AND that ours are not necessarily shared by others, right?
Thanks for listening to my rant. I apology for the time I made you wasted in reading this post.
Colorless green Cthulhu waits dreaming furiously.
On the desktop,in the last couple of years especially, Ubuntu has driven it a long way forwards, and I enjoy trying each new release. But several fundamental things still don't work well enough and the help when things go wrong is still fairly awful.
Printing - still too hard to get up and running.
Wifi connectivity - my laptop 'just works' for any required length of time with a solid Wifi connection in Windows at home, but in several distros of Linux it has to re-establish a connection every couple of minutes.
Battery life on laptops still sucks relative to both XP and Windows 7.
Suspend/resume, and Hibernation/resume. In Windows I just fold the laptop and *know* it will close down cleanly, and come back when I open it. USB, sound, video - all will still be working when it comes back. Not so in Linux.
Yes, I as a computer user and engineer of over 20 years experience can get Ubuntu to work for me. But it's just too hard to be worthwhile. And it's a shame, but I certainly can't recommend the technophobe people I support (family, friends) switch to Linux as things are.
Gee Whiz! I didn't realize my desktop isn't working. Month after month and year after year it felt like it worked just fine.
Viruses don't matter anymore. It's all about trojans now. A stupid person using Ubuntu is just as easy to infect with a trojan as a stupid person using Windows is.
The question is which anecdote echoes most readers' experience.
As popular as the game is, and knowing it can run on a *nix variant, Blizzard still won't produce a native Linux client. So why do you suppose that is?
They're a company, so the answer is simple: it's not worth the cost. Linux's install base among desktop users is less than 10% of what OS X has, and they've stated in the past that the only reasons they develop OS X versions of their games are because some of their developers prefer OS X and because maintaining multi-platform compatibility helps them find bugs that they wouldn't find otherwise. The OS X version makes them a small amount of money, but its primary usefulness is as quality control.
(by the way, OS X is not a *nix variant, it is a full-fledged UNIX)
Karma: Terrifying (mostly affected by atrocities you've committed)
the average user just doesn't give a shit and is unwilling if not incapable of tweaking the OS to accomplish otherwise simple tasks.
Absolutely. You've hit the nail right on the head. 95% of users out there are not going to RTFM, will not open the command prompt, and will not edit a config file. Not because they're stupid, or lazy - but because it's not their job. And the sooner developers realize this, the better.
It's not a question of "how can we make the stupid users figure out that 1% of the application experience so we don't have to code a step-by-step GUI configuration util for it?", it's a question of "how do we understand that the secretary/doctor/lawyer/manager *expects* the machine to work just like every other machine in his/her universe?".
The problem is not that the average office user / home user is stupid. The problem is that they're used to their coffee-maker, microwave, fax machine, and calculator being 100% operational out-of-the-box, and the computer should not be any different.
There is a significant difference in the mentality of Joe Q. User and Jim Q. Developer when it comes to the question of what's acceptable in a computer application, and until we IT professionals suspend our hubris for a minute and try to work out a solution that "just plain works", we'll keep running into the same brick wall and wondering why it's still there.
Of course, this is IMHO, YMMV, and so on.
P.S. Someone mentioned a dearth of audio software for Linux. Here's a small list: http://habrahabr.ru/blogs/sound_and_music/59815/#habracut
Google Transation: http://translate.google.com/translate?js=n&prev=_t&hl=en&ie=UTF-8&u=http%3A%2F%2Fhabrahabr.ru%2Fblogs%2Fsound_and_music%2F59815%2F%23habracut&sl=ru&tl=en&history_state0=
Many of his points are just embarrassing. I don't get it, what pride can you take in showing that you tried a hobbyist operating system that flies in the hands of a master and then demonstrating all the areas you failed to comprehend/jury rig ?
Its not supposed to work straight out the box, because you are supposed to find the way to use with in symbiosis - if you are a paid up windows user, please don't complain when you bring home a free DIY airplane and you glue the decals on the windscreen.
"Drop to the terminal", you've already lost most users.
I appreciate the power of the terminal, and many of the folks on this site also do, but users simply don't care about the "power" of it, they care about simplicity, straightforwardness, and software.
Linux isn't yet simple, especially when people are used to doing things in Windows or even Mac (a bit), those two platforms work surprisingly alike for installing software (double click!), for finding software (go to some website, download it), etc. When a user is used to going to Mozilla.org and downloading the latest Firefox, and then tries doing that in Linux only to find that they have to drop to the terminal and do a install, they already are ready to wipe and format and put Windows back on so they can play their games and surf the web.
Windows has taken years to get a cohesive (and still not quite there yet) and unified GUI. Mac took a long time too, and it's pretty darned good. Linux is a compilation of GUI, and while it's pretty good LOOKING, it's not unified across every window, every application, etc. Plus, breaking the habit of people downloading apps from the web and going to repositories is counter intuitive for a lot of people.
If you want to make a dent where Windows is king, you have to adopt some of the peculiarities of the OS in order to adopt. You give them training wheels and the rest they learn on their own. If you fail to provide that, then they just give up and say it sucks -- like the returns OEMs show from getting *nix laptops. Even with distros that work out of the box from OEMs, people are returning them. Sure the distro can be done better, but the odds of that happening are slim, so my thought is that Linux itself needs to change at its core to help the users bridge the gap.
But I've been hearing "Linux on the desktop" for so many years now I just laugh about it now. Given the treatment of non-Linux users by Linux users (berating, combative, defensive, angry, etc), there's good reason why it never catches on, and it's because the userbase for Linux are a bunch of assholes.
The price is always right if someone else is paying.
... was the point of the article going over your head. The key phrase (from your own post) "hobbyist operating system". The point of TFA was that Linux isn't ready for the masses, not that it isn't ready for geeks. Sure, it "flies in the hands of a master". The point is that very few people are masters, and very few have the time or inclination to become masters.
Right. Which is why it's not ready for the desktop (at least for ordinary mortals).
At the time, I estimated that we were around the 50% mark toward that goal (lots of missing device drivers, buggy OpenOffice, no high-quality equivalent tools for photo editing, page layout, video editing, and much more). In short, anyone using a Linux desktop would need to have another machine to accomplish these other tasks.
In recent weeks, I have installed SLED 11, openSuse 11.1, Fedora 10, and Ubuntu 9.04 on several netbooks, notebooks, and boxes. My goal (once again) was to make one of these systems my everyday workhorse machine, one that I could recommend to friends and family for all of their computing tasks. While the situation is much improved from three years ago, we are still quite a way from reaching that elusive 100% goal. For myself and my family, I would guess that we are in the 80's, but gamers would give a much lower score.
Installation and setup is vastly improved. The desktop layouts, particularly GNOME, are reasonably familiar to users of other platforms. Individual applications, notably OpenOffice and Firefox, have come a long way. The usability of system update mechanisms ranges from the smooth (Ubuntu) to the challenging (SuSE). (Development tools are outstanding, but that isn't the issue here.)
However, I had to install restricted drivers to make wireless work, had to install commercially licensed Flash to be able to view many websites, and still found myself without programs for video editing, page layout, and photo editing that compared well with their commercial counterparts (e.g., Scribus vs. MS Publisher or Pages). Watching commercial DVDs occasionally required the use of terminal commands to download and install software, not to mention the associated legal issues. Webcams and microphones were unreliable at best, making it impossible to do video chat or broadcasting (e.g. uStream) with web-based applications.
So I renew the challenge to make it possible for average computer users to do 100% of their work using open source software. That means moving development efforts up from the operating system and infrastructure level to concentrate on creating high quality, easily used applications. That also rules out using WINE or VirtualBox to run proprietary apps.
Let's create personas and scenarios for different types of users, identify their needs, and build the needed applications and drivers. Let's also continue to push device makers to supply Linux drivers. Let's find a workable solution for Flash and SWF-based web content. (Gnash isn't quite there.) In that way, we can make some progress toward that magic 100% number that would allow people to do all of their computing on a Linux desktop.