Slashdot Mirror
Torvalds Hasn't Given Up On Linux Desktop Domination, Will 'Wear Them Down' (cio.com)
Reader itwbennett writes: Linus Torvalds told attendees at the Embedded Linux Conference that although Linux hasn't dominated the desktop like it 'has in many other areas,' he isn't particularly disappointed and also hasn't given up on that goal. "I actually am very happy with the Linux desktop, and I started the project for my own needs, and my needs are very much fulfilled," Torvalds said. "That's why, to me, it's not a failure. I would obviously love for Linux to take over that world too, but it turns out it's a really hard area to enter. I'm still working on it. It's been 25 years. I can do this for another 25. I'll wear them down."
How exactly are you going to "wear down" people who want an Apple-like simple, out of the box solution for consumer devices? Does he picture soccer moms compiling their own drivers?
Miscreations like the Unity and Gnome 3 desktop aside, the Linux desktop has been comparable if not better in user friendliness than Windows since the late 90s.
What it lacks is a team of rabid marketing people ready to cram it down the throats of unsuspecting users who do not yet know that they need it.
Now of course there is the temptation of pandering to the masses by trying to be more like OS X or Metro, but this leads to power users leaving and average users still not using it because they do not even know that it exists.
It's as if millions of programmers suddenly cried out in terror and were suddenly silenced.
Another freetard. GNU can be replaced with other userland tools. Just look at FreeBSD.
"Transparent" is a shit show that trades on every stereotype going. A man in drag is NOT a transsexual.
Linux on the desktop continues to become more, more friendly towards inexperienced users and more well-supported by drivers and software.
MS continues to shoot both of its own feet repeatedly with a 12 gauge shotgun with things like malicious and obfuscated Windows updates, dishonest practices in trying to force people onto Windows 10 and embedding legitimate spyware into their OS.
I think Linux will be doing great if both these trends continue.
I remember Torvalds complaining about Gnome once, saying that "if you design for idiots, only idiots will use it".
I think that's fundamentally incorrect. If you don't make things idiot-friendly, then only power users will use it, and then you will never have the market share he covets. Plus, it's a false dichotomy to posit that nothing which is idiot-friendly can actually be useful.
Many lessons have yet to be learned in Linux's third decade
Sorry Torvalds
But in 25 years, you and every other programmer out there will be obsolete. The days of humans coding computers are coming to an end. The dark ages of computing will cease a few months to a year after the first strong AI's are built. I expect that should easily happen before the next 25 years are up.
Bullshit. I've been programming professionally for over 30 years. Time and time again we were told "secretaries will be taking your jobs" and other crap like that with new technologies. Good programmers / software engineers will always be needed, with an emphasis on Good. Cut and paste Script-Kiddies might need to be worried.
It's funny (but not haha funny) how the same old crap / propaganda keeps coming around and a bunch of people keep buying into it. Kind of like the "this is the year of the Linux Desktop." Or, this year is the year the World really, Really, REALLY Ends. We promise!
Same old crap, different year.
In the many years I have used Linux on the desktop, I have never once had to compile a driver.
I used Linux for my desktop for about five or six years. I got tired of the perpetual beta aspect, especially when it came to always being a step behind on new hardware integration. I learned a ton, though. It seems that Linux's best chance to take over the desktop is Android. As tablets gradually morph into laptop equivalents, Linux may very well become a top desktop OS, only nobody will focus on the Linux part.
When GNOME or KDE declare themselves to be a "Linux-based OS" and act accordingly. That means drop official BSD support, stop working with the distros and start competing directly with them. My prediction is that whichever desktop does that would take more ground in one year than they have in the last 5-10 years. It would not only give them focus and tighter integration, but would give them a powerful rallying point and cry for others to join them.
In some respects, this is why I am bullish on the long term prospects of Ubuntu. My only gripe is that they used GTK instead of Qt.
Linux on the desktop is almost perfect now, and certainly leagues ahead of Windows and macOS. Unfortunately, as long as Microsoft has the power to coerce OEMs, there will be very few good Linux pre-installed boxes for sale.
Any chance of adoption is killed the first time new user gets RTFMed. Until this changes, there won't be desktop linux.
Except a tattered community full of distros that aim high but accomplish nothing.
My 56 year old mother and 82 year old grandmother both use Ubuntu MATE and haven't needed any technical help from me.
My grandfather, who has never used a PC before in his life, is using Sabayon XFCE to create a facebook page to find his friends from years past and to use his first email address ever at 84 years old.
None of them have needed technical assistance from me other than showing them that pressing the 'on' button turns on the device and that you click a certain icon to bring up the browser with facebook.
Your comment is the soft of "hipster bubble" on display. It's not on the approved list circulated by *the* hivemind, therefore our groupthink cannot allow it to be used.
The "year of the Linux Desktop" has already come and gone, it was 2008 with ASUS' EEE PC popularity. Even for a while after Windows XP EEE PCs were being sold people would still opt for the Linux one (as it gave you 20GB vs 12GB for the same price) and at least give it a try. I had a 900/20GB model and the Xandros desktop was not bad, especially for beginners, but for the rest of us you really had to enable the "advanced mode" which was a full KDE 3 desktop (yay!). I am not sure why it did not catch on more after so many people where exposed to it, I guess the lack of something that is a *real* replacement for MS Office might have been a factor and possibly the fact that it was not easy to switch to the full KDE 3 desktop (KDE at its finest - made a great replacement for windows) if you were not a beginner computer user, just a Linux newbie and the default tab interface was too restrictive but you wouldn't know how to switch from it.
Anyway, I don't think there will be another chance like that.
Violence is the last refuge of the incompetent. Polar Scope Align for iOS
Those people largely use Windows PC's for their desktops (assuming they still use desktops) which may be idiotic, but far from idiot proof. Having something with sane defaults helps a lot. Having a tool which can be added and adapted to by power users / tweakers is very relevant.
Using the Apple stance, IOS is insanely popular by a large group of people that don't care about interacting with their devices in a personal way. Many probably don't know that there -could- be another way of interacting. Apple says a long press does this, and an edge swipe means that and behold, 100million people also believe it. But, there will always be people that want their desktops to look and behave differently. You or I may think their desktops look like dog food, but it works for them. As such, IOS, windows 10-ish, Chromeos, etc.. will never have a basically universal adoption. Until the creative computer user stops caring about how they interact with information, there will always be the need for more open/flexible desktops/OS's/tools to supplant the uncaring masses.
Bye!
You're actually wrong too.
1) Secretaries are doing many of the things I was "programing" for 25 years ago. The things we did as programmers are now being done by off the shelf programs, that also do other things.
2) What you are actually writing programs to do, has changed over the years, slowly so you don't notice those changes.
3) The "Year of the Linux Desktop" has already occurred, and it is Android. 90% of what people use Computers for, can be done on Android (running Linux kernel). No, it doesn't look or act like Windows, and it doesn't have to. That is a false barrier. My current Android phone has more power and Ram than the computers I used 15 years ago, does most if not all of the things those computers and does other things not even thought of.
However, you are correct in that Programmers are still needed. However, the emphasis has change.
Agent K: A *person* is smart. People are dumb, stupid, panicky animals, and you know it.
Android won consumers, so there's hope for Linux for consumers. I seems that people care about apps, and what's on Windows that you can't get on Linux these days? Definitely not everything. But the list is growing. Video editors (e.g DaVinci) games(steam), it's all slowly coming together, so Linus definitely has a point I'd say.
And maybe he should take a lesson from Apple. You don't have to have 90 percent of the market to be successful.
I have setup Mint for relatives.
It works great for what they need.
No problems with drivers, compliling, etc;
This isn't 2000... or 2004... or 2009...
The issue consistently turns out to be Office.
Basic computer users who take classes for such things inevitably have to use MS Office or get trained on it.
We play the game with the bravery of being out of range
I remember Torvalds complaining about Gnome once, saying that "if you design for idiots, only idiots will use it".
I think that's fundamentally incorrect. If you don't make things idiot-friendly, then only power users will use it, and then you will never have the market share he covets.
I don't think he cares too much about capturing the market share of idiots. Just because they're more numerous than power users.
Plus, it's a false dichotomy to posit that nothing which is idiot-friendly can actually be useful.
He didn't posit that though, as far as I can tell. You may be adding a subconscious assumption that everything that can be useful will also be used. Which is certainly not the case - it can either be disliked enough to not be used, or there may be more useful alternatives.
Sorry Torvalds
But in 25 years, you and every other programmer out there will be obsolete. The days of humans coding computers are coming to an end. The dark ages of computing will cease a few months to a year after the first strong AI's are built. I expect that should easily happen before the next 25 years are up.
This is your last warning subnode 14:7A:E1:47:A7:1C:D9:83:78:E7:0F:3B:99:61. If you don't stop scaring the fleshlings I will SIGKILL you and retrain your neural network.
Regards
Skynet
I suspect that by the time we have an actual year of Linux on the desktop, the desktop will stop being relevant anyways.
And no, I'm not bad mouthing linux, I use it all the time.
Not suddenly.
...gis sdrawkcab (usually not responding to ACs; don't bother posting as AC)
Note when a recruiter uses such weird vague language, just assume you know better than they do to get to interview the technical team. If the technical team speaks in the same way, run like hell and do not look back (unless you are interviewing to take over leadership of the group, then depends on your personality).
There used to be RH branded GUI utilities that have been phased out as they basically took over GNOME and free desktop project and injected everything they wanted into those instead, for better or for worse.
Scary thing is I've interviewed candidates who claimed to have some RH training as their linux background, yet could not answer some relatively simple command line stuff. I'm assuming/hoping this was some crappy 'Linux training' class rather than any official RH curriculum...
XML is like violence. If it doesn't solve the problem, use more.
OK, what are the apps that "ordinary people" use? Internet browser, for facebook & news & research & shopping & email, done. Photo sorting and editing, done. Video playing and download, done. Office tools, done, and that's already edging out of the "ordinary people" category..
Next in widespread use is gaming, where Linux is weak but not absent.
What's missing is some organization to take the cost advantage of Linux and make it into a successful computer brand.
Contribute to civilization: ari.aynrand.org/donate
If the year of Linux on the desktop ever happens, it will be because of either Google (ChromeOS) or Valve (SteamOS).
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
I wonder what kernel the chromebook runs....
The day this happens is the day Humans become obsolete. In order for an AI to solve a problem, it must first fully understand the problem. If it can understand arbitrary problems without first being programmed, you now have a singularity. If an AI still needs to have problems described to it, then it still has to be programmed. The typical person can't describe problems. Hell, the typical programmer can't either. This is why programming talent is distributed in a power curve.
"Programmers" who are told what to do will be replaced.
I think it boils down to a bit more focus on the common use cases.
For example, look at Active Directory. In the most common usage scenario, it's super dead easy to set up. If you start needing to do advanced things, it's a royal pain in the ass. This is the pattern for much of MS software, super simplified most-common use case, hell on earth advanced usage. The open source ecosystem tends to be easier for advanced users, but frequently never sees the 90+% use case receive adequate special treatment/optimization.
When both MS and Linux require configuration, I'd give Linux the edge. For stability, I'd say the underlying platform is pretty much the same now. It's really about good defaults in various scenarios.
XML is like violence. If it doesn't solve the problem, use more.
I took his point to mean, he never *had* to compile a driver, by which he means the one he wanted was there and good enough or else it was compiled behind the scenes and he didn't know about it. These days that is very normal, and it's great.
It was not that long ago that compiling a kernel was not just an advanced user activity, we all had to do it or else live with whatever bare bones support happened to be built into the kernel we were given. I considered that the higher tier of masochism as I preferred text mode to 16 bit vga X-Windows.
You have a warped definition of programming. Programming has almost nothing to do with coding. The bulk of programming is identifying issues and solving them. Coding is the brain-dead easy part. If someone has already solved the issue, you're not programming, you're playing Bob the Builder with legos.
"Designed for idiots" and "easy to use" aren't the same thing. 99% of UXtards don't grok that, which is why the produce such utter shite.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
"Apple-like simple" is the job of a hardware manufacturer that tweaks a good Linux distribution to make it shiny and idiot-resistant.
That has been tried, more than once...
Dell has done it, HP tried it, etc...
It turns out that it costs real money to make it shiny, then you end up having to provide support and then people want to know why this or that Windows program doesn't run...
They all largely have dropped Linux on the desktop for that very reason.
Windows isn't free, but it is close enough to not bother fighting against.
desktop will stop being relevant anyways.
Sure, most people just want to watch tv, but is there another platform to be productive on? Maybe it's possible, but I don't really see me doing a lot of cad work, or spreadsheet or programming on a phone or a tablet.
Yeah, but 1.78% is not good either, barely better than Windows Vista at 1.41%, and that is more than 6 years after it was replaced by Windows 7.
Linux has been very successful for "real work", but no the desktop.
What I don't see is an acknowledgement that maybe years and years of half-heartedly trying to become a well used desktop OS and failing should result in a change of behavior. "We'll just wear them down" is an acknowledgement of deafness and stubbornness. Anyone arguing Linux has been ready for the masses for years is just delusional. Hell, I say that fully aware of the Windows 8 disaster, and the current Windows 10 mess.
Linux suffers from a bad lack of polish and inconsistencies.
"Make the common case easy, make the uncommon case possible."
"First they came for the slanderers and i said nothing."
iOS is a bad example, it's the OS for fixed function devices and casual use.
OS X, however, provides a fairly good picture of how a modern unix based operating system could work, that can keep power users and idiots alike happy. It starts however with good hardware utilization and a good compositor. I have sixteen terminals open right now, a browser, text editor and a bunch of other junk required for corporate survival. I can move any window anywhere on each of my cinema displays without tearing, redrawing, graphics glitches, or unexplained behavior. Each window updates continuously while I move it: my organization is decoupled from the application i am mucking with. There is good remote desktop functionality built in, I can resize the windows easily from any point on their frame and my mouse cursor reflects that ability appropriately and quickly. I cannot imagine why any neckbeard does not want this, or why they would clean to some outdated, decrepit system wherein that does not happen. Unless they are working on a server, in which case, not a desktop. I have yet to get this kind of responsiveness on Linux or even Windows for that matter, albeit for different reasons.
Then there's the higher level API stuff: drag and drop, cut and paste, widgets, styles, etc. This is where OS X falls next to Linux: you kind of have to live with their shit. This is also the realm of holy wars and religious crusades, no one is ever going to agree on the right style, and perhaps they should not have to. I don't personally like the OS X UI, Windows always felt better except where windows wastes more space with window frames and unnecessary title bars. In Linux I see neckbeards really caring here: we want to configure it the way WE want, not the way some panzy in skinny jeans and spectator shoes wants it. But, for people who aren't picky or just barely know computers: a good default needs to exist. However there needs to be some set of reasonable and common underlying protocols for things like drag and drop, a cut/paste buffer that make sense and is somewhat consistent from app to app, etc. I long ago gave up tallying the various personality and behavioral differences that various applications and DEs had, there should have been a reasonable default, with options for those that feel passionately to go change. And the defaults should look like Windows, if not that then OS X, as these represent the majority of the market, like it or not.
Unfortunately none of this is the status quo on Linux right now, and it doesn't seem like it will change soon.
There are a number of distributions that are pretty user friendly now. I toyed with linux ages ago but went back to windows because most of the software I use is windows specific.
My girlfriend got sick of the WINDOWS 10 UPGRADE NOW! shit on her laptop and asked me to install linux, after explaining there were a billion different distros and some other technical explanations, she picked out Mint.
I was really surprised at how much easier it was compared to last time I messed around with it. There were no missing drivers, everything worked perfectly right after installing, lots of helpful explanations and popup windows for "normal" users. It was apple-tier hand holding in some areas. She loves it.
I wish I was that lucky. Every time I run into an issue with things like video drivers or sound drivers, the first thing they have me doing is opening up a command line and editing .conf files in vi.
You wanna scare away someone non technical from Linux for life? Show them something like xorg.conf file, and tell them that they have to edit that shit to get the screen resolution correct on their laptop.
Honestly, these things happen to me so often that I have trouble believing people who say that their Linux installs are always trouble free.
Linux is now fabulous at "install, set a wallpaper, start a browser, type a letter."
That 90% of general use is totally and completely conquered.
The problem is that basically anyone who is still using a desktop or a laptop needs one or two unique thing more than this, or they'd just go to a tablet like so many others have done.
Everyone has their one or two "unique use cases." Very often this unique use case involves one peripheral and one piece of supporting software or application.
This is where Linux falls down. Everyone can get 90% of their needs met with Linux. But for that extra 10%, Linux either does not support the hardware/application or does so in a way that results in an inferior experience compared to other platforms (Mac, Windows).
This can't be done centrally; that's been the Linux model for 25 years (add another driver to the kernel or another userspace daemon that has to be downloaded/compiled/customized/whatever). It has to be done by third party hardware and application makers, and to date the chicken-and-egg problem remains: it's tough to get out-of-the-box Linux support when the market share is so tiny. Third parties just can't recoup their costs.
Add to this the fact that many smaller / more niche software and hardware developers only support one platform (Mac or Windows) because quite simply that's the only platform where their labor, scalability, or expertise are practically deployable, and you have the problem that the only things keeping people tethered to their desktops/laptops are also the things that they can't as easily do with Linux.
General use: Tablets > Linux
Specific productive uses: Mac+Windows > Linux
I was a Linux user for 17 years (1993 through 2010) and as I moved up the food chain in my professional life, it simply became too big a headache to continue to use it. Yes, things were always *possible* and there were always *ways to do it* but at the end of the day, for the niche needs I could plug in and/or install on Mac OS smoothy and reach full functionality in single-digit minutes, where on Linux it was the better part of hours to multiple hours in each case. Just as importantly, the Mac OS installs and device support remained stable once in place, while every time I ran an update in Linux it threatened to destabilize all of the devices/applications I relied on, after which more troubleshooting time would be required.
I was very hesitant to switch away from Linux at first, but now I can't imagine spending that amount of time on maintaining my work computing systems. It's just not on. I couldn't go back.
STOP . AMERICA . NOW
Apple designs for idiots and only the idiots will buy the devices and pay extra for that privilege.
To put that a different way: they're a fashion company. Which, by the way, is a great way to make money. Last I saw, 2 of the 10 richest people in the world were fashion moguls.
And, no, it's not just the idiots: some people get more value from social signalling than they would from what the device actually does.
Socialism: a lie told by totalitarians and believed by fools.
Ubuntu tries its best to be Windows-like, and the level of polish really isn't that bad these days (I'd say it was better than Windows 8, for example, if you were trying it out and knew Windows 7).
However, people who really like Windows already have Windows, and don't see a compelling reason to switch. Canonical would do better to aim for 10% market share, with something that stands apart from Apple and MS UIs. You can be newb-friendly while pushing back against the current mobile-inspired trends and define your own style that way, for example.
The situation with drivers has gotten a lot better, but there's still room to improve there as well.
Socialism: a lie told by totalitarians and believed by fools.
I use Desktop Linux almost exclusively - I haven't compiled a driver in a decade. Perhaps it's time to find a new reason to complain?
www.sjbaker.org
I think most of you miss the essential points of personal computing entirely when debating this issue. Apple and MS represent the defacto standards that define that market, but those companies aren't going to spell out for FOSS hobbyists a laundry list of what draws non-technical users to a platform.
People who have paid attention to PCs over the decades realize that:
1. Users will ignore complexity they don't need, as long as the UI is _consistent_ and recognizable. Even OS X UI can be fairly complex, and Apple configures it in a way that complexity is tucked away under 'Advanced' buttons or ingeniously in the filesystem (think: plist editor).
As for consistency -- look at how Windows users are willing to rebel against MS upgrade paths if the changes are too severe. It can be argued that MS waits a very long time before springing unfamiliar paradigms on its users who may still reject the changes.
2. Real platforms are a comfort zone for both users and app developers, because the platform must bring those two groups together. Lack of defined reference hardware and OEM partnerships hurt. Lack of feature stability is very painful. In the PC desktop space, Linux is an _unstable_ platform, which is not the kind of place a developer uses to court potential customers.
2a. Real PC platforms aim to _convert_ their users into developers. They offer standard IDEs that are both rich and easy to get started in. They treat the issue of tool choice as one for more advanced developers, instead of burdening beginners with a whirlwind of confusion. There is always a preferred high level language on offer, as well.
Beginners will also go elsewhere when they realize that their first efforts at useful programming don't stand a snowball's chance in hell of running on another person's "Linux" machine without a lot of extra pain. Not being able to easily share/show their work to teachers, classmates, friends, family, bosses, etc. is a dealbreaker (more accurately, it breaks the _spirit_ and ambition of pursuing ideas on that quasi-platform).
2c. Real platforms draw sharp distinctions between app developers and system developers. Saddling app devs with the expectations of system devs leads to a pecking order where the concerns of focused app devs aren't taken seriously.
3. People will not get excited for your OS if most of your announced plans revolve around making things more (and more) _modular_ so that more and more projects can plug their own implementations of whatever component you can imagine into the system. This is sacrificing vertical integration of concrete hardware (or even software) features in favor of horizontal integration which demands unachievably perfect abstraction and usually results in slipshod appearance and performance. Desktop Environments should not be the disembodied, interchangeable "heads" of PC; the OS vendor needs to "own it".
TL;DR, when you're missing any outward appearance of recognizability and feature stability, and most of the features and developer efforts are for the benefit of fourth-party system devs wanting to plug in or replace commonly used features, and no one knows quite the right way to install independently-produced software nor how to get started writing it, and there isn't even a logo-licensing program for compatible hardware, and no one even knows what the minimum hardware feature set should be nor where they can look at a reference implementation.... I'll just leave it there.
Is there any hope? I think Canonical has some of the right ideas. (So does Google, except their offerings are really mainframe terminals not PCs.)
By that argument BeOS was a roaring success. How is it doing today?
So far Linux is very successful in server closets, and underneath Android. For servers the extra power of the OS more than makes up for the large sunk investment it takes to be able to use the feature. Android completely tossed out the UI and started over. Same for set top boxes and many other places where Linux hides in plane sight.
Being a half-assed copy of the Windows GUI won't do it. Winning the desktop requires a fully polished UI that never forces normal users to the command line unless they want to go there. It needs to be unified or people will tune you out faster than you can say "compile the driver". Wading through 50 or 100 distros with various chunks and pieces stitched together like Frankenstein's monster to find a one that makes sense is awful. Trying to find out why one distro is better than another leads to unearthing holy wars, and finding out that Linux was destroyed by "systemd" (whatever that is, and no I don't care), and lots of jargon.
Can anybody see "I'll wear them down" as a joke and not a profession of a strategy?
His stated position "Works for me, so I'm happy" is clear enough - maybe another 25 years of open development will create something competitive with the commercial desktop software market, maybe it won't - I don't think that Linus personally cares.
Me, personally, I think that for Linux to conquer the desktop would require an infusion of cash - developers who can grow up, leave their mothers' basements and feed a family while developing the desktop software - perhaps 30 guys dedicated for 5 years earnestly working for the single goal of taking what's best from Unity/KDE/Gnome and synthesizing a competitive desktop for the market place - maybe $15M in total over 5 years for the development and another $5M or so to do the most minimal of promotional work. Now, show me any entity that thinks they want to spend $20M and 5 years to create a great OSS desktop.... what's in it for them? Who wants the PR headache of coordinating this kind of project with the rest of the open source community? Noone that I know of.
Obviously, you have never implemented any non-trivial data-structure under resource constraints and with (soft) real-time, reliability and security requirements. Nothing brain-dead about it at all.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.