In the Depths of the Cloud, Open Source and Proprietary Leviathans Fight to the Death
Jono Bacon Asked Google Home ‘Who Founded Linux?’ You Won’t Believe What Happened Next!
Red Hat's Women in Open Source Award Winners, 2017
Imagine an Android Phone Without Linux Inside
Linus Torvalds Talks to Debian Users
Mozilla Relents, Thunderbird Can Stay
Heed the Prophet Stallman, oh Software Sinners!
August 15th, 2016

Why Desktop Linux Still Hasn’t Taken Over the World

The reason why use of the Linux desktop has never taken off has nothing to do with the operating system and everything to do with money.

“The desktop hasn’t really taken over the world like Linux has in many other areas…,” said Linus Torvalds in April at the Embedded Linux Conference in San Diego. “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.”

Linux desktopWhy, after 25 years, is the GNU/Linux desktop still near the bottom of the “market share” list of consumer operating systems? It’s certainly not due to quality. Those of us who use Linux on a regular basis and who have experience with other operating systems as well, pretty much agree it’s the best. We also know it to be an industry leader, with features showing up in desktop Linux years before they make their way into, say, Windows. By all rights, Linux should be known by techie and non-techie alike as the superstar of desktop operating systems.

So why hasn’t that happened? One word: marketing.

The only way to successfully bring about widespread adoption of any consumer product is by spending big bucks marketing it. That’s never been done with Linux, and for good reason. There’s no money in it — even if wildly successful. It would be, as business people like to say, throwing good money after bad. That’s because when it comes to monetizing Linux, its biggest weakness is its greatest strength in every other way: its free nature.

Let’s say Canonical were to get serious about marketing desktop Ubuntu. There’s little doubt in my mind that if they wanted, and were willing to spend the necessary dollars marketing, the distro could take a large slice of the operating system pie and even get OEMs on board to produce affordable Ubuntu boxes, probably quicker than most people would think.

One way they could do this would be by using an old mainstay of the advertising business, co-op advertising, which is often used as a carrot on a stick to get retailers to carry a particular product. In the case of Ubuntu, Canonical could leverage co-opt advertising not only to get Ubuntu machines into stores, but to get OEMs onboard to offer computers and laptops with the distro preinstalled for the stores to sell.

With co-op advertising, the manufacturer (in this case Canonical) agrees to foot a large portion of the advertising bill for any ad that mentions the product in ways that meet criteria spelled out by the manufacturer.

How might this work for Canonical?

Let’s say it approached an OEM such as Dell about offering Ubuntu on desktops and laptops that are selling well with Windows installed. Canonical could offer to help Dell make sure Ubuntu would work perfectly on these machines, and maybe even throw in a little limited technical support to those who buy them. They would also offer to help with the marketing with a co-op advertising deal that would not only apply to any advertising efforts made directly by Dell, but to any advertising done by retailers carrying the product as well.

Let’s say Canonical agrees to pay 75 percent of the advertising costs, which is not an unheard of amount. This co-op advertising offer would stipulate that the focus of the ads would have to be on Dell’s Ubuntu machines and that the ads must extol the virtue of the distro as a premium operating system, maybe using a tag line such as “a cut above.”

Dell could probably be convinced to go along, as they would have next to nothing to lose. It has the machines anyway and will continue to profit from Windows versions. For the Ubuntu editions, Canonical will do all of the heavy lifting with making sure the OS works well and will pick up the lion’s share of the marketing cost. Getting local retailers to stock the machines should pose little problem, as they’d have the incentive of free advertising. Many local radio, TV and newspaper outlets pitch local businesses on available co-op plans with schemes that rebate the percentage that would be the retailer’s stipulated share of the cost.

It could be done but it won’t, because it would cost Canonical tens of millions of dollars, without a prayer of return.

Even if Dell began pushing millions of computers out the door, there would be very little money in it for Canonical. It’s hard to sell for a profit something that can be downloaded free. The Red Hat business model of making money through support won’t fly because consumers computer users have never spent money on support. An app center for proprietary software won’t work because consumers pretty much don’t buy software anymore (besides, almost everything a Linux user needs is available for free), and affiliate marketing plans with the likes of Amazon are also unlikely to earn back enough to even come close to covering expenses.

I’m sure Canonical looked at many avenues to monetize desktop Ubuntu before coming to the conclusion that it’s best to focus marketing efforts on the server, cloud, and perhaps eventually, phone. There’s also little doubt that the saavy business folks at Red Hat continue to take good long looks at the prospects for monetizing desktop Linux, and if they thought it possible, they’d have an easy-to-use distro on the street tomorrow. The OEMs aren’t willing to really get behind preinstalled Linux on low cost machines, because consumer acceptance would require gambling a lot of money in a market that’s currently shrinking.

Until somebody can figure out a surefire way to monetize a desktop distro, GNU/Linux will continue to be primarily used by free software advocates, people who demand the best on their desktop, those trying to keep an older machine productive and by those who’ve had-it-up-to-here with Microsoft.

The good news is that the last two groups are growing in number.

The following two tabs change content below.
Christine Hall has been a journalist since 1971. In 2001, she began writing a weekly consumer computer column and started covering Linux and FOSS in 2002 after making the switch to GNU/Linux. Follow her on Twitter: @BrideOfLinux

138 comments to Why Desktop Linux Still Hasn’t Taken Over the World

  • Eric Espino

    Is because the business services that you can access from the desktop out of the box without complicated setup procedures. i.e. are Active Directory and OWA. Also is the inability to provide VBA compatibility for the open source office suites.

  • There are a few ways to make money on a free (as in beer) operating system. Apple’s is the most straightforward: it makes its money from the hardware, and (in iOS’s case) from taking a cut of every piece of software sold.

    Google goes about it a different way: it doesn’t make a profit on the hardware (at least, not usually), and its ecosystem isn’t as locked-down as Apple’s, but it has a similar app store that grants it a similar taste of app sales. Beyond that, though, it’s all tied to a Google account, which feeds Google’s real business: data analysis.

    Microsoft has largely copied Google’s methods in Windows 10. And while Windows 10 hasn’t set the world on fire the way Android has, it’s managed to pull a pretty massive userbase, largely from taking its existing, even more massive userbase and offering a free update (sometimes whether the user wants it or not).

    Canonical’s tried similar techniques and met much greater resistance than MS, Google, or Apple, because its existing userbase is different from those other companies’. When Ubuntu added search ads to the program menu, there was a huge outcry and it was eventually abandoned; now Microsoft has gone and done the same thing, and while privacy advocates are justifiably upset, most end users seem unconcerned. Ubuntu’s app store, similarly, was met with a collective shrug by users and publishers alike, and eventually abandoned. And while MS’s app store hasn’t been as popular as iOS’s or Android’s, I don’t think it will prove to be as big of a dud as Ubuntu’s, either.

    To some extent, I think there’s a catch-22 at work here: Ubuntu has a different audience than Windows, and it will continue to do so simply because it already does. It has a different audience from MacOS, iOS, and Android, too; about the only OS I can see its audience competing directly with is ChromeOS, and again, it doesn’t have Google’s financial incentives to give away the OS for free and then make a profit harvesting user information and taking a cut from app sales.

    It’s also almost certainly missed the boat on becoming a viable competitor in the mobile space, unless it can introduce compatibility with Android apps.

    It *does* have a potential “in” with convertible devices, the old “plug your phone into a dock and it becomes a desktop” premise. Canonical is farther along in this type of convergence than any other company except MS, and MS has left an opening by all but abandoning Windows Phone. But it’s still a longshot, and I expect that despite its best efforts, Canonical’s share of the mobile market will continue to be even less than its share of the desktop market.

    Server/cloud, as you note, is another story.

  • Dion Dennis

    What I’d like to see is a sustained set of viral videos, not necessarily from Canonical or Dell (top down), but more of a PR-campaign, from the user, up. For example, who among the rich and famous uses Ubuntu, Linux Mint, Fedora, etc? Or, who among the potentially charismatic and talented, a potential YouTube star, uses Linux? And how? What about a “100 days of Linux” campaign, with such videos, and either relatively high-profile users (Cory Doctorow, for example) or potentially break-out charismatic users or programmers? I’m not aware of such an organized campaign, with videos with good, compelling, imagintive production values, highlighting both mundane and novel uses of Linux. (There are perhaps hundreds of thousands of videos on Linux, but these aren’t geared toward persuading non-Linux users).

    That is where I’d think about starting.

  • Dion Dennis

    (Typo Correction) What I’d like to see is a sustained set of viral videos, not necessarily from Canonical or Dell (top down), but more of a PR-campaign, from the user, up. For example, who among the rich and famous uses Ubuntu, Linux Mint, Fedora, etc? Or, who among the potentially charismatic and talented, a potential YouTube star, uses Linux? And how? What about a “100 days of Linux” campaign, with such videos, and either relatively high-profile users (Cory Doctorow, for example) or potentially break-out charismatic users or programmers? I’m not aware of such an organized campaign, with videos with good, compelling, imaginative production values, highlighting both mundane and novel uses of Linux. (There are perhaps hundreds of thousands of videos on Linux, but these aren’t geared toward persuading non-Linux users).

    That is where I’d think about starting.

  • Bob W

    From my point of view, I think businesses who are not familiar with Linux and FOSS will be more inclined to play it safe with MS and Apple and their offerings even though the cost are much greater. Plus compatibility with all their outside business partners come into play or perceived compatibility. One last thing that many businesses will opt for Apple or MS is that many FOSS applications are not as polished and can be quickly unsupported.

  • mikef90000

    Hmmm, I thought Chromebooks were slowly doing this.

    As far as ‘regular’ desktops, I agree with the IT comfort factor. The middle aged guys really need to be shown how well (or not) Samba 4 and LTSP would fit into their environment.

    Ultimately it is all about the applications. Too many shops buy MS Office on Windows when only a small percentage of users needs critical functionality (Excel macros, etc). Red Hat too deserves some blame due to their lack of lightweight DE (xfce, lxde) offerings.

  • James

    Nope. As a Windows user who switched to Ubuntu a year ago, here are the problems:

    Drivers. I still struggle to get stuff like HP printers and Wacom tablets to work properly on Ubuntu. Hunting down and installing graphics drivers is complicated. I have even resorted to a Windows dual boot, as it is just easier operating most peripherals such as printers and scanners there than to do it through Linux.

    Convoluted troubleshooting. So many problems and fixes require fiddling with terminal commands. Nobody in Mac or Windows has used a terminal interface since DOS. My wife flat out refuses to install Ubuntu because of this.

    Advice. There is little consistency online in how to do certain things. I often joke that if there is a problem in Ubuntu, I can find ten ways to fix it – and none aren’t technical. I waste so much time finding the best fix. Mac and Windows are far simpler for novices to troubleshoot.

    Software. Only next month will there be a Firefox that can support Netflix. And installing Chrome is complicated, involving adding repositories since the download approach always fails. And the application manager only supports Chromium, which doesn’t support Netflix. This is one of many software examples I can name. But me a beer and I’ll tell of my misadventures with Skype.

    Linux desktop is far from ready for primetime. It’s my primary system and I love it. But it’s not easy enough yet for average users.

  • Eddie G.

    I said it before, and I’ll say it again. I don’t think there even NEEDS to be a Year Of The Linux Desktop! We’re moving past that now, and if there WERE such a thing? it’s already come and gone. Desktops have declined in sales and everyone’s abuzz with touch-screens, tablets, the cloud and wireless EVERYTHING! So for those of us who are “veterans” of Linux-use, we should focus more on what’s coming, and not worry so much about the YOTLD. Because even if it were to happen tomorrow? It really wouldn’t chamge anything, Microsoft and Apple have way too much monet at their disposal for marketing and advertising that it would be the equivalent to dropping one drop of water into the storm ravaged sea!
    This is by no means an insult or some form of negative thinkining towards Linux, not by ANY means! I’m just trying to state that the only thing that matters is that we promote Linux to as MANY people as possible! Until something “negatively impacts” both Microsoft and Apple who have had decades to build their cash reserves for whatever purposes they might need it for, then Linux WON’T be on every desktop. And wehn you factor in the amount of people who seemingly “trust” Microsoft, and refuse to go/look elsewhere for their OS, or the ones who have signed their soul over to Apple, then its easy to see that the numbers won’t change for quite some time. Is the Linux home user base growing? By leaps and bounds, EXPONENTIALLY! But for now, it’s not going to surpass MS or Apple. At least, I don’t think we will….

  • madman

    “” It’s certainly not due to quality. Those of us who use Linux on a regular basis and who have experience with other operating systems as well, pretty much agree it’s the best. “”

    That right there is a major issue with Linux software. The people using it and making it never spend a lot of time looking elsewhere. I love Linux but as an artist I can tell you the programs you have access to on windows (that don’t run in wine) are vastly better then the ones on Linux, and I’m not talking about photoshop. Sure PS is pretty good but you can pretty much do the same thing in GIMP with some learning and save a lot of money on the other hand when it comes to the real painting power like Painter and ClipPaintStudio the best you have on linux is Krita which is a LONG ways away from being on the same playing field and doesn’t save you enough money to care about. Same goes for 3D software and CAD. Yes you have options on Linux, but they don’t actually bold well vs whats on windows and even macOS. Linux is a great day-to-day OS and for that it’s awesome, but when it comes to a lot of art and design work windows 7 tends to win out by a mile.

    Overall this is just a single field, I’m sure there are other fields that Linux software quality is lacking just as much or more. The sad fact of the matter with desktop linux is that it has horrible high end software support and the community projects tend to fall short for everything thing not super main steam.

  • Askfor

    That is a wrong question to ask. Right question is why would Linux desktop take over the world.

    I am using Linux because I like UNIX and Linux is close enough. I have been using FreeBSD, as well. If I wasn’t a UNIX person, I’d probably use Windows. There is much more software available for Windows. Also, importance of video games should not be underestimated, and there are much more of them on Windows.

    Windows have been improved, they are not so insecure any more. They are not slow and bloated any more when compared to desktop environments like GNOME or KDE. Also, they are not so expensive, either, especially OEM.

    Average users don’t care for issues like scripting, because majority of them can’t write a single line of code.

    Linux advocates often emphasize Freedom, but I don’t think that philosophy is an issue for average user.

    In the end te result is that average users don’t use Linux. Linux needs some sort of selling point, something distinctive and desirable in the eyes of average users. The key is to offer something which is not available anywhere else, whic might be difficult. “We have the same as others” is not much of the selling point. It is like Russian cars, they are probably decent today, but few people buy them outside Russia and CIS.

    Linux advocates should start by analyzing how the people are using their computers. How often they use it for work and how often for entertainment. What applications are used most often and how frequently.

  • Ron

    Though I’ve been an avid Linux user and evangelist for many years, it has been difficult to get people past the “pain point” of seamless document exchange.

    Whatever you may think, MS Word is *the* standard across most enterprises. If you want to do business with them, you need to be able to properly handle the MS Office suite of file formats. Sadly, despite great advances in recent years, LibreOffice and OpenOffice are still not there yet.

    And I can tell you that my own product, “8th”, is cross-platform. But 90% of my users are using Windows, not Linux or OS X (even though I only develop on Linux and OS X).

  • James

    I totally agree. Many Linux applications are not as good as their Windows/Mac alternatives. Two apps I still miss a lot are Irfanview and Teracopy. Nothing on Ubuntu comes close.

  • James

    I’ll add another Ubuntu story. Recently I tried to upgrade from 14 to 16, only to discover I have a lot of broken relationships or something like that. I don’t fiddle with the OS – these were the result of simply trying to install different drivers, applications or whatever advice I followed to try and fix a problem. Consequently I ended up with a rather stunted and ill-performing OS. I tried to fix these using online advice and terminal commands – and succeeded in breaking the OS.

    I ultimately had to reinstall the system completely. Now I am very paranoid about installing anything new. I’ve set up a Windows box and use that for the heavy lifting – gaming, graphics manipulation, video editing, etc. Ironically it is harder to ruin Windows than Linux. Ubuntu purely serves as a place for my browser, downloads and business – as it is far more secure. I don’t dare add new repositories or install drivers that aren’t native to the installation, because that is a slippery slope.

    Linux Desktop is very far from ready for normal users. The real problem is the Linux community don’t seem to get this. The above fantasy that it’s all marketing related just reinforces my opinion.

  • oiaohm

    madman with cad it depend what field it is. The top factory design cad is Linux only it provide a web interface to windows and os x and everything else. Yes native client on X11. Of course that is not free software and it about the most expensive cad out there.

    Same happens with 3D software as well. There is a lot of the high end 3D software that is Linux only. The stuff that is 3000 dollars+ a seat. So you could say there is a issue with reasonably priced software in many fields. Not always a quality of software problem.

    Really I am not sure on high end software support being poor the mid range between high end and low end is missing in action.

    Lot of cases my gap filling is android emulator running like Autodesk SketchBook Pro and MS Office not wine these days because android emulators are so complete.

    Even with the android phones and everything else Linux Desktop has held is market share. Even OS X is fairly much holding it market share in this so called post desktop world. The party that has lost market share is fairly much Microsoft. Old IBM study said less than 20 of users need specialist software so Microsoft still has a lot more shrinking to-do.

    James sorry to say Ubuntu is not known for the best update process you find a lot of high end software is either debian or redhat enterprise because those two have a track record of doing distribution updates without busting.

    Some of the issues on Linux Desktop have been working out how to do stuff securely. flatpak and snappy are about sorting out application installation issues but those are only possible due to adding different techs to the kernel.

    So some of the delay on Linux Desktop is tech. Due to the poor security state of X11 for a long time developers classed Linux desktop as nothing but a toy to play around with. Serous Linux desktop will start when X11 can be replaced by Wayland.

  • Claude

    In fact, it’s all Linus Torvald’s fault.

    He never should have made Linux free.

    Companies who use Linux for their products, use it because it’s free. It lower their cost. And they can do anything they want with it.

    If Linux would have not be free in the first place, software for Linux would have not be free either. Developpers would have been paid for their work and not relying on fundations to continue.

    If Linux would have cost money to modify it, people would have been able to sell their modified Linux. It could have become an industry. An Open Source industry, but not free.

    Linux will never win the desktop computer because it’s free and nobody can make money out of it. And Linux is not only one, but hundreds of distributions. Too complicated to offer support.

  • James

    Thanks for that insight. I don’t really agree that MS is losing ground, but that’s a different debate.

    Anyway, your points are valid and actually just underpin my problem with the above column. Linux has far bigger drawbacks than marketing. It’s hard to manage, its applications are either threadbare or pretty sophisticated, and its update process (though vastly improved) is still a far cry from the seamlessness of Windows and iOs.

    Interestingly, I see Google is developing a new OS in the hopes of bypassing Linux problems and offering updates more directly.

    So, Linux people need to get out of this fantasy that Linux isn’t ruling because people don’t ‘get it’. Instead, Linux still doesn’t get end users. Ubuntu is by far the closest I have come to a Windows or Max experience – and those define the user expectation benchmark – yet even it is still to technical for the average user.

    (Btw, anyone going on about desktop being dead should try to think a little deeper than Gartner stats. There is a difference between a dying market and a saturated one.)

  • oiaohm

    James there are many formal ground studies in many countries showing that people are more using their phones and stopping using PC as much or even getting rid of it. So I am not looking at Gartner stats on there own. Of course areas where Windows has reduced a lot you find large government systems running Linux so Linux desktop usage in those locations has grown.

    Main reason Google cannot push out software fast with Android is not Linux but vendors with drivers. Google rolling own OS kernel is not going to fix that problem. Google is only experimenting with a OS kernel at this stage.

    Reality is arm before arm64 is pure hell requiring your kernel optimized for the exact soc chip to work. So there are some serous problems hardware side as well that until Arm64 there was no real solution either.

    “its applications are either threadbare or pretty sophisticated”
    Notice what you said. Mid range applications missing in action happen a lot.

  • James

    Please point me to one of those studies. I have read several and I have spent plenty of time on the ground looking at the trend. I live in the mecca of mobile phone adoption and yet I am not witnessing a reduction in PC use. Instead I see people who can’t afford PCs go for mobile phones (thus an expanding user market) and I see declining first hand sales of PCs. But I don’t see a reduction in usage and I have seen an increase in second hand/refurbished markets.

    The only two employee types I have encountered who are exclusively on phones are highly mobile workers (like package couriers) and very high-end executives. Even in that last pool I never see a c-suite only use phones. They love their Macs and hybrids. In the mid tier practically everyone still rely on laptops and desktops. The change is that companies are reusing laptops more instead of buying new ones. I have yet to encounter a company that successfully adopter a mobile-only approach.

    The ‘death of the PC’ narrative is mainly a vendor-driven one, based purely on sales volumes, and one that gets pundits and analysts excited. But that is not the whole story.

  • Flan OBrien

    “Linux advocates should start by analyzing how the people are using their computers.”

    Exactly. Coupled with the inertia we all feel about changing anything.

    95% of home/personal users use a browser a media player, and occasional word processing. Since these are bundled with Windows why would they change? Arguments about security, stability are in the clouds for such users.

    Business users are in general of the same type as described above but in addition may have some non-web legacy app that they run in addition, that Linux has no answer for because it is bespoke. The IT managers are subject to “Never got fired for buying Microsoft”.

    Thus, almost all the market is covered. 5% use Linux (due to curiosity, intelligence?) and the rest use Mac/Windows.

    Change would take at least a generation, provided educational institutions did their job and provided IT managers stiffened their spines.

    Not likely to ever happen.

  • Is “leverage co-opt advertising” some sort of Freudian slip? Is Micro$oft touting Linux on Window$ “co-opt advertising?”

  • oiaohm

    https://blogs.unicef.org/innovation/how-mobile-phones-are-changing-the-developing-world/
    James the biggest differences is when you start going through unicef data on developing countries. Lot of the developing countries desktop computers is coming restricted to government and big business with small business and home users making do with mobile phones mostly based on cost and function. If a mobile phone does everything a business needs why not.

    The developing world companies have the advantage that they are starting out mobile phone and tablet only without legacy stuff to convert. So yes there has been death of the PC in a big way in developing countries.

    “But I don’t see a reduction in usage and I have seen an increase in second hand/refurbished markets.”
    Cost cutting we have seen this in developing world countries before the PC market in those countries started disappearing backwards when supply dried up. Something to remember second hand/refurbished have less life remaining than new but this causes a knock on effect developed countries with big pools of existing PC are not exporting due to in fact reusing them so not as many old pcs are going to developing countries so now the developing countries second hand computer market drys up.

    Death of the PC is happening at different rates in different countries some more forced than others. Something else to be aware of most of the developed world with reducing in PC sales is heading to the point not as many new machines entering the countries pool of PCs as PC dieing due to old age or metal recycling.

    So what is going on here is vendors and death of PC to know that without the second hand computers going to other countries it going to get hard to grow new markets and existing market size will have to reduce. All the cuts appear to be coming out of Windows PC by web survey numbers. As you even said c-suite and the like love their Macs. We are not seeing Mac or Linux users give up their PCs first. Mac and Linux users are kind the ones that it looks like you have to pry there desktop out there cold dead hands they are not letting go of it. Also remember Linux users commonly use hardware to hardware fail all the time.

    Timing of google new kernel/os attempt lines up with the tp-link vs FCC problem where it looked possible FCC might out law open source in radio using devices but in fact FCC has ruled the exact other way that tp-link must provide open source option just fix up the radio issue. We will have to wait to see if that was just a back up plan that turned out no longer to be required or if it something more serous.

  • Jedinovice

    Well, I did my bit! Here is a link to my demo of Mint Linux KDE.
    https://www.youtube.com/watch?v=Q8hDYiGprWs

    I created it as a means of clearing away the pervasive myths regarding Linux – it’s so hard to use, you can’t use the internet with it, you can only install by compiling from source, etc, etc.

    It won’t bring converts on its own; it’s REALLY hard to break through people’s comfort zones, but it should at least get rid of the misinformation.

    Spreading the link around can’t hurt, right?

  • Andy Mitchell

    The main reason Linux hasn’t conquered the desktop market is simple. Windows comes preinstalled on nearly every new PC sold. This was the marketing master stroke by Bill Gates. The average person doesn’t want to, or is unable to install an OS. They’re happy to use the one that comes with the machine. After all, they’ve paid for it. Windows has a cash value, in the collective mind of a consumer based economy, it must therefore be better than something that has little monetary value. To many, the Linux desktop is a poor man’s alternative to the “real thing”. The Linux distros that try to parody Windows, don’t help Linux’s credibility with the consumer either. When a new user finds the similarities are only skin deep, they become disinterested with Linux.

    Also, from reading the comments. It’s obvious that many people don’t understand how to setup and administer their Linux systems. They’re trying to use Windows methods and techniques on their Linux desktop. They’re effectively making the simplest of tasks over complicate and difficult for themselves. For example installing user software, drivers and adapting to a more efficient and logical ways of administrating their systems. This is another area where Windows wins. It removes the necessity for the average user to understand the OS. The fact that this approach is fraught with dangers is irrelevant. An OS that’s perceived to be idiot proof will always win.

    Linux will never supplant Windows on the desktop. It really doesn’t matter either. The fact that the Linux desktop hasn’t monetary is it’s strength. It not in a sales competition with its proprietary counterparts. It’s not judged by its profitability. The Linux desktop will remain superior and it will be developed for the foreseeable future. It will always be available to people who refuse to be sucked into the unethical proprietary OS world. That’s all that really matters.

  • joncr

    Many people don’t know that companies like Red Hat and Suse began by selling shrinkwrapped boxes of desktop Linux on retail shelves. That wasn’t sustainable because FOSS licensing requires making source available, which means someone will always rebuild your source and serve it up gratis.

    In other words, desktop Linux isn’t even in the market — can’t be — due to ideological licensing restrictions.

    There’s a long list of other things that act as drags on Linux acceptance, including “better than you” ideological stances by FOSS zealots who label anything and anyone who wants to make a profit as “unethical”.

  • Spencer Kelly

    Simple answer – MS-Office
    Most businesses I’ve come across run on MS-Office and it’s just assumed that you have MS-Office to handle any documents they send you. Here’s a hint – they won’t be in open document format.

  • Qben

    It isn’t supposed to happen. Not for that one reason and not for that other one reason, but for all of the reasons displayed by commentros above. Some DO miss programs that do special things where eqivalents aren’t there for Linux (or are not as good, or …). Some DO have hardware and driver issues under Linux that they do not have under other OSs. Yet others DO find distributions, repositories, software management complicated under Linux. And others DO find the plentitude of GUI confusing. Other DO need an a§§ to kick if something does not work as expected.

    Etc.

    There ARE operating systems that are (at least) good enough in all of the single areas listed above. And “good enough” is what satisfies the majority of people.

    So whether or not there is a buck to earn with Linux does not matter. It also does not matter that it is free (of charge, or in some cases free as in free). The average consumer is not interested in free code and does not care about the costs of the Windows license (in his mind, he buys a “computer” anyways not an operating system). Those why DO care (i.e. a slighly growing minority) use LInux as Desktop OS. And no marketing ever will change this.

  • joncr

    >>” Those why DO care (i.e. a slighly growing minority) use LInux…”

    No evidence supports this. It amounts to an ad hominem slam on Windows and OX S users.

    FOSS licensing is the poison pill that keeps Linux desktop out of the market. Not failing in the market. OUT of the market.

  • Mike

    @James

    > “I don’t fiddle with the OS – these were the result of simply trying to install different drivers, applications or whatever advice I followed to try and fix a problem. Consequently I ended up with a rather stunted and ill-performing OS. I tried to fix these using online advice and terminal commands – and succeeded in breaking the OS.”

    I’ve seen that exact same scenario a thousand times – on Windows. I find it hilarious that people here talking about the ‘seamlessness’ of Windows. Windows breaks for any and all reasons…often for no reason at all. If rebooting doesn’t fix it, people typically take it to some godawful place like Best Buy to get it fixed.

    Enough with the bullshit.

    Linux is easier to repair than either Windows or OSX. This is an empirical fact. That the average non-technical person can run into trouble by randomly following online advice (often for the wrong distro) is not an indictment of the OS, but rather of the fact they couldn’t fix those other operating systems either.

  • antifanboy

    “So why hasn’t that happened? One word: marketing.”

    You’re wrong. It’s actually a combination of user unfriendliness(having to fumble around in the command line terminal in order to accomplish simple task that would take a few easy mouse clicks in Windows or OSX), poor support(much fewer software applications and games, as well as having to hunt down drivers), and the mentality of the FOSS community itself(cult-like behaviour, zealotry, fanaticism, fanboyism, viewing computing as a political act, etc. http://hugequestions.com/Eric/Linux/Linux4.html). No amount of marketing will ever fix this.

  • Mike

    @antifanboy

    Hilarious, everything in your statement is either provably false, or just…ironically…fanaticism with no basis in fact.

  • As many people already wrote here, the problem with Linux is Applications. Most Linux OSes (Debian, Ubuntu, RHEL and OpenSuSE) do an amazing job stitching a really good operating system from a huge selection of open source applications. Most of these applications are dedicated for servers, and only a small portion towards Desktop.
    As such, there is a cyclic problem of chicken and egg type. There not many users, hence not many desktop applications. The ecosystem is very small, and very few companies develop software for the desktop user that they can make money out of.
    The exception is software developed for the Developer Audience, for example Slack had from the beginning a Linux desktop client. Another example I can think of is PyCharm, IntelliJ family of editors. Most vendors simply don’t see how they can cash out Desktop application for linux.
    The 5% or so of people using Linux are ideologists, people who don’t need much more then a browser, or people who use it as a product (like XMBC media station) or developers. The common for them all is that Linux delivers what they need, or that they are willing to accept these short comings of Linux.
    I am willing to admit I am one of them. GNU\Linux as a Desktop OS still sucks, no matter how much marketing you will put here. Chromebooks don’t even get close to Windows, regarding the amount of software available and what you can do with them. Never the less, I think the situation is steadily improving, and not because of marketing, rather because the more developers use Linux, the more applications are available and some of it will eventually drip down to the normal consumer market. It might take 10 more years, but we will get there!

  • rclark

    Marketing is only a part of the problem…

    Drivers (printer drivers especially) I feel are the Achilles heal of Linux. For example, I just upgraded from Mint 17 to 18 on one of my desktops. It ‘still’ did not have drivers (out of the box) for the printers I own. Had to go to the Brother website and track them down and then took a bit to get both printers up and running.

    That said, for our use, LibreOffice does all the needs doing. Reads and writes Windows docs and spreadsheets just fine when needed and export to PDF out of the box. VLC handles video, Firefox the Web. Thunderbird the mail. For our home desktop use, Linux does it all… Without having to be rebooted for ‘updates’ every week (which aggravates me to no end at work). My linux server on CentOs handles the data server side of things. Just sits there and ‘works’. Linux runs on my Laptops, and Raspberry Pis too. Just no need for anything else for home use.

    Other than the driver issue, I never found Linux ‘unfriendly’… But then I started with DOS way back when before OS’s started to be dumbed down.

  • Andy Mitchell

    @ antifanboy:- I disagree with your synopsis. For the most part, the community is very welcoming and helpful. I grant you that some factions are somewhat elitist. These are very much in the minority. Take the Peppermint Community Forum as a prime example of a friendy and a welcoming attitude: http://forum.peppermintos.com/. It’s not necessary to fumble around in the command line. The majority of distro have advanced GUI options, just like in Windows. The command line is just a quicker and more definitive way of completing a task. Even Windows are now using, bash. It’s no longer the 1990s where every distro was dependent on command line proficiency. Most of the many Linux desktops are easier to use than Windows. Of course many people base their dislike and critism of Linux after 10 minutes of use, compared to their 10 years plus of experience with Windows. Those who spend the time to familiarize themselves to their Linux distro will agree, Linux is easier. Support is better than what’s available with Windows, namely the forums. These are far more than chat rooms. They’re huge knowledge bases and in many cases, provide fantastic technical support in easy to follow steps. They’re free to use too. Driver support is much better out of the box than Windows. That’s common knowledge. In all of the four years of using Linux, I’ve never had to install a driver. Everything works out of the box. I agree with you that some factions of the FOSS community are inflexible to newer ideas. This again, is mostly the minority. The link you’ve provided is a prime example of the pomposity and abject ignorance of many Open source critics. They make a lot of noise and can substantiate little to non of it. To me the, author appears to have the majority of issues, rather than the subject of their venom.

  • Dion Dennis

    Not addressed in all the many replies is what seems like the presumed conflation of “The World” with the usage habits of North Americans. A bit wider glance around “the World” reveals, for example, that Linux, and FOSS software is under wider adoption, in Europe, Asia and Africa. (Tux Machines links to announcements that, for instance, the French, Italian or Lithuanian police or armed forces have moved to LibreOffice, which is one step in the direction of Linux, given the FLOSS nature of LO).

    And some cities (Munich) and countries (Iceland, Macedonia) and other institutions, such as the French National Police or schools, for example, in India, have moved, in large numbers, to Linux. It’s not hard to infer that once those institutional footholds are gained, individual desktops follow.

    As far as I can recall, both the Russians and Chinese are moving toward a Microsoft-Independent world, although this has little to do with a love of FOSS, and more about creating their own intranets of tracking and control, free of the trade-drain and telemetry. And, of course, it would be wrong to underestimate “The Snowden Effect” in accelerating that move, as well, in non-authoritarian political systems, as well.

    In any case, perhaps little less American-centric vision is required to see that “the World” has a different configuration of computer use than the United States and its cultural colonies.

  • Mike

    @Dion Dennis

    Well said.

  • franklinr

    As a computer user since 1983 with CP/M OS I have used every version of the PC world OSs except Mac. But I am a dual user with Windows 7 on one desktop PC and Linux Mint 17 on another. I write books about Moving Over to Linux and teach a regular class at a local Senior Center in Windows. In my opinion, the problems are:

    1. People over 50 are seen as non-entities, who can’t possibly understand computer use when in truth millions of these people use their PCs daily to do many tasks and are using Windows XP, 7, or 8 but won’t go to version 10. Those who write articles on the internet have parents and grandparents who do this. Not just in North America but in UK, France, Sweden, Brazil, Phillippines, India, etc. Their number is legion.

    2. Mac developers continue to promote the idea that Linux is the “Evil Empire.”

    3. Many Windows PC users are afraid of Linux because it has so many strange-sounding words in it.

    4. Most Linux devs like to say that Linux is for US and nobody else. If you can’t use the “sudo apt-get” command properly you have no business on Linux.

    5. The myth that ubuntu is the only distro available when several higher level distros exist that are very GUI based and work much like Windows including all the necessary drivers. My Linux Mint 17 runs on ubuntu 17.1 but I never see it.

    6. Most civil governments, city, county, state, have not been introduced to Linux for the above reasons. They need to be shown that their hardware/software budget would drop to nearly zero if they made the switch. The federal government is starting to do this.

    7. Hardware manufacturers don’t offer a computer with Linux pre-installed because they would lose their Windows license. They can’t have it both ways. There is a need for a maker to ignore MicroSoft and offer a Linux computer line.

    8. Most IT professionals are not taught Linux in school, they must learn it on their own.

    It would be good if we could follow the UK model of introducing the Raspberry Pi into schools for the students to learn computer use and programming in Raspbian, a Linux distro.

  • I’m seeing a lot of arguments against Linux as a fit for end users because of its poor support for Wacom tablets, CAD programs, games, etc. I think perhaps there is a misunderstanding here about what a typical end user *is*.

    A typical end user needs a browser and an office suite. Comments about the compatibility issues between MS Office and LibreOffice are fair; that’s an area in which typical end users may well be locked into the technology they’re currently using. But Wacom tablets, CAD programs, even games — those are for specialists, just as surely as the command line is.

    If you’re a specialist who needs Windows, or a specialist who needs MacOS, that’s understandable. But don’t assume that Linux isn’t ready for general users because it doesn’t meet *your* needs.

    @Claude: Not to go all “Actually, it should be called GNU/Linux” on you, but it sounds like you’re not really aware of the history of the operating system you’re talking about. Linux Torvalds did not create the operating system; he created a kernel that was compatible with the OS components created by the GNU Project, and released it under the GNU Project’s free software license. He’s not the guy who decided the OS should be free, he’s the guy who made it work as a complete system. That’s a major contribution, and we wouldn’t be having this conversation if it weren’t for him. But the OS that most of us just call “Linux” actually predates the Linux kernel by several years, and its free software ethos isn’t some kind of afterthought, it’s the central purpose that the whole thing was built on.

    (And yes, to some extent I’m conflating free-as-in-cost with free-as-in-freedom here; I’ll get to the difference in a moment.)

    You say that if Torvalds hadn’t made the kernel free, it would have seen wider adoption. I think that statement fundamentally misunderstands the reason the kernel has been as successful as it has.

    Because it’s not simply free-as-in-cost, it’s also free-as-in-freedom. Anyone can access and modify the source code of the Linux kernel and the OS that uses it. Not only that, but it’s copyleft — which means that when IBM or Intel or Red Hat or Canonical or Google makes modifications to the Linux kernel, they have to share them with everybody else.

    And that’s why the Linux kernel is as good as it is: *because* it’s had so many contributions from so many people and companies over the past 25 years or so.

    It’s not a bug, it’s a feature.

    If Linux weren’t free, we wouldn’t be having this conversation, because it would never have become as mature, sophisticated, and complex as it has. Even if it had gotten investors and a good development team behind it, it couldn’t have had the kind of success that it has, because it wouldn’t have had the global network of developers that it does.

    And, not for nothin’, if its cost were similar to MS’s and Novell’s, then it would never have been a viable competitor to either one of them. You’ve seen what happened to other contenders in the server space: Novell gave up on Netware and switched to Linux, Apple quit selling server hardware. There are still other options in that space; Sun continues to chug along despite Oracle’s best efforts. But it’s mostly either Microsoft or Linux now, and Linux got to where it is by being free.

    (Speaking of Apple, it’s had quite a bit of success forking other free software projects, such as FreeBSD and KHTML.)

    And then there’s Android, which uses the Linux kernel but not the GNU OS components. How do you think it became the most popular OS on the planet? Because Google gave it away for free. Apple wasn’t opening up its OS to third-party hardware vendors; third-party hardware vendors wanted to compete with Apple; Google provided a working OS so that they wouldn’t have to develop their own in-house.

    If Linux weren’t free, it wouldn’t currently be running on over two billion phones and tablets.

    No, it’s never managed to make major inroads on the desktop. Just servers, mobile devices, embedded systems, the IoT…

  • Typo: “Linux Torvalds” should, obviously, be “Linus Torvalds”.

  • Bill

    Follow the money. Nobody buys Linux pre-installed. Everyone buys Windows machines, giving money to Microsoft and nothing to Linux manufacturers. Even hard-core Linux advocates buy Windows or Apple machines and then install Linux on it. There’s no money for Linux.

  • Bill

    @rclark

    Perfect example. rclark buys Windows machines and peripherals, and expects them to work perfectly with Linux. If you want Linux to work, buy Linux machines and peripherals.

  • archuser

    I had to use Webex to connect to my training and it is not supported on Linux , though i was able to install it, it could not hear voice , installation was complicated.

  • antifanboy

    @Andy Mitchell well explain to me why the last time I tried to use Linux on my desktop(six months ago) I have to spend four days hunting down drivers for my wacom tablet, only to find out I have to do a lot work in the command line terminal just to install it and several dependencies to get it to work. It halfway worked, but when I asked to make the tablet fully work, I got a big RTFM instead. Oh, and don’t get me on about the cult-like community who are willing to drink stallman’s KoolAid at the drop of a hat…

  • rclark

    “… rclark buys Windows machines and peripherals…” and “Even hard-core Linux advocates buy Windows or Apple machines and then install Linux on it”

    Not precisely true. My two current Linux Desktops were built from the ground up by me (motherboard, PS, memory, etc.). No Windoze OS installed. True, I didn’t check compatibility with Linux when I bought the ‘pieces’ (one AMD system, the other Intel system) as I expected it to all work with Linux — which they did. Installed Linux Mint on one and LUbuntu on the other and we were cooking. The Brother printers (one Laser, the other Inkjet) were bought because we were dis-satisfied with HP and Epson products over the years, so tried something different. Just no Linux drivers on supplied CDs… FWIW, the ONLY issue was the now resolved printer driver issue. No other driver type issues were encountered — a good thing (Linux has come a long ways)! However printers are ‘usually’ a major part of a home system and I can see where a newbie my be intimidated with the install (say CUPS install with //localhost:631 or use SAMBA to link to a printer on a Windows Box).
    Now the laptops were hand-me-downs where I did overwrite installed Windows with Linux. Again no issues. Easy and straight forward, everything worked including WiFi out of the box. Even added SSDs in them to make them much faster.

    Very satisfied with my Windows free world (at home that is!).

  • Purple Library Guy

    Some interesting comments here, and quite a bit of nonsense.
    Just for the record, I am an “average” user: I browse, do some word processing and the occasional spreadsheet, play some games but not intensively, have some peripherals but no flippin’ Wacom tablets or whatnot. I am not very comfortable with the command line. And ya know what? I find Linux easier to maintain than Windows.

    Partly because I use Mint, maybe. I can’t believe people were talking about updates being hard, for instance. Wha? On my screen there’s a little applet on my taskbar. When the system is up to date, it shows a little green checkmark. When it isn’t, it shows a little red X. Click the X and it says I’ve got updates do I want to install them; by default it has all the updates checked and I just say yes and it installs them and says OK you can close the updater now it’s done, and then there’s a little green check mark again. Note that it does not take over my computer and ram the updates down my throat, making the whole thing unusable while it’s doing this. Note that it lets me vet the updates if I’m the kind of guy who knows enough to do something like that, but I don’t have to. I just read a book by a comedian guy who has probably never heard of Linux, mostly about a completely different topic; somewhere in the middle he starts talking about how his computer has this software called, he believes, Microsoft Gestapo, which periodically lines the system up against the wall and tells him to update or else, then locks him out of his computer for ages at a time. He goes on in this vein; it is not only Linuxheads who get annoyed by the way Windows updating works.

    Thing is, I’m not trying to do Xtreme stuff. So I use the drivers it installs automatically and I install my software from the software centre by clicking on it (or I install it from Steam if it’s a game, just like anyone else). I haven’t messed around with a package that’s not in the repository in years. If you don’t try to do stupid Windows stuff like download and install random programs from the internet, Linux is way easy to maintain.

  • Purple Library Guy

    Meanwhile, on the “why it hasn’t eaten Windows”–Some people are grasping at explanations centred in the nature of Linux. Like supposedly it didn’t succeed because it was open source, or because the community has a bad attitude, or because (drivers, software etc) are lacking.

    Most of this is nonsense. Linux would have just died if it weren’t open source, and probably would have quietly stagnated if it weren’t copylefted. That is, it would have died like BeOS, OS/2, Amiga and so on, all closed source operating systems once liked by many. Or if it was open source but with a permissive license, it would have stagnated like BSD. Back then most open source people didn’t trust non-copyleft licenses, because the bandwagon effect that GPL software eventually created hadn’t given open source the momentum that nowadays seems to protect permissive software against closure attempts.

    The simple fact is that taking desktop market share is very difficult. Marketing and money are certainly a big part of it. We live in a society ruled by money and dominated by marketing, and it just amazes me when I see people acting as if that stuff didn’t count as an explanation for things. It is also true that some things, like software ecosystem and driver availability, are less good than they could be–but again, this isn’t because Linux is any particular way, and certainly not because Linux is open source. To the contrary, things would be far worse for a hypothetical closed source Linux. It is because any OS which is not dominant will have fewer third-party things, like drivers and software, than a dominant one. Duh. Even Macs don’t really have the software and peripherals ecosystem that the Windows PC does. The amazing thing is that Linux desktop software is even remotely comparable, given such a small market share.

    And yet it is possible for an OS to gain market share. MacOS did; it didn’t become dominant, but it increased its percentage significantly from where it had been. It did so largely on the strength of massive, very well done marketing campaigns. Mac computers did and do have strengths, but without a marketing campaign branding and defining those strengths to the audience, Mac would probably have gone down the drain. NeXT did, and yet the Mac that succeeded under Steve Jobs was basically NeXT, the next generation.

    Linux probably can’t succeed in the same way as Apple. Linux has its own strengths, and will have to succeed in ways that use them. One avenue is probably specialization. Rather than try to be a size that fits all, one strength of Linux is that it can be so many things to many different people. For instance, someone should start a project to talk to all the different groups around the world who are trying to preserve a small minority language. Inuit and other first nations types, small African ethnic groups, Basques and so on and so forth. Big software makers aren’t interested in any of them individually as a market, but there are lots of these little groups and many of them are working hard to keep their languages. With Linux and open source software, if the big boys don’t want to make a Haida or Tsimshian translation of the OS or the software, they can do it themselves. There are masses of groups with specialized needs that mass marketing has no interest in filling, but that the “scratch an itch”, “adapt it to your purposes” ethos of Free Software could help. Little groups, insignificant groups, but add them all together and you’re starting to talk significant numbers. Just one example of where Linux could make inroads.

  • tracyanne

    >>>In any case, perhaps little less American-centric vision is required to see that “the World” has a different configuration of computer use than the United States and its cultural colonies.<<<

    Yes it's interesting that "The World" to American citizens, even those whom I otherwise respect, is actually the US, and sometimes as an after thought Canada and Mexico.

  • Andy Mitchell

    @ antifanboy:- I’ve just tried my wife’s Wacom Intuos tablet on two of my installations. I’ve never used this peripheral before. I’m dual booting on a Dell Latitude e6410 Core-i5 560m 2.66Ghz CPU. Partition number one is running Peppermint 7, (Ubuntu 16.04 based). It’s using kernel 4.4.0-34. The tablet works perfectly out of the box. Partition number two is running SparkyLinux,(Debian Testing “Stretch” based). It’s using kernel 4.6.0-1. The tablet also works perfectly out of the box. I have a second machine running Manjaro 16.06.1 that I haven’t tested. It’ll work though, as it’s also using kernel 4.6.*. You didn’t mention which distro you were testing.

    If you received a RTFM reply, your using the wrong distro for a newbie. Move on to a new user friendly distro. I personally hate the RTFM replies. They’re not helpful and do little to attract new users. Four days is excessive amount of time to find a driver though. If the driver installation does require the use of the terminal, it’ll be cut and paste for the most part. So it’s no big deal.

    I do agree that the Stallman purists are basically cutting off their noses to spite their faces. I go with the philosophy, if I can use Open source, I will. If there’s closed source software that does a great job, or there’s no Open source alternative, I’ll use that. Teamviewer and Intel microcode are great examples of closed source software that are well supported and work really well. I’m not going to boycott them for some unattainable utopian dream. For the average user, there really isn’t a problem using a Linux desktop. They’re easy to use, well supported and the forums are for the most part, inviting and helpful. It does require the user to adapt to the new environment. This is normal for everything that’s new though.

  • ProperitoryOS

    Not due to “quality”???? I’m sorry it is DUE to QUALITY. Linux is a fragmented HACK nothing more nothing less. The desktop remains slow with an extremely dated, overly complex stack (X.org) and a solution (wyaland and yet another fork) is still years away. Most of the desktop is a blatant copy of OSX (read: Ubuntu). Libre Office remains stuck in a 2003-look-and-feel era, much like the Linux desktop.

  • Askfor

    When one installs and starts to use what is called “modern Linux distro”, the desktop cries out: “Look at me, I am am very much like Windows”.

    Reminds me of those poor people of west Africa which were mistreated be Belgian colonial forces under King Leopold II. Common punishment for rebellion was chopping ones hand off. Today Belgians are long gone, but locals are still chopping each others hands when at war with each other.

    Recently I installed Slackware 14.2. It is a simple straightforward distribution. Requires general understanding of UNIX concepts, but it definitely has identity, a personality of its own. It is clearly different from Windows and does not try to imitate Mac. No, it is not fit for casual computer user, but it could be a starting point for something different and new. A brand of its own.

    There is no need to imitate Windows, Wine is probably good enough for those who need Windows compatibility. Linux needs to develop identity of its own. Hiding the underlying complexity is the way chosen by Miscrosoft and Apple, and I think it is wrong. Take K3b as an example. It is not trying to push the complexity under the carpet, but offers shortcuts for most common tasks. If one needs something special, he or she still needs to learn something about optical media burning. However, 90% of the needs are handled with just a few mouse clicks.

  • There’s a simple answer to every question; here’s this one:

    Linux hasn’t taken over the desktop because the desktop is going away.

    Accept it; get over it.

  • tracyanne

    Two examples of successful consumer Linux products.

    Android Phones and Tablets are so successful, because… Marketing.

    Chromebooks are successful, because… Marketing.

    Linux as a consumer Desktop or Laptop product is otherwise almost non existent, and where it does exist it is not marketed widely.

    The only reason Linux hasn’t taken over the Desktop is Marketing. Where devices featuring Linux are marketed well they are successful.

  • Purple Library Guy

    jimmike, I’m really not interested in trying to play Civilization V on a phone.
    All the stuff about how the desktop is going to disappear, whether the replacements do desktop things any good or not, is moronic.

  • rclark

    I sure don’t see the desktop computer going away. Who wants to type anything of any length (or write applications, or do spreadsheets, presentations, database work, play games, or cad, or, well, the list goes on…) on a phone or tablet, or even a laptop? No thank you. Bigger the monitor(s) the better, and local private storage (cloud computing isn’t for me). A good graphics card, and a nice Natural keyboard to boot. As for Linux, I like having the choice of how the desktop manager looks and feels. Cinnamon, KDE, Gnome, LXDE, XFCE, etc. Choices. Not a desktop that’s rammed down your throat as they know what’s best for you…. I like the choice of which flavor of Linux to use. I like LTS CentOS for my server, Mint and LUbuntu for my desktops. Have several others loaded in VMs to kick their wheels so to speak as well. Just can’t beat it.

    Back to lack of marketing … I can see that. With the ‘now’ generation, who today wants to spend time loading an OS? You don’t buy a blank smartphone, go home, load the OS, and check it out do you? I think that is a good analogy of the problem with Linux on the desktop and in general for the masses. Most people just want it ‘now’, what it looks like at the store, what does the salesman say it can do, and then buy if meets their needs. M$ has that market cornered. Heaven forbid you might have to spend some time researching looking for a driver…

  • Askfor

    Whoever plays First Person Shooter games on phone, I am calling him a masochist…. Some people do, indeed, but I wouldn’t want to be one of them.

  • oiaohm

    rclark first mistake phone equal being stuck with only a small screen.
    http://www.pcadvisor.co.uk/how-to/google-android/how-connect-android-tv-summary-3533870/
    Reality android phone stack of connectors with a HDMI screen and bluetooth keyboard and mouse. You end up with something very PC like.

    Askfor playing quake 3 on a android phone with HDMI screen and mouse connected it not too bad at all. So apparently I am a masochist. Wait I guess you were also forgetting you can add larger screen and keyboard and mouse to a phone. Its one of those things you do when you have gear setup for a presentation and you find out due to key person being delayed you have over 2 hours to waste. Yes expecting modern GPU heavy First Person Shooter to work would be pushing bolder up hill then wondering why you end up dead under it.

    Issue with phones is limited GPU and limited in device storage and limited cpu. But compared to what we used when XP was released current day phones are huge and people still get a lot of things done with old XP machines.

  • Mike

    The shift to mobile today parallels the shift to microcomputers (PC’s) back in the day.

    Enterprises with big mainframes had teams of people who suddenly had to contend with micros because people were using them outside work. Micros were considered toys and not fit for ‘serious work’. Sound familiar? PC’s rapidly increased in capability until they replaced big iron in all but a tiny niche of uses. Desktops are already dead to most consumers…except for an ever shrinking niche of uses. As tiny computers (phones, IoT devices, what have you) increase in capability, they will steal more and more of what’s left. The practical upshot of this is whoever controls mobile will control computing for the next couple decades. That almost certainly will be a Linux powered OS. Windows and OS X are already dinosaurs and don’t know it.

  • NonSequiTourDeForce

    The main problem in large enterprises is Excel and stupidity.
    Until LibreOffice will offer a bug-compatible version of VBA so that the enormous amount of macros in existence will be migrated seamlessly, Linux will not take over the corporate desktop.

    As for stupidity, a lot of places have jumped into the “cloud” bandwagon, without realizing that all those services (including M$ ones) are hosted Linux machines.

    I also worked at Linux-only shops where hardened Windows users were simply told to stop whining and get on using the Linux desktop.

  • oiaohm

    NonSequiTourDeForce serous-ally you don’t want bug for bug compatibility with MS Office VBA.
    http://www.spreadsheet1.com/excel-vba-bugs.html
    And that list here is the tip of the ice berg of Excel VBA bugs that do nothing useful other than driving you nuts why Excel has now done something stupid to my data or crashed. You want compatibility with VBA macros where MS Office is not completely brain dead.

    Personally I would tolerate a converter asking questions do you want to use like excel incorrect leap years and so on in the converted macro. Yes there are a lot of cases were excel is in fact doing maths wrong due to some historic excel hack to deal with some foolish design bug.

  • The best candidate for a popular Linux desktop will be the merger of Chromebooks and Android. Chromebooks bring the desktop form factor for serious work, and Android brings a huge-and-growing base of applications and avid users.

    Both already have extensive marketing and dedicated hardware free of aftermarket driver issues.

  • @Purple Library Guy–
    No one WANTS to support, or cares about, people whose idea of computing is to play games. The idea that they do is moronic. For YOU and your ilk, the desktop is going away. Forever. Back to your basement and wait for it.
    Accept it.
    Get over it.

  • Purple Library Guy

    I expect you can do all that–hook up a big screen and a full keyboard and peripherals to a phone. At which point you have . . . a desktop.
    jimmike on the other hand turns out to be just a troll. Back to my basement? Back under your bridge!

  • Nathan

    What Linux needs is a big software company (like Corel) to partner with a hardware company (someone who is not in a license agreement with MS) and dedicate themselves to Linux. They must provide an all in one solution similar to Apple. They must provide a complete package. Including hardware, OS, and complete software solutions. Corel would need to make a Linux compatible version of ALL their software. Then they can sell the hardware with Linux preinstalled for a small initial profit on the hardware. Then sell proprietary software for a larger profit. To be successful, they must build/market their own Linux distro to be sure they have all the drivers for the hardware they support/sell. It needs to function smoothly. Then they must have a complete app store. This can include free open source as well as proprietary closed source applications.

    Obviously this would take a lot of money and resources. The majority of the profit would likely end up being on selling the proprietary software. It would have to be modeled similar to Apple as a hardware/software company. I think Corel was heading in this direction in the 90’s before they ran into financial trouble and MS purchased a part of them and forced them to sell off the Corel Linux OS and they ceased development of their Linux software. I still think this plan could work. But they really need to be dedicated to a full Linux suite of software.

  • CFWhitman

    It does indeed look as though James is using Linux as though it were Windows.

    Generally, with an Ubuntu based system, with the exception of printers and scanners, any drivers that you need should either already be present on the system, or offered in the “Additional Drivers” control panel. You certainly shouldn’t have to go find graphics drivers on line.

    For printers and scanners, it’s good to check compatibility beforehand. These devices don’t use conventional specific drivers (that is the only actual drivers involved are not device specific, but generic ways of connecting to a class of device), but communication filters for the software that runs them. A lot of filters are included, but not all.

    Also, you’re expected to install most of your software from the distribution repository. Judicious use of third party repositories for additional software or newer versions of a few programs does not generally create any problems during upgrades. Extensive use of third party repositories for getting newer versions of many applications that are included in the repositories probably indicates that you should switch to a rolling release distribution because you apparently want to live on the bleeding edge.

    If you use Linux as intended, it is much, much easier to maintain and upgrade than Windows. Windows “upgrades” tend to leave your system a mess with a bunch of extra hard drive space used.

    I never upgrade my own installs of Windows because it doesn’t really upgrade. It installs a new version of the system in a new folder, tries to move over what it can (the registry making this effort sketchy at best, though msi packages have at least made moving software possible; they didn’t used to even try), and then leaves what’s left in a renamed folder. The Windows 10 upgrade seems generally the most successful at actually moving your stuff over, but it’s still messy.

  • CFWhitman

    My take on ‘desktop Linux taking over the world’ is that it’s basically a comparison between things that aren’t in competition with each other.

    Proprietary anything (not just software and hardware) exists only as a bubble around the sum total of human knowledge. After a time it gets replaced by non-proprietary open technology as that becomes part of general knowledge, and proprietary stuff moves on to the next frontier. The desktop is headed in that direction too; it’s just a very slow process. That doesn’t mean the desktop will necessarily be dominated by Linux, but it will eventually be based completely on open technology.

    The proprietary technologies aren’t really in competition with the open ones. They inevitably get replaced by them as time goes on.

  • Go easy on jimmike, the poor thing has to type all his comments with his thumbs. 🙁

  • franklinr

    Mike, you are right when you say “Enterprises with big mainframes had teams of people who suddenly had to contend with micros because people were using them outside work.”

    Back in the late 1970’s I was working at HP in San Diego and they had their mainframes there for all their work. But a couple of engineers brought in a Commodore PET to run some operations. It was called “the toy.”

  • Mike

    @franklinr

    Yeah. Nobody took micros seriously.

    We hear the same tired refrains today: “That’s just a toy. You can’t use it for serious work.”

    People in IT who dismiss mobile as not capable of replacing desktops are going to have their lunch eaten when they aren’t looking. 🙂

    Linux rules mobile and Linux rules IoT. Sooner than you think that’s all that will matter unless you want to be the 21st century equivalent of an AS400 operator (i.e. Windows server admin or desktop technician).

  • @ProperitoryOS: “Libre Office remains stuck in a 2003-look-and-feel era”

    Thank Christ.

    The next person I meet who thinks the ribbon interface made MS Office better will be the first. And probably a Microsoft employee.

    @TracyAnne: “Android Phones and Tablets are so successful, because… Marketing.”

    Not exactly.

    Samsung, LG, et al are successful because of marketing. Android is successful because it was the best option available for Samsung, LG, et al to hitch their wagons to.

    “Chromebooks are successful, because… Marketing.”

    I would say they’re successful because of their price, weight, and battery life, but marketing doesn’t hurt.

    “The only reason Linux hasn’t taken over the Desktop is Marketing.”

    Well, no, not the *only* reason; decades of habit and vendor lock-in are part of the issue too.

    @Purple Library Guy @Askfor: Gamers are a niche audience who have a tendency to vastly overinflate how important they (we) are to the PC market. The number of people who play even major hits like Call of Duty on PC are dwarfed by the number who play League of Legends on their phones. I don’t believe the desktop market is going away because there are still major day-to-day tasks where a full keyboard is vastly more efficient than an onscreen one, but that’s got nothing to do with Civ 5.

    It sure is *nice* to be able to play Civ 5 on Linux, though.

    @jimmike: Oh snap, a joke about people living in basements. The first time you wrote that sick burn, was it with a hammer and chisel?

  • Correction: LoL isn’t a phone game, I was thinking of the similarly-alliterative Clash of Clans.

  • CFWhitman

    @Mike
    Well, there was a difference when talking about microcomputers next to mainframes. The people who said they were toys were talking about computing power because the interfaces were essentially the same.

    With current computers compared to mobile devices, there are places where mobile is an advantage, and in those tablets and phones are already dominating. However, there are certain places where a monitor/mouse/keyboard is still a big advantage. Of course that doesn’t mean that mobile devices can’t begin to take over those places, but they will have to be hooked up to a monitor/mouse/keyboard in order to do so. Of course, they also have to become powerful enough to do the tasks, but that’s certainly plausible, even if they’re not all the way there yet.

    Remember, though, that mainframes haven’t departed, they’re just a lot more niche than they used to be (and some are referred to as “supercomputers”). Desktops will likely be around for a while yet.

  • CFWhitman

    I’ve heard a lot of people talk about the importance of video games to platform popularity. In real life, however, I’ve always seen video games follow platform popularity, never lead it (of course dedicated gaming hardware is different).

    In reality, the amount and quality of video games available serve as a measurement of how popular your platform has already become rather than helping it grow.

  • Purple Library Guy

    Just for the record, I mentioned a game simply because it was one obvious and common use case. I’m not a heavy gamer, nor do I think gaming is the core of computing. Although gaming is certainly far from marginal, and I’ve seen masses of comments over the years saying “I might shift to Linux except I can’t play my games on it”.

    CFWhitman, as to your last comment on games following vs. leading . . . well, the current case of Linux seems to be a counterexample. Linux has not gained noticeable desktop market share lately, and yet over the last few years game availability has skyrocketed; percentage of PC games available natively on Linux has gone from 0%plus rounding error to around 20%. This level seems to have stabilized for the moment, although there is some indication it may increase again if the new cross-platform Vulkan graphics framework becomes popular. This increase seems to have been driven by two factors: First, a push from Valve, and second, a move by various game engines to be more cross-platform, reducing the barriers to cross-platform game production. Technology matters here–if it takes 5% of your game production effort to make your game be on another platform, it is unequivocally not worth it to port the game to get 1% of sales–but if it takes 0.1% of your game production effort, then the same port is certainly worth it, if of trivial importance. Currently we seem to be somewhere in between, such that ports to Linux may or may not be worth it–which means that the situation is probably sensitive to relatively small changes either in market share or technology. Hopefully those will go in good directions.

    There are lots of barriers caused by network effects, which are just inherently very hard for minority OSes to overcome. Games is one, and not an insignificant one. Drivers is another, but overall it’s amazing how effectively that barrier has been overcome; not perfectly, but considering that even Apple deals with that one by hiding from it (simply not allowing outside gear to connect), the amount of stuff you can use on Linux with its low market share is incredible. General software is another. It improves Linux’s chances every time one of these barriers is reduced, and that certainly includes gaming becoming viable on Linux.

    As to mobile . . . sure, eventually I expect that computing power will be so powerful and compact that your desktop’s CPU, graphics card etc will all fit in a phone-sized package, and what will happen is if you bring your phone within range of your desktop (that is your screen, keyboard, mouse, printer, external hard drive etc), it will give you a prompt asking if you want to fire that puppy up, and suddenly your big screen will display a desktop-style OS (NOT the same as what you see on your phone even though it’s driven by the phone) and maybe in the background some synching-up with the hard drive will happen. But even so, that will not represent the death of the desktop. You’ll still be computing at a desktop for a certain set of computing needs.

  • Mike

    @CFWhitman

    > “Remember, though, that mainframes haven’t departed, they’re just a lot more niche than they used to be (and some are referred to as “supercomputers”). Desktops will likely be around for a while yet.”

    Bingo, my point exactly.

    Desktops will be around a while yet…just like mainframes continue to be. They will live as a niche item that Joe consumer doesn’t give a rat’s ass about. The next gen of IT people will look at desktops exactly the same as the current gen look at mainframes, i.e. a dusty old computer used only by people stuck in the past. That doesn’t mean they won’t have valid uses, just those uses will become more and more esoteric over time.

  • tracyanne

    @Thad
    August 17, 2016 at 4:07 pm

    >>>@TracyAnne: “Android Phones and Tablets are so successful, because… Marketing.”

    Not exactly.

    Samsung, LG, et al are successful because of marketing. Android is successful because it was the best option available for Samsung, LG, et al to hitch their wagons to.<<>>“Chromebooks are successful, because… Marketing.”

    I would say they’re successful because of their price, weight, and battery life, but marketing doesn’t hurt.<<>>“The only reason Linux hasn’t taken over the Desktop is Marketing.”

    Well, no, not the *only* reason; decades of habit and vendor lock-in are part of the issue too.<<<

    Good marketing overcomes any and all Vendor lock in, poor or non existent marketing doesn't, the Chromebook and any and all Android devices are evidence of that.

    With good marketing, all those, it's not available on Linux issues would disappear, almost overnight.

  • tracyanne

    Something went wrong with my last post, I think it’s my use of >> I’ll redo it, because I can’t edit it

  • tracyanne

    @Thad
    August 17, 2016 at 4:07 pm

    [[[@TracyAnne: “Android Phones and Tablets are so successful, because… Marketing.”

    Not exactly.

    Samsung, LG, et al are successful because of marketing. Android is successful because it was the best option available for Samsung, LG, et al to hitch their wagons to.]]]

    Yes Exactly.. without marketing those Samsung, LG, et al phones and Tablets would not be successful, yes they could be using another OS, but they aren’t, they are using Android, and they are using Android because Google Marketed Android, to them, in such a way that they immediately saw it’s advantages, to them. So Android, a Linux OS is successful because of Marketing. Google did not just throuw the OS out there, they agressively marketed it to the Phone and Tablet OEMs

    [[[“Chromebooks are successful, because… Marketing.”

    I would say they’re successful because of their price, weight, and battery life, but marketing doesn’t hurt.]]]

    The their price, weight, and battery life of Chromebooks would not be apparent, to those buying them, without marketing. How many people are aware of the advantages of any desktop version of Linux? which is NOT marketed at all.

    [[[“The only reason Linux hasn’t taken over the Desktop is Marketing.”

    Well, no, not the *only* reason; decades of habit and vendor lock-in are part of the issue too.]]]

    Good marketing overcomes any and all Vendor lock in, the Chromebook and any and all Android devices are evidence of that.

  • tracyanne

    Good marketing overcomes any and all Vendor lock in, poor or non existent marketing does not, the Chromebook and any and all Android devices are evidence of that.

    With good marketing, all those, it’s not available on Linux issues would disappear, almost overnight.

    This inability to re edit posts is super annoying. When are you going to fix this Christine?

  • oiaohm

    tracyanne you have a mistake.

    Android is successful because it allows a lot of vendor locking. You find out how much when you attempt to put generic firmware on to devices. The custom interface from each android vendor is normal of this. Samsung, LG, et al don’t market android but there own variation of it.

    Android being open source manage to partly unify the multi OSs that were used in feature phones.
    https://www.kickstarter.com/projects/andromium/the-superbook-turn-your-smartphone-into-a-laptop-f/
    Things have changed with the introduction of USB C connector. The USB C connector allows power to go in one direction while instructions are going in the other. So items like superbook are now more possible where the phone is charging yet the phone is controlling the screen and keyboard and other items connected to it.

    So a phone + 100 dollar bit of hardware is basically a laptop. There is going to be a lot of usage overlap between what is performed on a desktop and what people perform using a phone and a dock.

    Linux desktops where they are will most not be effected by this change. OS X desktops minor amount of change. Windows is the one on the chopping block. Question is how many windows users only use windows for minor tasks so really don’t need a Windows PC at all and since they have a smart phone its really not worth them investing in a desktop computer due to the fact smart phone can dock and become one.

  • Denis D

    “Desktop Linux Still Hasn’t Taken Over the World” –
    In order to discuss this we should come to common understanding of what is “Linux Desktop” and what does it mean to “Take Over The World”

    IMHO, marketing is very important thing that a lot of “techie” people just don’t get it. Read “Crossing The Chasm” and “Inside the Tornado” by Geoffrey. It explains how to dp marketing and how to sell High-Tech products to mainstream customers. The only way Linux can succeed on desktop if and only if a sound and specific marketing strategy will be developed. Who is the Desktop Linux for? (target market/niche). What does it do better then others?(value/proposition) A compelling reason to buy. To have even a tiny market, but all of it, is better then to have a little share of the whole market. Be practical. Be THE THING to use in the certain user cases.

    As an example. I work at a graphic design company, and we use Corel products. We are forced to use Windows because there is no Linux driver support for our Vinyl cutters. If there will be on market a little box that runs the software with no issues whatsoever and the peripheral hardware works as it should, we would switch in a heartbeat. But there are none. Got to deal with Window$ – slow computer, viruses etc.

    Until developers of distributions will make their mind and decide, who they projects are for, and what is the long term path their taking to get there, Desktop Linx will be only for Techies/Visionaries/Enthusiasts etc and stay as a niche/not ready yet product. And I’m not even talking about Whole Product solutions

  • Steph

    I have two other reasons for you: the Microsoft Office suite and the predominately Windows-based gaming industry.

    I simply can’t do the full range of tasks with a Linux office suite (specifically excel), or play the library of games that windows supports. And my gaming peripherals all lack Linux support.

    These are not the comments of a windows fanboy. Believe me, if I could ditch Microcrap once and for all, I would. And I hope someday that will be a viable option.

  • tracyanne

    @oiaohm
    August 17, 2016 at 8:19 pm

    [[[[tracyanne you have a mistake.

    Android is successful because it allows a lot of vendor locking.]]]]

    No I didn’t. Vendor lock in is one of the major advantages that Google marketed Android to the OEMs.

    The devices are marketed by the OEMs to the consumer as something that enriches the consumers life, by being convenient and easy to use or as a status symbol or what ever the marketing people perceive as the best way to market the device.

    The majority of people who buy Android devices neither care about Vendor Lock in, nor even realise it exists.

    [[[ Question is how many windows users only use windows for minor tasks so really don’t need a Windows PC at all and since they have a smart phone its really not worth them investing in a desktop computer due to the fact smart phone can dock and become one.]]]

    I would argue that question is irrelevent, so long as those people believe they need to have a Windows Laptop or Desktop, they will buy them, and they will continue to believe that so long as Microsoft’s Marketing people can successfully key into consumer’s perceived needs. The moment consumers realise they can and do, do everything on a mobile device, they will forgo buying any laptop or desktop device, and stay with their iPads and Android Tablets etc.

  • tracyanne

    @Steph
    August 17, 2016 at 8:35 pm

    [[[ have two other reasons for you: the Microsoft Office suite and the predominately Windows-based gaming industry.]]]

    Those are marketing issues, not missing software issues.

    1/ Microsoft aggressively markets Microsoft Office, and they don’t sell a Linux version. No one markets LibreOffice.

    2/ No one markets desktop Linux, either as a gaming machine (with one possible exception, that at the moment seems to me rather half hearted, in other words poorly marketed), or as a consumer desktop machine, as a result there is no market for the gaming industry to attempt to market games to.

    Where Linux OS are successfully marketed to consumers (Android, ChromeOS) you will find the gaming industry is happily marketing their games, you will also find Microsoft attempting market whatever products best fit those devices.

  • oiaohm

    Steph I am sorry to say
    *I simply can’t do the full range of tasks with a Linux office suite (specifically excel),*
    This is your limitation. Please remember Libreoffice calc is used by many governments now todo every single spreadsheet task no Microsoft excel at all.

    Microsoft Office is way less of a limit since you can run the free version now in android emulator for those documents that don’t work. So as good as you get in MS Office compatibility is now open on Linux desktops as long as you are willing to run a android emulator in the cases libreoffice does not work.

    One things that governments have found is python in libreoffice is scarily powerful. One of the big problems with people coming from excel to libreoffice is determination to keep on using basic.

    Changing from Windows to OS X you have a percentage of hardware fail as well. Also you have gaming peripherals that use custom software fail with Windows versions updates. If all your gaming peripherals don’t use standard methods that Linux supports this does suggest future problems.

    tracyanne the reality is there are parties marking LibreOffice.
    https://www.collaboraoffice.com/
    For example this one and they are doing quite well with there contracts thank you very much.

    Also the claim no one markets Linux Desktops is also wrong.
    http://www.dell.com/us/business/p/xps-13-9350-laptop-ubuntu/pd

    Most linux desktops are not marketed at gaming or general consumer desktop but professional web and server development.

    This is where the idea is only about money is wrong.

    tracyanne basically about time you pull you head out the sand and wake up how much commercial marketing there is on Linux solutions.

    tracyanne something else to be aware for google did not market Linux to most of the vendors making Android phones. They were making vendor locked versions of Linux before Android appeared. So its not a feature Google sold to phone vendors. Its a feature that has been a on going bane to Google. All android did was unify a stack of independent vendor locked Linux distributions into 1 distribution with a common framework.

    So really not much you are saying is anywhere near correct tracyanne.

    Libreoffice is development many times faster than OpenOffice because Libreoffice is in fact being marketed and this is resulting is more and more developers working on it employed full time.

  • tracyanne

    @oiaohm
    August 17, 2016 at 10:39 pm

    [[[tracyanne the reality is there are parties marking LibreOffice.
    https://www.collaboraoffice.com/%5D%5D%5D

    So I see. They are marketing it as a business tool, which is good, as it is a great replacement for Microsoft Office, for all businesses, large and small, and the general public too.

    What I don’t see is that they are marketing LibreOffice to the public, the consumers I spoke about. Which means that by the majority of people… and small businesses, Microsoft Office, which is marketed agressively to the public, the consumers I spoke about, is still perveived as the only valid Office Suite tool, and one of the primary reasons why Steph and the majority of computer user believe
    they can’t do the full range of tasks with a Linux office suite.

    [[[Also the claim no one markets Linux Desktops is also wrong.
    http://www.dell.com/us/business/p/xps-13-9350-laptop-ubuntu/pd

    Most linux desktops are not marketed at gaming or general consumer desktop but professional web and server development.]]]

    I did not originally state there is NO marketing of Linux for the Desktop. What I said was

    [[[tracyanne
    August 16, 2016 at 9:58 pm

    Two examples of successful consumer Linux products.

    Android Phones and Tablets are so successful, because… Marketing.

    Chromebooks are successful, because… Marketing.

    Linux as a consumer Desktop or Laptop product is otherwise almost non existent, and where it does exist it is not marketed widely.]]]

    Please note that “…and where it does exist it is not marketed widely.”

    [[[This is where the idea is only about money is wrong.]]]

    In Business, it is always about the money. To believe otherwise, is at best, naive

    [[[tracyanne basically about time you pull you head out the sand and wake up how much commercial marketing there is on Linux solutions.]]]

    My head is, I assure you well and truley out of the sand.

    [[[tracyanne something else to be aware for google did not market Linux to most of the vendors making Android phones.]]]

    Really? So if Google didn’t market Android to them… and

    [[[They were making vendor locked versions of Linux before Android appeared.]]]

    Why in the world did they bother adopting Android, if Google had not marketed, and marketed agressively to those vendors.

    why would they bother adopting Android, if they did not perceive advantages to themselves like:

    android unifying a stack of independent vendor locked Linux distributions into 1 distribution with a common framework

    They are not in business for your benefit, they are in business for ther own, which also means they are not in business for Google’s, which means that even if

    [[[ Its a feature [vendor locked Android devices] that has been a on going bane to Google.]]]

    Google will still market that as a benefit to those OEMS who were already vendor locking Linux based operating systems, because it’s in Google’s interest to get as many OEMS on board selling devices that in the end benefit Google, even if there are some issues that google has to deal with. A cost benefit analysis will always show that “on going bane” as trivial beside the profits generated.

    [[[Libreoffice is development many times faster than OpenOffice because Libreoffice is in fact being marketed and this is resulting is more and more developers working on it employed full time.]]]

    That’s right, primarily, it’s being marketed to Developers… Not average consumers, which means it is not seen by the public, the consumers as a viable replacement for Microsoft Office… which is aggessively marketed to exactly those people.

  • oiaohm

    tracyanne you are completing forgetting Limo in phones.
    [[[Why in the world did they bother adopting Android, if Google had not marketed, and marketed agressively to those vendors.]]]
    No google did not. All google did was provide a pure MIT licensed userspace. Limo and lot of the others before had LGPL and GPL licensed bits so forcing vendors to release bits they had tweaked. So now vendors could take a different base to Limo and not release stuff. So no marketing to vendors was required. Google did aggressive marking to consumers to allow the vendors supporting them to sell more product. Its the wrong idea that Google had to market to the vendors. Google based the android design purely around what the vendors had been asking for. Even the java language syntax support in android comes from the fact vendors were having to pay for java licenses all the time. Make exactly what the vendors wants and they take it fairly much.

    Please remember companies like samsung still work on items like Tizen as their own Linux and it lives with LGPL and GPL core parts. The reality changing from limo stack on soc chip to android on that soc chip was basically nothing. Some of the limo devices made in 2007 got android updates in 2008. Some of the first android phones were made before android existed and by swapping to android meant not having to pay java license. Why such a big dispute between Orcale and Google over java.

    Libreoffice is targeted at businesses, governments and education in most of their marketing. This makes sense. Sorry to say in the office suite market general consumer is not worth worrying about as they are sheep to businesses, governments and education as general consumer has to be compatible with those. Also businesses, governments and education are the areas who are willing to pay for developers to work on improving Libreoffice. Something to remember no developers no features never will be useful.

    So exactly what is wrong with marketing to developers when the product has some major issues.
    http://www.infoworld.com/article/3060246/security/ubuntu-snap-doesnt-have-the-security-issue-x11-does.html

    Remember X11 is a known security disaster. Linux in server and phone world sells as a fairly secure OS. So you push current insecure designs of Linux Desktop on general consumers who from a security point of view are idiots and Linux suffers from major security problems because of it. Result would be Linux current markets would get damaged. Basically until wayland and mir can successfully land there is no point marketing Linux to everyone. Businesses are not going to endanger will paying markets for no good reason.

    So the reason why Linux desktop does not exist yet is companies are not willing to risk the income from well paying markets.

  • oiaohm

    tracyanne you are completing forgetting Limo in phones. All google did was provide a pure MIT licensed userspace. Limo and lot of the others before had LGPL and GPL licensed bits so forcing vendors to release bits they had tweaked. So now vendors could take a different base to Limo and not release stuff. So no marketing to vendors was required. Google did aggressive marking to consumers to allow the vendors supporting them to sell more product. Its the wrong idea that Google had to market to the vendors. Google based the android design purely around what the vendors had been asking for. Even the java language syntax support in android comes from the fact vendors were having to pay for java licenses all the time. Make exactly what the vendors wants and they take it fairly much.

    Please remember companies like samsung still work on items like Tizen as their own Linux and it lives with LGPL and GPL core parts. The reality changing from limo stack on soc chip to android on that soc chip was basically nothing. Some of the limo devices made in 2007 got android updates in 2008. Some of the first android phones were made before android existed and by swapping to android meant not having to pay java license. Why such a big dispute between Orcale and Google over java.

    Libreoffice is targeted at businesses, governments and education in most of their marketing. This makes sense. Sorry to say in the office suite market general consumer is not worth worrying about as they are sheep to businesses, governments and education as general consumer has to be compatible with those. Also businesses, governments and education are the areas who are willing to pay for developers to work on improving Libreoffice. Something to remember no developers no features never will be useful.

    So exactly what is wrong with marketing to developers when the product has some major issues.

    Remember X11 is a known security disaster. Linux in server and phone world sells as a fairly secure OS. So you push current insecure designs of Linux Desktop on general consumers who from a security point of view are idiots and Linux suffers from major security problems because of it. Result would be Linux current markets would get damaged. Basically until wayland and mir can successfully land there is no point marketing Linux to everyone. Businesses are not going to endanger will paying markets for no good reason.

    So the reason why Linux desktop does not exist yet is companies are not willing to risk the income from well paying markets.

  • oiaohm

    Libreoffice is targeted at businesses, governments and education in most of their marketing. This makes sense. Sorry to say in the office suite market general consumer is not worth worrying about as they are sheep to businesses, governments and education as general consumer has to be compatible with those. Also businesses, governments and education are the areas who are willing to pay for developers to work on improving Libreoffice. Something to remember no developers no features never will be useful.

    So exactly what is wrong with marketing to developers when the product has some major issues.

    Remember X11 is a known security disaster. Linux in server and phone world sells as a fairly secure OS. So you push current insecure designs of Linux Desktop on general consumers who from a security point of view are incompetent and Linux suffers from major security problems because of it. Result would be Linux current markets would get damaged. Basically until wayland and mir can successfully land there is no point marketing Linux to everyone. Businesses are not going to endanger will paying markets for no good reason.

    So the reason why Linux desktop does not exist yet is companies are not willing to risk the income from well paying markets.

  • tracyanne

    oiaohm
    August 18, 2016 at 12:47 am and August 18, 2016 at 12:49 am

    [[[tracyanne you are completing forgetting Limo in phones. All google did was provide a pure MIT licensed userspace. Limo and lot of the others before had LGPL and GPL licensed bits so forcing vendors to release bits they had tweaked. So now vendors could take a different base to Limo and not release stuff. So no marketing to vendors was required.]]]

    Of course it was, that’s how Google “Sold” Android to the Vendors. They didn’t simply throw it over the wall and wait for someone to realise the advantages to their business, they actively pushed all those advantages, including the MIT license. That’s called Marketing.

    [[[So the reason why Linux desktop does not exist yet is companies are not willing to risk the income from well paying markets.]]]

    Yeah, that’s basically what I’ve been saying. Marketing, good Marketing requires Money. If a business isn’t willing, or doesn’t have the money” to put money behind a product, or even a potential product, it doesn’t get Marketed. Most companies that want to push Linux, and indeed Open Source Products, don’t have the sort of money necessary… even Canonical doesn’t have the sort of money necessary to push the Linux desktop as hard as necessary.

    [[[Libreoffice is targeted at businesses, governments and education in most of their marketing. This makes sense. Sorry to say in the office suite market general consumer is not worth worrying about as they are sheep to businesses, governments and education as general consumer has to be compatible with those. Also businesses, governments and education are the areas who are willing to pay for developers to work on improving Libreoffice. Something to remember no developers no features never will be useful.]]]

    I never disagreed with what you are saying there. I have merely pointed out the simple fact that for Linux on the desktop to gain the sort of “market share” that Linus Torvolds, and the article, is talking about requires Good Marketing.

    [[[So the reason why Linux desktop does not exist yet is companies are not willing to risk the income from well paying markets..]]]

    Absolutely nothing, that is the way FOSS is supposed to work. What won’t happen is general consumers recognising the advantages of the product.

    If consumers don’t recognise the value of FOSS products, desktop Linux will never gain the sort of “Market Share” this article is discussing. To do that requires good or aggressive Marketing, of the sort that Apple and Microsoft, and the Android OEMS, and Google with their ChromeOS/Chromebooks engage in.

    If consumers don’t recognise the value of FOSS products, potential users of desktop Linux will continue to say the reason they don’t use Linux is because it doesn’t have Microsoft Office or Games or some other thing that hasn’t been marketed to them, and desktop Linux will forever remain a toy, a thing that geeks play with, in the minds of those people.

  • CFWhitman

    @PurpleLibraryGuy
    I want to clarify a little bit. I wasn’t saying that there was nobody at all that held back from switching to a platform because of the lack of availability of games, but that I’ve never seen those people have a significant effect on market share. They’re not the fat part of the graph. Basically, I’ve never seen the availability of games appear to influence the market share of a general computing platform.

    “CFWhitman, as to your last comment on games following vs. leading . . . well, the current case of Linux seems to be a counterexample.”

    The current situation with Linux does not seem to be a counterexample as far as I can tell. As I see it, Linux gaming has grown for two reasons.

    One reason is that, according to sites that track Web hits to show market share, Linux desktop share has about tripled in the past seven or eight years, going from about 0.7% to about 2.1%. So some of the growth in Linux gaming is because there is a bigger audience. It should be noted that the relationship between game availability and market share isn’t 1 to 1. Once audience size reaches a certain critical point, it draws the attention of certain smaller game publishers, then again at another point larger ones, etc. If game availability influenced market share, then market share should show corresponding jumps to game availability. Instead we see a slow, but steady geometric progression (that is, it’s normal for a rise in market share to gain speed as it goes along) of the rise in market share of desktop Linux.

    The other reason is somewhat incidental. Computing platforms (including dedicated gaming platforms) have diversified in recent years, so cross-platform development makes sense for game publishers. Once you are already developing for multiple platforms it becomes easier to throw in one more, Linux, among them. You basically pointed this out yourself, but only because I think you misinterpreted what I meant.

    To clarify what I meant in my original comment: Market share seems to be the biggest influence on game availability, though not necessarily the only one. Game availability always seems to have a negligible effect on market share of general computing platforms. As an example, look at the huge game market there was for Commodore platforms as opposed to IBM PC compatibles in the eighties, yet games followed users to PC compatibles and eventually abandoned the Amiga even though it was still a superior gaming platform at the time.

  • Bo

    There are a lot of lousy Linux distros.That is freedom. I have been using XP,/,8,10 and I don´t think they are ready for the desktop or that they are user friendly. Many people hate windows 10. So what to use when 7 or 8 gets no support anymore? Maybee Linux or what else? I have had good experience with a lot of distros and I have only used the prompt with Windows 10. Why would I need 20 to 30 GB of crap to do the simple things I do on a computer. Beos and Amiga are dead, except for Haiku and Aros which have hardly any drivers, or Solaris which was turned into some kind of database/cloud OS. Nobody develops a new OS so we are stuck with expensive Macs, horrible windows 10, or Linux.

  • @tracyanne: “Good marketing overcomes any and all Vendor lock in, poor or non existent marketing doesn’t, the Chromebook and any and all Android devices are evidence of that.”

    Well, no, they aren’t. StatCounter’s numbers for July 2016 put ChromeOS at 0.38% of the desktop market. Not of the desktop *and* mobile device market combined, just the desktop market. That’s not breaking vendor lock-in, it’s a rounding error. MacOS has more than 20 times as many users as ChromeOS; would you say that *it’s* broken MS’s vendor lock-in?

    Android isn’t an example of breaking vendor lock-in either; Google broke into the smartphone industry before it took off. Smartphone sales figures in 2007 (the year Android was released) were around $122M; by 2009 they were approaching $300M and last year they were over $1.2M.

    Android didn’t break vendor lock-in, it got in on the ground floor, before any vendor lock-in occurred.

    Windows Phone, on the other hand, demonstrates that hundreds of millions in advertising *isn’t* enough to break vendor lock-in, once it’s happened.

    The best example I can think of offhand of advertising breaking vendor lock-in is Chrome. And that was a free download advertised at the top google.com.

    So yeah, good advertising can break vendor lock-in — if you’re advertising a free product at the top of the most-visited website in the world.

  • oiaohm

    Thad if you look closer to what happened when smartphones took off its a different thing happens than what is first expected.

    Just before smartphones take off there is a huge stack of symbian and Linux feature phones. Both at the time open source OSs. Android appears as those just swap over their software giving google explosive growth with no major work. Android market is huge because of how many low cost vendors exist and those low cost vendors came from the feature phone market.

    Thad
    “Android didn’t break vendor lock-in, it got in on the ground floor, before any vendor lock-in occurred.”
    No this is wrong. Feature phones were totally vendor locked in most case forget getting software from anywhere else. When feature phone makers swap to android makers they reduced there vendor lock-in level a lot. I have a early telstra android phone from feature phone vendor the android it runs is written into a rom chip. Yes a rom chip no changing it ever and it forgets everything bar stuff stored in the sim card if you pull battery. It took a little while to convince feature phone makers that they had to make android devices with flash and any form of update feature. So google has been battling up hill and still cannot get phone makers to update on time.

    Windows Phone was not successful in the feature phone market before android because it was too expensive. As Windows phone demos you can market all you like if no one will make the hardware for you forget it.

    Yes ChromeOS is smaller than general Linux Desktop numbers.

    tracyanne the big issue that is neglected over and over again. The biggest reason Linux Desktop cannot take off is vendors who make hardware don’t trust it as desktop. Part of fixing that trust issue is get rid of X11.

    Bo
    “Nobody develops a new OS so we are stuck with expensive Macs, horrible windows 10, or Linux.”
    There is a reason.
    https://www.linux.com/publications/estimating-total-cost-linux-distribution
    Got 1.4 billion worth of kernel development time. People don’t understand how much quirk information Linux kernel and other OS kernels have to have to work. Quirk being if running on X hardware don’t do Y or system will stop dead. Fairly much every time someone suggests they are going to be writing a new OS you go check on it in 3 years time because there are good odds it will not succeed.

    Microsoft attempted singularity os from 2003-2010 as a successor to Windows NT base that basically end up having to be thrown away after few billion dollars of cost to Microsoft because they could not get the hardware support and other things right. Also Microsoft put quite a bit of marketing into singularity as well. Failed to win trust result death.

  • Purple Library Guy

    I think everyone here is using “Vendor lock-in” in unusual ways. Normally, “Vendor lock-in” is not the same as something like a device not being alterable or hackable, nor does it mean anything like “Large established market share” although trying to accomplish vendor lock-in WITHOUT such market share is typically suicidal.
    Vendor lock-in is a situation in which, once you start using a product, it is difficult to switch to anything else because of features the vendor puts in the product. A typical example is software which uses file formats that make it very difficult for competing software to access or create them, thus meaning that if you switch to competing software you will have to recreate files from scratch. Or, say, making use of a non-standard language for scripting that only the vendor has a compiler/interpreter/whatever for. Not to name any (Microsoft Office) names.

  • tracyanne

    [[[[Windows Phone, on the other hand, demonstrates that hundreds of millions in advertising *isn’t* enough to break vendor lock-in, once it’s happened.

    The best example I can think of offhand of advertising breaking vendor lock-in is Chrome. And that was a free download advertised at the top google.com.]]]]

    Advertising and Marketing are not the same thing. Marketing uses Advertising, but it also uses many other forms of communication.

    From Wikipedia:

    Marketing is a widely used term to describe the communication between a company and the consumer audience that aims to increase the value of the company or its merchandise or, at its simplest, raises the profile of the company and its products in the public mind. The purpose of marketing is to induce behavioral change in the receptive audience.[1] The American Marketing Association most recently defined marketing as “the activity, set of institutions, and processes for creating, communicating, delivering, and exchanging offerings that have value for customers, clients, partners, and society at large.” [2]

    The techniques used in marketing include choosing target markets through market analysis and market segmentation, as well as understanding methods of influence on the consumer behavior.

    From a societal point of view, marketing provides the link between a society’s material requirements and its economic patterns of response. This way marketing satisfies these needs and wants through the development of exchange processes and the building of long-term relationships.

    In the case of nonprofit organization marketing, the aim is to deliver a message about the organization’s services to the applicable audience. Governments often employ marketing to communicate messages with a social purpose, such as a public health or safety message, to citizens.

    Also from Wikipedia:

    Advertising is a form of marketing communication used to promote or sell something, usually a business’s product or service.

  • oiaohm

    tracyanne
    https://www.google.com/googlebooks/chrome/
    Sorry you have it wrong. The comic book about chrome was meant to explain the technology ideas behind chrome then have other browsers take it up. Chrome was a tech prototype intended to die and it turned into success. Problem was it was design with security and all the other good things from the ground up.

    Chrome on tech grounds grew to quite a large market share with very little marketing from google.

    The reality is you cannot expect to dig you way deep into a market with a core part of your product security flawed. One of the biggest hindrances to Chrome OS is the fact it been using X11 as well. Google is planing on ripping out X11 and putting freon in with future chromeos. Ubuntu is looking to go Mir to fix X11 problem. Most of the Linux desktop world is looking at Wayland to replace X11. And android already got rid of X11 from the start.

    Chrome is an example make a good secure and functional product the market will find you. Issue you cannot call the current Linux Desktop a good secure functional product.

    Purple Library Guy
    “I think everyone here is using “Vendor lock-in” in unusual ways. Normally, “Vendor lock-in” is not the same as something like a device not being alterable or hackable, nor does it mean anything like “Large established market share” although trying to accomplish vendor lock-in WITHOUT such market share is typically suicidal.”

    Would have been helpful if you had got phone makers to listen to that one. Blackberry and others have vendor locked in features without having a established market share then used those features as sales points to win market share and it worked for a while. Long term correct it suicidal but there are many examples of groups pulling it off for short term gains.

  • Mike

    I don’t think X11 has much to do with Linux adoption or lack thereof.

    I think there are two big issues:

    1) Awareness.
    Most people I know have no idea what Linux even is. This could be combated by both marketing and having pre-installed Linux machines for sale, as long as people aren’t expecting something that would run their Windows software. Android and Chrome succeeded because nobody expected them to run Windows software.

    2) Lack of cross-distro application packages. As a developer, it’s difficult to make your application available for multiple distros without a lot of work. Users typically can’t find an application on github and download a binary package and run it on any distro, creating a hurdle Windows and Mac users don’t have to deal with. While there are severe security problems with doing just that which cause no end of headaches for those other users (bundled malware, trojans, etc.) we are talking about things that would increase market share, not necessarily what would be best in the long run.

    Personally I’m fine with the way things are going for Linux. Marketshare shows a slow but steady uptake. Increasing people’s awareness of software freedom needs to be a priority, because if people don’t appreciate it, then Linux will eventually degrade into a cesspit of DRM and proprietary code no better than those piles of malware known as Windows and OS X.

  • oiaohm

    Mike issues with sound servers, X11 and many other things have blocked formation of cross-distro applications packages for a long time.

    appimage leading to flatpak and snappy. Both flatpak and snappy are going after sandboxing the cross-distro applications. Reason why these cannot complete is X11 cannot be secured so we need Wayland or Mir before snappy or flatpak can be declared ready for consumption in a big way.

    So lack of functional cross-distro application packages leads us back to X11 again.

    I guess Mike you did not understand how much of a super big road block X11 is. Normally many complaints about Linux lead back to X11. Like I don’t want to use command line problem is a result that X11 is not secure anything expecting security ends up done on the command line instead of graphically.

    Long long list of Linux faults leads back to one single area over and over again.

  • tracyanne

    @oiaohm
    August 18, 2016 at 6:56 pm

    [[[Sorry you have it wrong. The comic book about chrome was meant to explain the technology ideas behind chrome then have other browsers take it up. Chrome was a tech prototype intended to die and it turned into success.]]]

    It turned into a success because, in spite of your protestations, they marketed it well.

    Security of a product has now effect on how well it is taken up by the general public. Windows is a prime example, Android is another, because in spite of it’s very real security issues the gneral public continue to buy Android based devices.

    It isn’t the security of Linux, which in spite of any issues with X11, is still orders of magnitude more secure than Windows, that stops the general public from switching to Linux Desktop computers, it’s their almost complete lack of awareness of Linux in any form, even those using Android devices are unaware that it is Linux.

    That is a marketing issue.

  • oiaohm

    tracyanne what is shocking is Windows and Android can pass government certifications for security anything X11 cannot.
    –It isn’t the security of Linux, which in spite of any issues with X11, is still orders of magnitude more secure than Windows–
    So that is purely incorrect. True until you fire up X11 then it is false. X11 design equals security instantly toasted.

    X11 design allows a java/flash applet running inside a browser to in fact control everything on the system. Linux Desktop using X11 is mostly safe from attack because it is in such low volume.

  • Timon19

    oiaohm,

    I used to work on projects that are classified, and we used a mix of Windows and Linux. All of the graphic-intensive applications aside from real-world scene rendering and Google-maps-style mapping was done on Linux machines. And actually, some of that real-world scene rendering was even done on Linux boxes, now that I think about it.

    In fact, every one of those classified projects is using X11 extensively. The government (DSS) has a specific procedure for certifying such things.

  • Nathan

    Its the lack of commercial software. Period. Its really that simple. Everyone has one or 2 pieces of software that they need and if it doesn’t run in Linux they just keep using whatever they are using now.

    So the real question is, how do we get commercial software on Linux. I was hoping WINE and Codeweavers was the solution. Or at least a transition piece. But after some 10+/- years of trying, the software I need still doesn’t run. So the only real solution is for software providers to support Linux directly.

    So how do we get companies to write software for Linux. Its the old chicken and egg thing. No one has interest in using Linux because there is no software for it. No one writes software for Linux because there is no interest in it.

    I gave one solution earlier. If someone like Corel can figure out a way to corner the commercial linux software market, it might provide enough profit to port all their software to Linux. But its a big gamble.

    It seems to me the best hope is for cross platform apps to become sucessful enough that companies have no reason to not support Linux, Mac and anything else. I really hope that things like appimage, flatpak or snappy provide a way for commercial success on Linux. Without an opportunity for profit, we will never see desktop linux go mainstream.

  • Mike

    > “No one writes software for Linux because there is no interest in it.”

    There are thousands of open source applications that say otherwise. There are far fewer proprietary and commercial offerings but even that is changing evidenced by Microsoft planning on releasing SQL Server for Linux. I suppose what you really mean is “No one writes commercial desktop oriented software for Linux.” Although there has been an uptick in the amount of proprietary games available for Linux lately.

    Unlike a lot of Linux proponents, I don’t see mainstream adoption as a worthy goal in and of itself. Rapid mainstreaming of Linux without users understanding the benefits of truly free software will only result in pressure on Linux to adopt the same draconian DRM and anti-piracy measures that infest proprietary operating systems like Windows and OS X. Linux is doing quite well, with good hardware support and application availability. It could always be better, but as long as a certain critical mass of hardware drivers are available, it will survive. Somewhat selfishly: That’s all I require. If more people come to truly understand and appreciate the freedom of Linux and free software, then that’s great, but not a requirement. Linux certainly does not need a huge mass of users howling for their DRM encumbered video streaming to work, at the expense of crippling the freedom of the underlying operating system.

  • tracyanne

    @oiaohm
    August 19, 2016 at 4:21 am

    That is a security issue that doesn’t stop linux from being used by the general public, and the general public is unaware of linux due to lack of marketing. If the general public was aware of Linux, they might be aware of the security issue, but they are aware of neither.

    That is a marketing issue.

  • tracyanne

    @oiaohm
    August 19, 2016 at 4:21 am

    That is a security issue that has no bearing on how aware of Linux the general public is. It also doesn’t stop the general public from using Linux.

    If the general public was aware of Linux they might also be aware of the security issue, they also might not, as most people seem unaware of security issues with Windows, but the fact is the general public is aware of neither Linux nor any potential security issues.

    That is a marketing issue.

  • tracyanne

    @Nathan
    August 19, 2016 at 10:15 am

    Its the lack of commercial software. Period. Its really that simple. Everyone has one or 2 pieces of software that they need and if it doesn’t run in Linux they just keep using whatever they are using now.

    Once again what you are describing is a marketing issue.

  • oiaohm

    @tracyanne
    August 19, 2016 at 5:28 pm
    ==That is a security issue that has no bearing on how aware of Linux the general public is. It also doesn’t stop the general public from using Linux.==
    This is wrong. Lack of marketing of Linux Desktop directly relates to security. OEM vendor is not going to majorily market something that when there big customers attempt to buy it they get told they cannot use it on some security ground due to a design flaw or have to do extra steps to secure it. Basically you cannot get major marketing dollars out the OEMs if the product due to security is going to give them black eyes. Software bugs are one thing but items like X11 are design flawed. Result of having to say design flawed is major reduction of trust its like pushing a known defective product like car or so on. So think do car dealers put massive advertisement into moving second hand cars that could contain defects the answer is no. Do new car dealers internationally market new cars with known design flaws the answer again is no. Yet for some reason you are expecting marketing around Linux to behave differently.

    So one of the biggest reasons there is not marketing dollars for the Linux Desktop where there is tones of marketing dollars for Linux Servers is nothing other than X11.

    Timon19
    August 19, 2016 at 9:17 am
    ==In fact, every one of those classified projects is using X11 extensively. The government (DSS) has a specific procedure for certifying such things.==
    I know about a few myself. Due to X11 not being certified as secure itself the result is there is a stack of extra rules to make sure outside network connections of any form cannot get anywhere near the X11 servers displaying stuff to end users. Most of those DSS stuff I have heard of are based on really old legacy Unix stuff before Windows existed. Sooner we get wayland or mir the better off a lot of these issues will be.

    @Nathan
    August 19, 2016 at 10:15 am
    For flatpak or snappy to work properly we need wayland or mir.

    Pointing to its a marketing issue is ignoring we have 1 major keystone bit of software that is broken in the worst possible way. X11 issue is that by design it is stuffed and was first known to be stuffed in 1986 yes before the first line of the Linux kernel was written. Linux kernel framebuffer driver system was the first attempt to get away from X11 but Linux was forced to use X11 due to video card vendors only providing drivers for X11 for Linux.

    So you want a party to blame for Linux Desktop failure and lack of marketing dollars look no further than video card vendors. And then the very ones who we need to fix it are video card vendors.

    Wayland on Linux comes out of a Intel graphics developer attempting to disprove the 1986 report only to find the only way to fix X11 was rewrite it from the ground up and produce something that is not X11 compatible. Fairly much what the Linux kernel had attempted to-do from the get go.

    We need to push hard on Intel Nvidia and AMD to get functional drivers out for wayland at least and possible mir. Please note swapping over to wayland and mir core does not mean having to give up X11 applications completely just would be X11 applications could be run in ways they cannot mess with every other application at once. Of course items like firefox/chrome and other things you will want ported to wayland as soon as possible.

  • oiaohm

    Here is a good one. How does X11 development at x.org work out what sections of X11 they can remove. Simple they intentionally break the code and if no one complains it must not be required. Totally not the method you want for application compatibility.

    The X11 behavior means we may need a X11 server per application. Why we need Wayland or Mir with flatpak and snappy is 1 for security 2 to make it functional to bundle X11 dependent applications with the version of X11 they want.

    Attempting to get marketing dollars is pointless. We need dollars for Wayland development. People fail to notice that X.org has almost gone under a few times due to lack of funding as well.

    Lot of effort has been put into making Gnome and KDE windows managers look good no where near as much has been put into the lower levels the must work well either. Developers not working in critical areas equal major issues.

  • Timon19

    oiaohm,

    Not much else to say other than you’re right (sort of), but mostly not.

    “I know about a few myself. Due to X11 not being certified as secure itself the result is there is a stack of extra rules to make sure outside network connections of any form cannot get anywhere near the X11 servers displaying stuff to end users.”

    That’s a plain “classified network” issue. Air-gapped in the delivery environment, for sure. However, at the non-TS level, in development, it is absolutely approved to have unclass Linux boxes connecting to the internet. They themselves just have to be air-gapped from the classified network or have an approved data diode connecting them.

    “Most of those DSS stuff I have heard of are based on really old legacy Unix stuff before Windows existed.”

    Sorry, no. Not true.

  • tracyanne

    @oiaohm
    August 19, 2016 at 9:08 pm

    [[[[This is wrong. Lack of marketing of Linux Desktop directly relates to security. OEM vendor is not going to majorily market something that when there big customers attempt to buy it they get told they cannot use it on some security ground due to a design flaw or have to do extra steps to secure it.]]]]

    Given that a Linux distributor, such as Canonical, already do those extra steps to secure X11, the Security flaw in X11 forwarding is more a theoretical one than a real one.

    So yes it is a security issue that actually has no bearing on the Marketing of Linux. It certainly appears not to bother Dell, who are a fairly large OEM, who rather poorly market Linux.

    From the point of view of security issues, you argument would presuppose no one would touch Windows, do to it’s very real security issues, however in spite of those issues, Window is very well marketed.

    And as Timon19 has already pointed out, DSS is not worried about this theoretical security issue.

    the fact is the lack of uptake of Linux by OEMs is a marketing issue that the Linux ecosystem needs to address. Canonical appear to have addressed this issue to a small extent, hence the Dell Linux laptops, and the OEMs such as BQ and Meizu building Ubuntu Phones and Tablets.

  • oiaohm

    Timon19
    August 19, 2016 at 10:36 pm
    However, at the non-TS level, in development, it is absolutely approved to have unclass Linux boxes connecting to the internet.
    This depends on country and I can tell you there is more than the USA. Australia and a few other countries its a absolute no go even for stuffer lower than non-TS due to requirements under privacy act. To make inroads in particular markets X11 need to be dead its faults put it on wrong side of legal requirements.

    The problem here is your fines for a privacy breach will tipple if you were using something that was known defective in design. I am not in the only country with this problem. Yes USA has Linux Desktops marketed more than Australia guess why.

  • oiaohm

    tracyanne
    August 20, 2016 at 12:09 am
    Take a lot closer at dell Linux laptops there is a lot of countries they avoid selling in. They avoid running global programs on them a lot.
    ==It certainly appears not to bother Dell.==
    So this is absolutely not true Dell is avoiding marketing Linux in countries with more strict rules that run into X11 security flaws.

    So no question a percentage of marketing dollars is missing because of X11.

    ==Security flaw in X11 forwarding is more a theoretical one than a real one.==
    That is not the flaw I am refering to. Its the fact X11 by design does not exactly align what window aligns to what application. So once an application is running on X11 it can access every other applications also running on X11

    http://www.hackinglinuxexposed.com/articles/20040513.html

    2004 write up and that stuff still works in current day X11.

    tracyanne there are script kiddy tools out there to take full advantages of the weakness since 2005.

  • oiaohm

    The thing that is the problem with X11 is you technically cannot have on a X11 server two applications running at two different privilege levels properly isolated from each other.

    Because X11 does not provide this kind of separation. You find this kind of separation in OS X, Android, OSi, Windows fairly everything bar Linux Desktops.

    Personal wanting to have classified and non classified on a single screen separated from each other does not work with X11 and that is the way the cookie crumbles.

    That turns out to be very key operational feature more often than not. While it not there there are going to be a lot of places that cannot accept Linux Desktop no matter how much marketing was thrown at it.

  • tracyanne

    @oiaohm
    August 20, 2016 at 12:25 am

    [[[So this is absolutely not true Dell is avoiding marketing Linux in countries with more strict rules that run into X11 security flaws.]]]

    I call bullshit on that. I’m in Australia, used to work for the ATO. X11 has no bearing on why AU gov doe not use Linux Desktops, as a consequence it has bearing on why operating systems using X11 are not marketed in Australia.

  • oiaohm

    tracyanne so did you see the X11 computers in the secure section of the ATO. Yet the same systems are not used outside. Why it that way comes from orders from the Australian Signals Directorate. So the way the ATO computers are placed depend on Australian Signals Directorate. So you were in the wrong section of the Australian government to know why they set of rules they did. Yes need to ask the ASD.

    I have had the fun of sitting down with development leads of Dell and HP and being able to direct ask them the question why. They do attend the Australian Linux Conference tracyanne. I know the answer better than you do because I spoke to the right people.

    Yes between asking ASD questions and asking developments leads questions you get the same answer X11 is trouble.

    The reality is the answers are not exactly nice. Interesting part when you ask to the right places in other countries you get the same problem answer.

  • tracyanne

    @oiaohm
    August 20, 2016 at 4:25 am

    [[[I have had the fun of sitting down with development leads of Dell and HP and being able to direct ask them the question why. They do attend the Australian Linux Conference tracyanne. I know the answer better than you do because I spoke to the right people.]]]

    Well your straight beats my pair.

    That’s what I call keeping your cards unnecessarily close to your chest.

  • Unbeknownst

    Erm, people, I think it’s a kind of exponential thing: it starts slowly, begins to catch speed at a lazy pace and then out of nothing it explodes like an infection. Don’t fret if it’s not increasing by the hour. Let’s just have some patience, shall we?

    Regarding what Linux is or isn’t, time and time I’ve people better than me being proved wrong because Linux simply changes. It adapts. It is pervasive. It goes everywhere and still keeps being Linux, because of the GPL.

    – once upon a time there was no killer application: then GIMP came;
    – once upon a time there was no good browser: then Firefox came;
    – we had formerly no good Office suite: we have now Libre/Openoffice and a couple of proprietary ones;
    – in the past we had to use to Windows for certain tasks; now Windows has to use Linux for some tasks.

    English started slowly, too, and people complained they needed to learn French for international businesses. But English got a firm footing and grew exponentially.

    If and when Linux becomes widely used, I still expect some places to be still using Windows in the decades to come. Not as it is now, but a better, more open Windows, trying to mimic Linux just as they have been trying to mimic Apple’s offerings since forever.

    Maybe I’m delusional, but in the last fifteen years I went from crazy weirdo to delusional — that’s a vast improvement in my condition 😛

  • Timon19

    oiaohm,

    Privacy stuff has nothing to do with classification. You’re mixing customer-facing, non-military things with something different.

    We’re talking past one another. DSS doesn’t give a rip, in general, about privacy, because it’s not their mission. Their mission is entirely different. The systems that I referred to have a completely different set of requirements, thus expectations.

  • oiaohm

    Timon19 really that are not that different when you talk to the OEMs. The issue if OEM push a product to general consumer that is not up to snuff the OEM ends up in the position where some department head that needs it secure ends up asking for the alterations to make the item secure and if it cannot they end up in the pits of hell. Remember OEM making produces is dealing with everyone Timon19 and prefer no matter the customer everyone fairly happy with them. So OEM are after items that suit everyone ends.

    So security and privacy issues is kinda a big thing if we want to get far with OEMs. This is has been ignored and why there advertisement from there is fairly much token.

    tracyanne being ATO backed go check something else.
    http://www.sbr.gov.au/software-developers
    To advertise accountancy software effective into Australia the program need to pass the ATO requirements to get list like on the SBR register.

    This would be indirect advertisement of the existence of the Linux Desktop. You start going through the list of software on the sbr site and compare to list of open source accountancy software and find them all missing.

    There is a lot of indirect advertisement around the world the Linux Desktop does not have because items on the Linux desktop are not up to government standards to be listed.

    Now if you could go around the world and Linux Desktop software was up to what government standards mandated then stack loads of advertising might get us somewhere. Current state is not lack of marketing but lack of development focusing on being conforming to government standards.

    Remember majority of people choose the OS of their home PC based on what they are using at work or school. Of course while Linux Desktop in different countries cannot meet privacy and security of business its not going to get far also while Linux Desktop is not meeting government standards around the world for accountancy its not going to get far. Remember accountancy normally means having to update the software every year out of alignment with distribution releases.

    Stage one get rid of X11 control of output its a barrier.
    Stage two focus in on making sure open source software is in fact up to standard in key stone areas like accountancy if it not market those to get resources to bring those up to standard.
    At stage three then consider marketing of the general Linux Desktop and maybe it will not be required as fixing up the other areas and getting more indirect marketing might do everything that is required.

    Its really simple to say Linux Desktop need more marking. The reason why Linux Desktop is not getting more marking is it not up required standard.

  • Mike

    @oiaohm

    > “The issue if OEM push a product to general consumer that is not up to snuff the OEM ends up in the position where some department head that needs it secure ends up asking for the alterations to make the item secure and if it cannot they end up in the pits of hell.”

    That’s a load of crap. OEM’s don’t choose Windows because it’s more secure than Linux, they choose Windows because of Microsoft extortion and because of existing consumer expectations – i.e. I have a bunch of programs: Will this new computer run my existing software (which uses Windows)?

    > “So security and privacy issues is kinda a big thing if we want to get far with OEMs.”

    If OEM’s cared one bit about security or privacy, they wouldn’t undermine it every chance they get with rootkits and self-signed security certificates. If they cared about privacy at all, they definitely wouldn’t use Windows.

  • oiaohm

    > If OEM’s cared one bit about security or privacy, they wouldn’t undermine it every chance they get with rootkits and self-signed security certificates. If they cared about privacy at all, they definitely wouldn’t use Windows.

    No you don’t get it. Let me explain something. OEM sell two versions of most systems with Windows on it. One with garbage and one without garbage. The undermined models allows them to sell more expensive not undermined models that cost the exact same amount to make. Lets just say OEM are greedy. To sell increased price models they need to be able to fix all the major security flaws including removing the ones they added.

    >>some department head that needs it secure ends up asking for the alterations to make the item secure and if it cannot they end up in the pits of hell.
    Notice that bit you quote from mine Mike its a upsell. OEM need the means to upsell secure version. They cannot upsell Linux while it contains X11. Yes OEM are all about money and getting the most money equals selling a secure product and to price that product up they sell security broken products as well.

    Basically you are making a big mistake if Linux is truly solid security wise you would see OEM upselling it way more.

    A lot are taking the wrong point of view since OEM sell broken products that they don’t care about security that is wrong. The issue here is OEM sell broken products so they can price more secure products higher.

    > I have a bunch of programs: Will this new computer run my existing software
    This point you miss if it like software you need to run you accountancy its even that open source accountancy programs are for windows, linux and OS X they are mostly not certified of course these programs will not be peoples existing.

    Remember lots of people have chrome and firefox in there stack of existing software. These are not barrier to migration and these are conforming to the standards in web browser field so most cases workable.

    The big thing here is even the person said they would swap software with accountancy software in Australia when the question comes what is the ATO certified replacement hello dead end because there is not a single open source bit that is certified and the commercial software on Linux is only the really high end stuff. This is not a problem just to one country either. The countries with the lowest take up rates of Linux desktop aligns to countries when you don’t have government approved accountancy software for Linux across all market price ranges.

  • Mike

    > “No you don’t get it. Let me explain something. OEM sell two versions of most systems with Windows on it. One with garbage and one without garbage.”

    No, it’s you who doesn’t get it. You cannot sell a system with Windows and claim to care about privacy. Privacy and security can only be achieved through open source. Anything else is blind faith that Microsoft (or other corporate entity) will not misuse its position of total control, even in the face of government pressure to harvest and/or manipulate data…in other words, burying your head in the sand.

  • oiaohm

    >>No, it’s you who doesn’t get it. You cannot sell a system with Windows and claim to care about privacy. Privacy and security can only be achieved through open source.

    http://www.computerworld.com/article/2580563/microsoft-windows/microsoft-opens-source-code-to-governments.html
    Mike just because we cannot look at windows source code does not mean big enterprise cannot. Yes then can do binary compare between what is provided and what would be generated.

    Mike basically welcome to the extremely dark side of upsell.

    So to government and big enough enterprise the difference due to Linux Desktop being open source and Windows source under NDA is not much different at all.

    So the Privacy and security card claim that blind faith has to be used with Microsoft is false. If you don’t want to use blind faith with Microsoft you have to pay enough money.

    So this means a Linux Desktop has to match up and exceed the built in features of the Windows Desktop to get places.

    Really Mike you are the one who did not get it to large enterprises and governments Windows is not a closed source product. Windows a proprietary bit of software with paid access to source code.

    If we are going to succeed against Microsoft we do have to understand what their product is. So no I am not burying my head in sand on what windows is Mike you have well and truly have.

    OEM have a upsell for a security worried so do Microsoft.

  • Mike

    @oiaohm

    Your description (despite being wrong, which I will get to in a bit) only promises privacy and security for enormously rich entities, being out of reach for all but the most wealthy of corporations and governments. That is not a viable solution. FOSS is the only REAL solution.

    But your basic premise is wrong anyway. Microsoft does indeed allow shared source access to some high-paying paying customers. However this is NOT complete access to the entire source of Windows, and you most definitely can NOT build it yourself and “do a binary compare”. They simply do not give that level of access. You can run a debug session using a copy of windows with symbols and use the online source portal for debugging, but that’s it. That does not give you a full picture of the OS, or prevent it from hiding certain things from you via the closed source visual studio debugger. This may be enough to give a warn fuzzy feeling to some bureaucrat, but any security claims are purely illusory. Only FOSS provides real verification capability.

    Additionally your continual harping on X11’s design flaws as real security issues are misleading in the extreme. Yes an X11 application can read the keyboard, screen etc. of another process. In reality this is not as big a deal as you seem to think. Sure, it’d be nice to have greater isolation between processes, but in reality it’s only a problem if you are already running untrusted applications with permissions capable of altering the system. With most distros using signed repositories of trusted applications, this is practically a non-existent scenario for typical users. If you are already executing malicious code, you’re already pwned. Windows greater process isolation in the UI will not save you here either. So what you are left with is smoke and mirrors – aka marketing.

    So what I hear is a lot of hoopla about nothing. Windows is less secure than Linux in actual operation, and provides no security or privacy guarantees whatsoever to 99% of all users, and some warm fuzzy illusions to a handful of entities with more money than brains.

  • oiaohm

    >But your basic premise is wrong anyway. Microsoft does indeed allow shared source access to some high-paying paying customers. However this is NOT complete access to the entire source of Windows, and you most definitely can NOT build it yourself and “do a binary compare”.

    This is so wrong its not funny. Microsoft sometimes shares the complete source with Unis as well. There was a version of a Dom0 for xen that was built at a Uni from Windows Xp source code of course without Microsoft approval it could not be shipped only demoed. Yes an enterprise development devices can pay for full source access include the rights to build the thing completely from scratch. NDA restricts from releasing the binaries or the source. So really wrong companies are allowed to build the shared source they are not allowed to deploy or ship the shared source. So anyone who does pay enough can build the complete source for prototyping or security reasons.

    > With most distros using signed repositories of trusted applications, this is practically a non-existent scenario for typical users.

    Problem is you are extremely wrong. One of the advantages Windows has over Linux is the means for users to practically install applications that microsoft does not ship. There is a need the current Linux Desktop does not service. When dealing with government departments around the world you will find that you are required to use X version of an application new or older may not cut it. Of course X version may have a security flaw so is untrustable.

    Remember global tax time does not align with Distribution package updates at all.

    –Windows greater process isolation in the UI will not save you here either. —
    Microsoft Enterprise Desktop Virtualization (MED-V) about time you look this beast up. So yes windows in enterprise desktop versions already have something like flatpak.

    If the malicious code has been run using Visualization containment as Windows enterprise provides and to make it integrate correctly into the desktop you need a UI that does separation the system is not pwned. So Windows enterprise already has a system for running untrustable applications.

    Sorry the Windows difference is not smoke and mirrors.

    Remember those signing off on marketing budgets more often than not will be using Enterprise version of Windows so will be feature comparing the Linux Desktop to Enterprise editions of Windows and currently Linux Desktop does not cut it.

    The ability to run untrusted applications securely is a end user need not a want. The ability of the end user to run multipliable of the same application just different versions is also a need not a want.

    Like it or not current Linux Desktop put head to head with current Windows Enterprise loses on security ground hands down. To turn this around X11 has to stop being the default item in-charge of the desktop and flatpak or snappy as to come to floor.

    No point claim Windows is less secure in actual operation compared to Linux is true against the version of Windows given to the low end market its currently not true against the enterprise windows desktops. We want it to be true this means pull head out sand and seriously accepting the faults.

    Yes high end enterprise have security and privacy guarantees out of Microsoft and those are the ones in charge of most of the marketing budgets. The reality is we have to aim to better than the high end Microsoft enterprise products not just be better than the low end garbage products to get marketing dollars. Linux Desktop does not have marketing dollars because our target is set way too low.

  • Purple Library Guy

    I’m beginning to think that the concept of “marketing” on this thread has moved from theoretical to practical. That is, someone is practicing it.

  • Mike

    @oiaohm

    Wow, so many chunks of BS I hardly know where to start.

    I know for a fact that Microsoft’s shared source program does not allow full compilation or modification of the Windows source. If you suggest there is some other way for companies to get access to full Windows source, provide a link to its existence. I know you can’t.

    > “One of the advantages Windows has over Linux is the means for users to practically install applications that microsoft does not ship.”

    Linux users can’t possibly install applications beyond what “ships” with Linux? Sure. Wanna buy a bridge?

    > “Microsoft Enterprise Desktop Virtualization (MED-V) about time you look this beast up.”

    MED-V? Really, that’s the best you can come up with?! An obsolete 32-bit only virtualization platform based on the decade old Microsoft Virtual PC software that Microsoft purchased and stopped development on. Applications aren’t contained, they’re just running remotely on a Win XP VM. A malicious app here would pwn the back-end host…that’s even better than getting a single desktop.

    Linux has genuine faults that need to be addressed in future…but current Windows (even Enterprise) has even more. I’ve seen organized Windows enterprise environments brought down because of a single click on an e-mail.

    You’d make a good Windows salesman to someone who didn’t know better (i.e. most people).

  • oiaohm

    Purple Library Guy the reality I do expect to be able to market Linux Desktop in time.

    Practical rules of marketing.
    1) Know your competition.
    2) Know your products defects.
    3) Know your products advantages
    4) Don’t lie.
    Basically Mike is not marketing does not know what the advantages/defects are. Win market you must have a trump card.

    What is the Current Linux Desktops trump card. The reality is it does not have one.
    1) Current Linux Desktop is not secure.
    2) Current Linux Desktop does not have the most applications.
    3) Current Linux Desktop is not truly stable due to issues with X11 and other bits.
    4) Current Linux Desktop is not easy for users to install third party applications to meet government and company requirements.

    Now lets look at picture after we get wayland and flatpak.
    1) Future Linux Desktop depending on compositor operation could be secure.
    2) Future Linux Desktop Still most likely short on applications.
    3) Future Linux Desktop should be possible to be fully stable.
    4) Future Linux Desktop should be fairly easy to install third party applications to meet government and company requirements without ruining system security due to flatpak.

    So fairly much chalk and cheese. If the future Linux Desktop can truly play the security trump card it will have means to properly win market share.

    Issues around not being able to get production ready drivers for wayland out of Nvidia and others is a major problem. Wayland not being fully feature complete is another.

    Saying companies should invest in Linux Desktop marketing is not going to get anywhere. We want them to invest we have to have a decent product for them to invest in.

  • oiaohm

    >>I know for a fact that Microsoft’s shared source program does not allow full compilation or modification of the Windows source. If you suggest there is some other way for companies to get access to full Windows source, provide a link to its existence. I know you can’t.]

    It called a “Joint Development Agreement” NSA has one along with many other parties way before 2003.
    http://www.wsj.com/articles/SB1042592554169993184
    Windows source complete goes out to many parties for third party security review under Joint Development Agreement. This can be parties who making devices as well so they can try out windows on their hardware and do modifications and submit them back. The Joint Development agreements predate Microsoft releasing anything called shared source.

    To do full security auditing the right to build it is in fact include so you can build it with your own security diagnostic tools. There has been a odd leak here or there out the problem like the windows 2000 source leak. So there are in fact two programs with Windows Source code. Shared Source and Joint Development. Basically Mike considered the existence of one.

    >>MED-V? Really, that’s the best you can come up with?!<>I’ve seen organized Windows enterprise environments brought down because of a single click on an e-mail.<<
    There are metasplot examples that go though thunderbird under Linux as well. So could the current Linux desktop be brought down by a single click on an e-mail the current answer is unfortunately yes.

    Now could this be changed after wayland and flatpak the answer is yes if it used. Was the windows enterprise environment using App-v to contain the e-mail client I would say no it was not.

    Security is only useful if it used. But to sell the product like a OS it better be possible to use it.

  • oiaohm

    >>MED-V? Really, that’s the best you can come up with?!
    MED-V is replaced by App-V. The beast joke refers to something Microsoft sent out to support paying customers telling them to migrate.

    Mike basically Microsoft started doing something like Flatpak with Med-V and has keep on doing it with App-V right into current day Windows 10. So this is an area where the Linux Desktop need to catch up yesterday. To catch up this is X11 stop being the item direct charge of the screen so that flatpak or snappy can be deployed providing proper security.

    Yes Microsoft stopped development on Med-V and started development on App-V.

    So no Med-V is not the best I could come up with but its the first time Microsoft put out anything in flatpak class.

    Next time you think you have a point Mike do me a favor and make sure its not like this one where you are being a idiot. I just gave the the point were Microsoft first added the feature. Just because Microsoft ends of life a part does not mean they removed that feature completely.

  • Mike

    @oiaohm

    > “It called a “Joint Development Agreement” NSA has one along with many other parties way before 2003.”

    Hahaha, nice try. I’m aware of that program, but it is even more restrictive in several ways than the shared source program. It isn’t open to corporate entities at all, only governments. It also only grants access to the source while in a Microsoft “Transparency Center” (hilarious doublespeak on Microsoft’s part, worthy of George Orwell’s 1984). You can NOT build a custom version of Windows on a machine under your control. You also cannot independently evaluate the accuracy of the build as all the machines used to evaluate it are under Microsoft’s complete control.

    > “Yes Microsoft stopped development on Med-V and started development on App-V. So no Med-V is not the best I could come up with but its the first time Microsoft put out anything in flatpak class.”

    App-V? Still just virtualization/streaming – not application containment. You really don’t understand the technologies you are espousing. It’s closer to Citrix which I worked on back in the 90’s than anything like flatpak. Another Fail.

    > “Next time you think you have a point Mike do me a favor and make sure its not like this one where you are being a idiot.”

    It’s pretty clear who the idiot is here and it isn’t me. You really are a moron and a Microsoft shill (unintentional or otherwise). Please keep trying though. I like debunking stupid claims and marketing BS.

  • Mike

    @oiaohm

    keep moving those goalposts. 🙂

  • Frank Andersson

    The main reason is: just about 2-3% of pc users can and will install any operation system. 97-98% won’t never do it. In that perspective Linux desktop has actually been quite successful. Majority of those installing any OS will do it with Linux.

  • Timon19

    All this kvetching about how X11 is insecure and Thunderbird allowing malicious code execution has nothing whatsoever to do with the dozens of systems I’ve personally worked on and deployed whose main operations – including graphics processes and classified data manipulation – are handled exclusively in Linux and/or Irix. All of which DSS HAD to sign off on (also, in most cases, USAF).

    The only stuff run on Windows was effectively a “front-end” for interfacing with a student in the actual system run primarily in Linux.

    Yet I’m to understand that it can’t be done (oh, and FWIW, my company also worked with the Australian government and used – guess! – Linux for the primary processes of the systems they sold them).

  • oiaohm

    Mike reaching at straws much. Microsoft “Transparency Center” don’t exist in 2003. To be correct countries and parties allowed to use Transparency Center normally cannot pass the requirements to get joint development agreement.

    Intel, AMD and Nvidia all have Joint Development Agreement access rights with Microsoft. Joint Development Agreements for the full windows source code is not locked only to governments either. So there another program you are totally not aware of Mike. I pointed to the 2003 reference for a reason you are looking for the oldest program Microsoft started sharing their OS source code under. The newer programs are more restrictive to get around USA export restrictions. Yes how do you show source code to countries by USA law you are not allowed to export particular security things to gets very George Orwell. People who have Joint Development Agreements with Microsoft can submit code alterations back for Microsoft review. So completely different level “Transparency Center” that is fairly much tick and flick or shared source programs that are restricted samples.

    https://en.wikipedia.org/wiki/Microsoft_App-V#Standalone_mode
    Mike I think you missed is stand alone mode. So it send out a image with all the parts include the OS core the application depends on to the client machine to run. Flatpak is all the userspace stuff. App-V is totally different to Citrix of the 90s. App-V is virtual images running on the client machine for desktop class applications where Citrix thin would be running server side. Two

    MED-V, app-v, flatpak, snappy are methods to deal with exactly the same problem. How to add an application to a system without disrupting the currently installed software and integrating so user does not notice too much and and without requiring constant network connection because they operate stand alone. Citrix of the 90 does not address this problem take away network connection thing fails right. Yes app-v includes a streaming as well as stand-alone.

    I just noticed with App-v 5.0+ Microsoft has decide to remove the abilities to lock permissions on applications to prevent applications from causing hell and introduced a copy on write thing. So Linux Desktop without X11 and with flatpak there is a sales point better security around isolated applications. But we don’t have that sales point to use at this stage.

    Mike stop accusing me of moving the goal posts because that is exactly what you have attempted todo to defeat my points. Reality you are taking lemons and comparing to the apples I am referring to.

    Timon19
    >The only stuff run on Windows was effectively a “front-end” for interfacing with a student in the actual system run primarily in Linux.
    Lot of cases the reason why Windows is front-end and systems are not Linux all the ways though is X11 faults get in the way at particular point.

    I am not saying Mil and others don’t use Linux Servers for backend tasks were the security side passes. The desktop side is being hindered by the X11.

    Frank Andersson yes Linux has done well with what it has got. The reality is we are to the knuckle down point. Knuckle down point is where you have to have the need features. If you don’t you cannot progress forwards no matter how much you spend on marketing.

    We also need to get brutal. If something is not up to standard we have to start demanding it be replaced or improved. X11 case only option is replacement. Once items like X11 are not the default we then can start selling security and win the debates. Not having to do the crap Mike does of attempting to compare to something I am not referencing to at all.

    Basically fix the security people marketing Linux Desktop will have something to work with. Currently attempting to market Linux Desktop you will get bogged down by the defects with no option to provide if they do this it secure.

  • oiaohm

    http://www.mupuf.org/blog/2014/02/19/wayland-compositors-why-and-how-to-handle/
    Read above this explains the X11 issue.

    Ok X11 application using Windows or OS X as desktop in mil why. X11 application can complete stuff up the X11 server and this is because of the way X11 is designed so not classed as secure or stable. So under Windows or OS X X11 server is not primary so user can go to task manager kill it and start it over again without losing all their work. Current Linux desktop X11 is primary one bad application X11 stuffs up only option is kill X11 server again except now users loses everything.

    X11 protocol is not suitable to use a primary desktop interface due to how bad it is. The fact Linux Desktop has got as far as what it has with X11 servers shows there is a decent demand out there.

    There is no point pan handling for more marketing dollars while X11 in part of the default offering. It would be better to be asking for development dollars to get rid of the barriers.

  • Timon19

    Dude, you just flat do not know what you’re talking about and you’re using your very narrow experience as if it is universally applicable. I put “front-end” in scare quotes for a reason – it’s because it’s really only the instructor’s interface to interacting with a student that he is training. The student sits and interacts directly with hardware driven directly by Linux boxes with graphics presented to him USING X11, which he is manipulating using said hardware. That graphics package is literally composed of several borderless windows on an X11 pipe on an extended desktop, split by broadcasting-grade hardware to be presented in several places in the student’s seat. The student’s “front-end” is entirely Linux-based and produced, with a huge X11 component. DSS and USAF are so confident that it’s OK that they’ve been doing business with the company for decades in exactly this way.

    You need to get off your hobby horse and re-think how you present your arguments.

    Unrelated, it would help your cause greatly if you could write coherently. I have absolutely no idea what your first paragraph in the latest post says because it’s a mess.

  • Mike

    @oiaohm

    Moving goalposts:

    First you claim enterprises and universities can get complete access to Windows source and do binary compares simply by paying more (an upsell you claimed). I disprove that and you retreat to a special government program for the likes of the NSA and Russia. The GSP program (despite initial claims by some news outlets) is the very same one that restricts access via Transparency Centers. Since the early news stories circa 2003 widely listed Russia and China as early adopters, while at the same time speculating on how much access would eventually be granted by Microsoft, it is interesting to see that those countries have decided to move away from Windows after the Snowden revelations…obviously they don’t trust Microsoft and are not satisfied with the information provided by Microsoft. Despite all this transparent access they supposedly have.

    Now you claim Intel, AMD, and Nvidia are part of another program. While I could see Intel being privy to some stuff at Microsoft (they don’t call it the Wintel monopoly for nothing). Nvidia and AMD simply don’t need that level of access to write video drivers. Go ahead and provide one single link that suggests that such an agreement has ever been made.

    You will simply move the goalposts again. In the beginning you were talking about an upsell for enterprises to validate Windows source…you’ve retreated so far you can’t even see the original goal anymore. 😀

    If you want to validate what a binary really does…you use FOSS (Linux). Anything else is, as I said before, smoke and mirrors.

  • oiaohm

    Mike Intel, AMD and Nvidia all three make motherboard chipsets so there is a very valid reason for there higher access.

    –First you claim enterprises and universities can get complete access to Windows source and do binary compares simply by paying more (an upsell you claimed). —
    At no point did I retreat from this of course you wish you had won that point. I have just being bringing in more example data on the same point. The uni that made Xen version of windows XP had a Joint Development Agreement not a share source agreement.

    I never moved from this point at all. Shared source naming starts 2001 but Joint Development agreements start with Windows NT 3.5 in 1994.

    https://www.microsoft.com/en-us/sharedsource/mvp-source-licensing-program.aspx
    Even this one under shared source does not mandate the source code be kept on Microsoft systems or that you cannot build the code for binary comparing reasons. Please note Mike did say this was not allowed under shared source to build binaries when in fact the license does not forbid that. Remember how Nvidia binary blob gets around Linux kernel license its built into driver and then not shipped.

    Note the yearly fee. Joint Development agreements and share source agreements have yearly fees. Location holding the source code has to be in the same list of countries.

    So there are three ways to look under the hood of Windows its not a 100 percent closed source thing.

    Joint Development Agreements are different to the shared source agreement with the Joint development agreement rebuilding the complete thing with a alteration is allowed. Like the UNI Xen Dom0 Windows XP example rebuilding core to work in that environment.

    >>If you want to validate what a binary really does…you use FOSS (Linux). Anything else is, as I said before, smoke and mirrors.

    FOSS lets you look under the hood without paying a yearly fee as Microsoft mandates to-do the same thing. Sorry your smoke and mirrors does not hold.

    Validate is simple if you have binary you must be able to have access to source and the license does not have to be FOSS.

    Timon19
    >You need to get off your hobby horse and re-think how you present your arguments.
    Not at all you need to rethink yours. Just because X11 does not have a blanket ban does not mean the restriction on usage is not in effect. So on a particular usage case they have got X11 past. Now how many other usage cases has X11 not being used because is failing security quite a lot.

    No Timon19 you need to get of your hobby horse of a very limit exception and look at the big picture. The security side has to be addressed to move forwards.