2015 – The Year Of The GNU/Linux Desktop

There has been a lot of good news in the web-stats of GNU/Linux this year. It’s become clear to me, despite what others write, that the Year Of The GNU/Linux Desktop can happen and it is happening this year. The reasons are many:

  • That Other OS is sputtering along with the shipments of legacy PCs
  • Canonical has connected with OEMs to ship Ubuntu GNU/Linux. e.g. Dell and HP are shipping many units in India and China.
  • Android/Linux has shown the world that */Linux works for them.
  • Most of the major countries have left ~1% share of page-views behind with the exception of China which has a special relationship with TOOS. Despite that, Asia has still risen above 1%.
  • USA, M$’s home field, has reached 2%.
  • Europe has been there a while but still grows.
  • Many countries in Europe have much higher levels of usage and they continue to grow.

Others have written that GNU/Linux can never make it on the desktop but it has. I thought this happened years ago but that was just the first cracks in the damned monopoly. This is a flood of adoption. The worldwide share has risen from 1.5% to 2% in one quarter. If these trends continue for the rest of the year, we could exceed 3% this year. Last Christmas, GNU/Linux was a gift. It will keep on giving next Christmas.

See for yourself:
We’ve had a phenomenal quarter of growth in the web-stats and there’s no sign of decline except in a few places like Uruguay where the kids have been out of school and the schools are major users…

See Top 7 Desktop OSs from 1 Jan to 22 Mar 2015.

About Robert Pogson

I am a retired teacher in Canada. I taught in the subject areas where I have worked for almost forty years: maths, physics, chemistry and computers. I love hunting, fishing, picking berries and mushrooms, too.
This entry was posted in technology and tagged , , , , , , , . Bookmark the permalink.

87 Responses to 2015 – The Year Of The GNU/Linux Desktop

  1. antifanboy wrote, “The point is that in 2008, you predicted that at 2010, desktop (keyword, “desktop”) Linux adoption will rise to 46% from 7%”.

    So, I was wrong, partly. There were many successes of GNU/Linux on the way, like Uruguay, 12.29% on 2009-August-7, Munich, of course, the French Gendarmerie, etc. GNU/Linux’ success, not mine. It’s doubtful that GNU/Linux will reach ~50% globally but it is possible and feasible. It’s doubtful because consumers will have choices like ChromeOS, Android/Linux, TOOS and MacOS, all of which are feasible, but no longer will there be a monopoly on the desktop as existed around 2005. So, my new target would be in the range 20%-40%. Probably 50% is out of reach. M$ and Apple can give their OS away for $0 which will enable it to compete on price. Consumers and OEMs will have to figure out the eventual share but it will be far above ~1% for GNU/Linux.

    Good news! Yesterday, on Christmas Island (pop ~2K), StatCounter recorded 31% GNU/Linux. From their data, it looks like not much was happening until this week but it’s happening even in the far corners of Earth. Get used to it.

  2. antifanboy says:

    The point is that in 2008, you predicted that at 2010, desktop (keyword, “desktop”) Linux adoption will rise to 46% from 7%(it’s really 0.7%http://gs.statcounter.com/#desktop-os-ww-daily-20080701-20101231), MacOS up 12% from 3%, and Windows(“Windows”, not “Lose”, “that other OS”) dropping to 41% from 90%. But when 2010 came and went, the stats was still the same as was in 2008 with desktop Linux still stuck at 7%(0.7%) and people started confronting you about it. However, instead of admitting you was wrong, you got defensive and started lashing out on them. And on top of that, in a (failed) effort to make yourself look good, you threw smartphones into the mix:

    “Well, that did turn out to be optimistic. “7” did pretty well retail. OTOH Linux exploded on Android and M$ lost huge share. There are still nearly 50% of PCs using XP so the predictions may still come true in 2011. The monopoly was certainly weakened in 2010 and ARM is expected to take as many computing devices as M$ in 2011 so it will not be long.”

    So to summarize this, you have proven time and time again that 1) you can’t take even small amounts of criticism and believing that you’re right all the time because of your grandiose sense of self that is rivaled only by Richard Stallman’s , and 2) this wasn’t the first time you was proven wrong and it certainly wasn’t the last.

  3. antifanboy wrote, “Explain this:
    http://mrpogson.com/2008/07/24/the-monopoly-will-end-in-2010/

    Uh, 2010 GNU/Linux’ share was 0.78% while in 2015 GNU/Linux’ share has been 1.61%, more than doubling. That doesn’t happen when monopoly gets its way. Consumers are finding GNU/Linux and other */Linux on retail shelves and all kind of individuals and organizations are adopting GNU/Linux and other FLOSS. In France the numbers were 1.7% and 2.49% with a new high of 3.81% just yesterday. I’d say the monopoly has ended on the supply side and the installed base is rapidly catching up.

    So, I was a little optimistic on the maths. Sue me. If you count all operating systems that StatCounter tracks, “7” is at 33.69%, “8.1” is at 9.25%, Android/Linux is at 20.78%, and GNU/Linux is at 1.2%. Total for M$ is 55%, a far cry from what M$ wants.

  4. antifanboy wrote, “What about the wishful thinking by Linux evangelists like you thinking every year from 1998 is the Year of the Linux desktop and getting it wrong 16 times and counting.”

    That’s a good question. I’ve documented on this blog serveral ways that 2015 is different from all those other years:

    • The magnitude of the changes seen are what we used to see in a year but in 2015 they are happening in a few weeks. It’s possible this is some effect of how StatCounter does things having changed or how StatCounters base of monitored sites evolve but the changes are so large it’s hard to imagine StatCounter not catching some blunder. Also, we know Wintel is struggling, so the production/installation of GNU/Linux systems stands out more, but what was fair for Wintel in their “Good Old Days” is fair for GNU/Linux today.
    • These changes are quite widespread and spreading further. In fact, even the weaker regions of the planet are approaching ~1% when that level was mainstream a few years ago. There doesn’t seem to be any significant decline of GNU/Linux anywhere. That other OS has nowhere to go but down and GNU/Linux is helping every way it can.
    • The levels we see in the most popular countries are very serious indeed, like ~2% for USA and India, ~3% for Germany, France, and ~5% in Italy. We are clearly at the level where the “positive feedback” that M$ used to boast was theirs for the taking is now belonging to GNU/Linux. e.g. USA has ~2% for GNU/Linux but also ~1% for ChromeOS a variant of GNU/Linux.

    This is wildly different than those other years. Change is bigger, faster, in a few months now than it was in years back then. I think there were other special years for GNU/Linux when the first GNU/Linux workstations shipped, then the netbook, then with DELL selling to consumers. There’s no single factor working today. It’s dozens of good things happening for GNU/Linux all in one year, like Dell&Canonical shipping in China and India, governments promoting FLOSS & GNU/Linux, ARMed CPUs becoming powerful enough for desktops, Moore’s Law making power/size/cost almost irrelevant to FLOSS, and, of course to the wild success of GNU/Linux on servers and Android/Linux on smart thingies. The skids are greased for GNU/Linux and it’s picking up speed, not downhill but down the valley sweeping all before it.

    We saw what happened last Christmas with GNU/Linux webstats jumping. That was consumers actually having choice on retail shelves in many places. With OEMs and retailers having seen that success and Chromebooks selling well, there’s nothing left to hold back GNU/Linux. The market for GNU/Linux is growing and TOOS is shrinking or static. ISVs will exercise their “independence” for real this year, 2015, and blow away the myth of the “must-have” app being unavailable on GNU/Linux. It will take a while but it will happen with a vengeance in 2015. There’s just no more lock-in with many using web-applications and other applications becoming available on GNU/Linux.

  5. antifanboy says:

    ““Never” implies a strong religious belief unsupported by the steady growth of GNU/Linux on desktops and everywhere else.

    It’s wishful thinking by the opponents of human progress. It is a sure signed evidence that if they truly believed what they say they wouldn’t feel the need to combat you.

    The future is open!!”

    What about the wishful thinking by Linux evangelists like you thinking every year from 1998 is the Year of the Linux desktop and getting it wrong 16 times and counting.

  6. Dr Loser says:

    To amplify Deaf Spy’s sadly abbreviated, yet entirely justified, effort to pour scorn on Fifi for the usual “I can google, me, but I just pick words out of the Wikipedia Bag at random!” attempt to claim something or other about NT:

    The reality is almost no Windows application is coded using just the Windows Native API.

    The reality is that no Windows application at all is coded thus, oiaohm. It’s not even a point of discussion. It’s just something that you dreamed up in your pointy little head.

    As Deaf Spy points out, the “Windows Native API” is not, actually, an interface to the application layer.

    It is fairly simple to detect applications using just Native as all functions start with one of the 8 prefexes.

    “Prefexes” presumably being the illiterate plural of “Prefect,” Fifi. Let us list those “prefexes:”

    * Nt or Zw
    * Rtl
    * Csr
    * Dbg
    * Ki
    * Ldr
    * Nls
    * Pfx

    See what I did there, Fifi? I did exactly what you do, as a matter of course — I copied the information out of a Wikipedia page. Well, actually, I did one better. I bothered to enumerate the stuff. And one better still, because …

    … I, personally, have programmed against this level of NT API code, Fifi. You have not. Not a surprise, since you are not remotely capable of programming anything at all.

    Which means that, in one or more of those cases, I recognise the prefixes (or “prefectures,” if you will, Fifi).

    And, you know what? I can confidently assert that no normal Windows program on Earth would ever go near them.

    Thing is, Fifi, I do this stuff for a living. You just randomly look it up on Google and hope it impresses the likes of a retired teacher in Manitoba.

    Have you considered chatting up female marsupials as an alternative? On the whole, they are very stupid indeed.

    Who knows, they might even be stupid enough to go out on a date with you.

  7. Dr Loser says:

    See xwayland

    Ummm … Robert … xwayland is just a way of adding yet another stupid dangerous layer below the standard X client cruft and the standard X server cruft.

    The whole point of Wayland is to get rid of this cruft.

    Still, whatever makes your belly-button palpitate, I suppose.

  8. Dr Loser says:

    Qt is odd. Its a API that can be High level abstraction ie above X11 or a low as win32/direct x abstraction(wayland server/clients) or decanted abstraction (kms/framebuffer) on Linux.

    Waiter: Did you enjoy your FLOSS meal, sir? The finest gourmet food that the world can offer, all for free!
    Cheapskate: Naturally I am entitled to such things. I am not a slave, you know. I am a fully paid-up blood sucking drone. Well, I say paid-up. I used a false address on the account. How do you guys make money?
    Waiter: Ah, M’sieur, or may I call you My Liege? Or possibly Tractor Boy? That is our Unique Servile Point. May I interest you in some apres-reality? Perhaps … a glance at the Abstractions on our Menu du Year Of Linux Desktop?
    Cheapskate: A quart of KMS or Framebuffer? I see where you guys make your money now. It’s on the mark-up for apres-diner service, isn’t it?
    Waiter: (smoothly) Quite so, Your Heavenly Majesty, quite so. But one has to admit that $50 for a fully-decanted Abstraction is quite the deal, is it not?
    Cheapskate: Oh well. I’ll just ask the wife for her credit card. She has a real job you know. We’ll paint the town red and go for the Framebuffer stuff! 2014 LTS Vintage, no less! Decant away!
    (Sounds of vigorous shaking, followed by a very loud fart.)
    Cheapskate: But, but … there’s nothing here but useless poisonous sludge!
    Waiter: Well, that’s pretty much what you expect when you decant a FLOSS abstraction, isn’t it?

  9. DrLoser wrote, “it’s hard to imagine FLOSS woolies taking Wayland to their hearts.”

    Wayland or something like it is required for some users of GNU/Linux who want better/faster graphics. I don’t have any objection as long as I can do my networked graphics, and there’s a driver for that…

    See xwayland

    As long as we are not locked out of doing what X does well, there should be no objection to getting better/faster graphics through Wayland.

  10. Dr Loser says:

    And here I feel the need to disagree strongly with Deaf Spy.

    Wayland may be a good idea, but is still vaporware.

    Wayland is clearly not a good idea. It’s a modern (proposed) solution to the top one of the top three present Linux systemic issues, which I would rank as:

    1. X
    2. SysVinit
    3. PulseAudio

    (And, entertainingly, PulseAudio was perpetrated by Poettering. What goes around, comes around, I guess.)

    Thing is, considering the amount of high-pitched screaming that systemd has generated (and given that there is a very simple migration path in that case), it’s hard to imagine FLOSS woolies taking Wayland to their hearts.

    Even if it does escape out of the standard Linux Condition and become actual release-quality software, as opposed to the endless betas that certain strange people like to faff around with.

    A theoretical good idea? Maybe. But a practical good idea? Very unlikely indeed.

  11. Dr Loser says:

    And, before you go there, please spare me the chroot clusterfuck.

    I wish I was as assiduous as Kurks. I have no thesaurus of oiaohm lunacies, but for some reason I seem to remember him advocating a combination of SELinux and chroot at one point.
    I’ll happily accept that he didn’t do so, however — the very idea is absurd. It was probably some other nutter.

  12. Deaf Spy says:

    Remember kurkosdr was also complaining about Linux API’s being unstable. Reality there are Windows API’s in just as bad of state.
    There is a slight difference, Ohio. On Windows, APIs are stable enough to let you run Chrome 43 on Windows XP, or to run Chrome 1 on Windows 8.1. Just like that. No need to use even compat mode. Try do that in Linux and watch the fun. And, before you go there, please spare me the chroot clusterfuck.

    Really comparing Qt to Flash, or Silverlight shows you are a idiot.
    That is very brave for someone who’s always been proven the idiot all over the Internet. Ah, and it is possible to develop an OS, based on Silverlight.

  13. oiaohm says:

    Deaf Spy the point here is a lot of API that kurkosdr refereed to on Linux were do not use directly as well. Remember kurkosdr was also complaining about Linux API’s being unstable. Reality there are Windows API’s in just as bad of state.

    Actually, I was mocking at your notion to promote Qt to a OS API. It is not. It is just a third-party library you may install and build upon, just like Flash, or Silverlight.

    This is where things get really strange. Are you aware that Qt applications can choose to run on frame-buffer or KMS directly under Linux. So able to completely bi-passing X11. So can Flash or Silver-light completely do this. Qt contains the frameworks to make your own Wayland compositors.

    So yes Qt is a third party library but it not exactly what you think it is. Remember glibc is also a third party library.

    Qt is odd. Its a API that can be High level abstraction ie above X11 or a low as win32/direct x abstraction(wayland server/clients) or decanted abstraction (kms/framebuffer) on Linux.

    GTK is also heading into this form on Linux.

    Really comparing Qt to Flash, or Silverlight shows you are a idiot. Comparing to Java might have been workable as Java OS has existed.

  14. Deaf Spy says:

    Well done, Ohio, you’re improving. Finally a valid, meaningful reference. Obsolete, archived, but hey, a real reference! I hope you keep up the good work, and progress, and refrain from ill fantasies.

    Now, let me bring your attention to the very first sentence of your own reference:
    These functions are subject to change with each operating system revision.

    In other words, Fifi, these functions may constitute an API only in the world of Linux, where API and ABI compatibility is never important. In the sane world, the sentence above means – do not use.

    Actually, I was mocking at your notion to promote Qt to a OS API. It is not. It is just a third-party library you may install and build upon, just like Flash, or Silverlight.

    Wayland may be a good idea, but is still vaporware.

  15. oiaohm wrote, “Smoke makes what is called buckyballs that can get inside harddrives and jam under the head of the driver(not particularly good) also happen to be highly conductive so layers smoke on circuit boards can cause short circuits.”

    Yep. The dirtiest PC I’ve ever seen in my life had road-dust glued together on fan-blades and heatsinks with tobacco-tar. It had to be scraped off. A wipe did nothing.

  16. oiaohm says:

    While Desktop Linux requires neckbeards with crumb-infested facial hair and an unhealthy addiction to stale coffee and cigarettes.

    kurkosdr Please attended a Linux conference. Yes you might see facial hair, yes you might see people with a coffee problem but there is one thing you will never see is cigarettes. You might if you are really lucking see a cigar being eaten.

    This is very much a highly interesting difference between attending a Linux conference and a Microsoft one. Microsoft one you see people going out side for a smoke. Even journalist attending Linux events have learnt not to smoke. Why Linux people smell smoke you will be treated like a leaper.

    Even the Unix guys of old almost never smoked. This is not about personal health. This is about health of the electronic equipment.

    Smoke makes what is called buckyballs that can get inside harddrives and jam under the head of the driver(not particularly good) also happen to be highly conductive so layers smoke on circuit boards can cause short circuits.

    Here is the bad thing you smoke a cigarettes today so you don’t breath out buckyballs you have to wait 14 days to be free of it.

    You can spot new reporter to a Linux Conference who smoke. You watch the crowd leave a circle of clear space around them. They also normally don’t get send again because they don’t get any interviews.

    The avoidance to smoke by Linux personal does cause problems when the CEO and the like smoke. Because they will do anything to avoid being in the same room as them.

    You could say Linux personal lives in the strictest no smoking culture. If you employ person to manage Linux that smokes you have made a major mistake because there is no point sending that person to a conference.

  17. oiaohm says:

    Deaf Spy of course you did not read my link. NTgdi.

    https://msdn.microsoft.com/en-us/library/ms648408%28v=vs.85%29.aspx
    This notes about the NTgdi functions. Yes windows 2000 SDK and before provided a library to build the syscalls into your application.

    Anything that is a syscall is handled by the executive.

    http://en.wikipedia.org/wiki/File:Windows_2000_architecture.svg

    Basic NT design. You cannot access the GDI parts in kernelspace without going threw the executive. Can you create a window and everything else just doing Windows calls to the executive yes you can.

    The reality here is low level functions are wrapped behind libraries on Windows all the time. My example from MSDN show direct syscalls being wrapped behind D3D8THK.DLL.

    Yes you can draw a graphical application under windows without the win32 subsystem loaded.

    Even after you micro-hybrid NT kernel fiasco, you simply didn’t progress at all. I would be very disappointed of myself if I were you, Fifi.
    Sorry Deafspy you should have know from that past debate what Microsoft defined as Executive. All syscall accesses go by the Executive under windows even if they are redirected back to user-space somewhere. This is very much the same as the Linux kernel does. Yes particular syscalls to the Linux kernel redirect back to the libc of the system.

    Basically Deafspy you are creating another fiasco by not knowing facts again. Instead you want me to provide cites when I am correct. Sorry you have just lost the right to ask for cites from this handle as well.

  18. kurkosdr says:

    they still need to hire executives to run That Other OS

    While Desktop Linux requires neckbeards with crumb-infested facial hair and an unhealthy addiction to stale coffee and cigarettes.

  19. Adam Queen says:

    kurkosdrloser95 its because M$ is using outdated technology , they still need to hire executives to run That Other OS , which is why their slower. GNU/Linux does all of it automatically , it has more power to user like the Bash.

  20. kurkosdr says:

    BTW, where does Microsoft gets those names? “Executive”? Does it also have a secretary and sniff nose candy?

  21. Deaf Spy says:

    Yes Windows NT Native API has a Wikipedia page.
    You still don’t understand the architecture of Windows NT, do you? Even after you micro-hybrid NT kernel fiasco, you simply didn’t progress at all. I would be very disappointed of myself if I were you, Fifi.

    NT native API is provided by the executive. The graphics subsystem is not a part of the executive. There is no native API for graphical apps. Unless, of course, you can provide a reference in MSDN.

  22. oiaohm says:

    Now, are we talking GDI, or DX, or WPF, or WinRT? Or do you mean some magic undocumented API born of your funny imagination, Fifi?
    By the way Deaf Spy is another stolen identity. It is possible to bring in the evidence of that as well. It is what TMR did bully people until they stop posting on-line then take over their identities. Yes Oldfart calls me a lier and fraud yet he was hanging around with a pack of worse. About time Oldfart as the one claiming a solid identity do the pull up this time.

    Windows NT Native API is what real DeafSpy use to code in. Not knowing what this is proves the identity as stolen. http://en.wikipedia.org/wiki/Native_API
    Yes Windows NT Native API has a Wikipedia page.

    The reality is almost no Windows application is coded using just the Windows Native API. It is fairly simple to detect applications using just Native as all functions start with one of the 8 prefexes. Yes all GDI calls should beand NtGdi and all user functions should be NtUser those are both are under the NT prefex. .

    Just like you don’t find applications coded using xlib or xcb or other low level interfaces directly on Linux the reality is you don’t find Windows applications using the low level interfaces of Windows in major amounts either.

    Just like pulseaudio and other things in FOSS having questionable documentation “Native API is not fully officially documented” is noted on the Wikipedia page. Fair comparing you find out that FOSS and Closed source are not much better or worse than each other.

    Really why should I respect and not insult a person who is using a stolen identity.

  23. LinuxGentlesir says:

    “Never” implies a strong religious belief unsupported by the steady growth of GNU/Linux on desktops and everywhere else.

    It’s wishful thinking by the opponents of human progress. It is a sure signed evidence that if they truly believed what they say they wouldn’t feel the need to combat you.

    The future is open!!

  24. Deaf Spy says:

    Windows desktop application using Windows NT Native API todo everything.

    Now, are we talking GDI, or DX, or WPF, or WinRT? Or do you mean some magic undocumented API born of your funny imagination, Fifi?

  25. Deaf Spy says:

    Software is text, usually, or bits somewhere, you know, information, something other than chaos.

    F.

  26. oiaohm says:

    kurkosdr remember X11 protocol has got smaller in recent years(last 15). Why the X11 protocol got some management and starting deleting never used sections from the standard.

    Wayland is rewrite that the FOSS developers with all commercial developers bar Nvidia that X11 was just so defective at a standard level that it was not fixable. Please note Wayland will be able to run X11 applications.

    Qt is just as much of a FOSS API as X11 is. Only different is QT has always had good management.

    kurkosdr tell the the last time you saw a Windows desktop application using Windows NT Native API todo everything. The answer is you don’t. QT on Linux is like what Win32 provides on Windows.

    Kurkosdr you need to learn to keep mouth shut on some things.
    https://www.adobe.com/devnet-apps/photoshop/fileformatashtml/

    Photoshop in fact uses Open Standards. Central management by Adobe but still an open standard. Adobe is also working on opening up Autocad formats.

    So Kurkosdr so one example of closed format you picked is 100 percent bogus and the other one is on very much the way of getting there. Reality Adobe wants government contracts using Open Formats is a requirement to getting lots of those contracts today.

  27. antifanboy wrote, “You said the same thing in 2012.”

    Yes. GNU/Linux has had many good years. Since that year many millions have found software freedom. 2015 will make 2012 seem like a “continental breakfast” compared to gluttony at McDonalds. I think growth has been nearly as great this quarter as in all of 2012. There’s still room for growth of course and the economy of GNU/Linux desktops could catch fire in which case it will be the Decade of the GNU/Linux Desktop… For me it was back in 2000 and I changed the lives of dozens that year. I’ve changed the lives of thousands since then by introducing them to FLOSS. Your fear is that the world will clone me a few thousand times… Oh, apparently that’s been done. I think we are heading to 100K “me”s in the world. Then there’s LibreOffice, Apache, VLC, and a ton of other software freeing the world from slavery. Monopoly is all but gone. Thank Goodness.

  28. antifanboy, claiming I am “religious” about IT, wrote, “will never happen”.

    “Never” implies a strong religious belief unsupported by the steady growth of GNU/Linux on desktops and everywhere else.

  29. kurkosdr wrote, “true “freedom over slavery” can be achieved by using open standards”.

    That’s not true at all. e.g. someone can own your data, software and even your computer if you can’t run, examine, modify and distribute the software. That other OS is a closed standard. So’s it’s EULA from Hell. No one has the right to tell me how many fewer networking connections my PCs can have below their technical capabilities. No one has the right to tell me they can install whatever crap they want on my computers. I am a free man, not a slave to Wintel.

  30. dougman says:

    Re: Of course, open standards isn’t always possible. See Photoshop and AutoCAD files.

    DO tell, how many people actually use these products?

    Besides that, both products offer cloud versions anyway. So anyone from ANY platform can utilize them.

    Open Standards benefits everyone. One single entity cannot lock-up or control it. This is why lots of people hate Apple, as they use proprietary connects for their devices.

    https://www.youtube.com/watch?v=jyTA33HQZLA

  31. kurkosdr says:

    Enough with the “FREEDOM over slavery” religious political ideology bullcrap that Stallman feeds you!

    And anyway, true “freedom over slavery” can be achieved by using open standards, not by the elevation of a technical instrunment (source code) to a holy item status. This is why so many companies ask for JavaEE programmers. And why new business apps are coded as Java apps (despite the documented shortcomings) or as Qt apps instead of using whatever is the FOSSie API of the year (X.org, Wayland, ALSA, PulseAudio etc).

    Of course, open standards isn’t always possible. See Photoshop and AutoCAD files.

  32. antifanboy says:

    PS: Enough with the “FREEDOM over slavery” religious political ideology bullcrap that Stallman feeds you! “The Year Of The Linux Desktop”never happened in the past and will never happen in the future.

  33. antifanboy says:

    You said the same thing in 2012. http://mrpogson.com/2012/03/22/2012-will-be-the-year-of-the-gnulinux-desktop/ Don’t play dumb!

  34. I hate to rub it in, but antifanboy wrote, “This year is no different.”

    He can read it and weep, too. BG’s master plan to put one copy on every PC in the world has failed miserably, even in his own back yard:

    Where have you failed, M$, that millions of USAians choose FREEDOM over slavery? USA is believed to have 254million connected people. 2.8% of that would be 7.1 million FREE people and probably several million FREE desktop PCs.

  35. Deaf Spy wrote, “Software is an information-processing tool.”

    Software is text, usually, or bits somewhere, you know, information, something other than chaos.

  36. Deaf Spy says:

    Do you know that software is information?

    Incorrect, Mr. Pogson. Software is an information-processing tool. You indeed have no clue what an information system is. How can you even dare preach what software people should use?

  37. antifanboy says:

    Let’s face it: EVERY YEAR from 1998 has been the “The Year Of The Linux Desktop”(not GNU/Linux. Who are you, one of Richard Stallman’s sheep?) and EVERY YEAR, wishful thinking Linux Nazis like yourself are proven WRONG. This year is no different. Get over it.

  38. kurkosdr says:

    Do you know that software is information?

    Do you know that everything is made of stardust?

  39. Deaf Spy wrote, “Do you know what an information system is?”

    Do you know that software is information?

  40. Deaf Spy says:

    To the extent that this blog (coded in HTML) and in particular my pretty graphics of web-stats (SVG) are software distributed to the world

    Can you tell the difference between “software” and “service”, Mr. Pogson? Do you know what an information system is?

  41. DrLoser, attempting to move goalposts again, wrote, ” You, Robert, cannot and do not and never once have “made World Software and World Hardware without monopoly.””

    To the extent that this blog (coded in HTML) and in particular my pretty graphics of web-stats (SVG) are software distributed to the world, DrLoser is wrong again. Technically, I hold copyright by default on much of this but I permit anyone to use, copy, distribute my stuff without objection. I’ve also distributed some ballistics software for many years. I know folks have used that because there has been discussion of the software between me and users. I am a professional in several ways: I have BSc (Hons) in Physics, M.Sc. and Cert. Ed. all being professional degrees, I was a member of IEEE back in the day, I am a published authour in peer-reviewed scientific journals and I was a certified teacher in several jurisdictions in Canada. Developing software was just a part of my accomplishments but I was paid to write software for many years, mostly in science and statistics but also in the field of education. Some of my writing was about hardware that I developed and distributed to the world freely.

  42. oiaohm says:

    On the other hand, those extensions helped Silicon Graphics make an X-based system that worked (IRIX), and brought in money before SGI commited ritual suicide anyway.
    kurkosdr you missed the closed source Unix vendors were not application compatible even at the source level. GNU made autotools to attempt to mask over the massive number of platform differences.

    The X11 fragmentation limited IRIX and other Unix OS market penetration. So applications written for IRIX X11 did not work with other X11 servers. So the disaster mess of X11 starts even before Linux starts from companies like SGI. SGI is the first company to work out that these Unix unique graphics extensions to X11 was stupid. IRIS GL from IRIX comes Opengl. But of course that never fixed the mess of having to support stacks of other graphical solutions for application compatibility.

    X11 is really example of if you keep backwards compatibility with everything you end up with something completely worthless.

    X11 mess and the Audio mess of sound servers and subsystems starts before Linux exists.

  43. kurkosdr says:

    Closed source X11 servers added there own extensions lot of cases independent to each other this is why X11 has 4 ways of doing copy paste.

    On the other hand, those extensions helped Silicon Graphics make an X-based system that worked (IRIX), and brought in money before SGI commited ritual suicide anyway.

  44. oldfart wrote, “he would not have even been able to integrate his Beast from white box crap parts had not commercial companies expended the money to fund the R&D for the original hardware that the Taiwanese and then the Chinese copied and commoditized.”

    Uh, Intel, IBM and others cooperated in the development of ATX so that they could sell product in volume with multiple suppliers for all hardware… They intended the copying and commoditization. It was and is good for everyone. Just ask all those who switched from UNIX OS on Sparc to GNU/Linux on Intel/AMD. Competition is good and allowing competition in the market for PC and server was one of IBM’s best moves. Unfortunately they granted M$ a monopoly on the OS… That nearly nullified the openness on the hardware side. Fortunately the world has moved on and the good times are rolling.

  45. oiaohm says:

    horrible as horible is correct as per Australian dictionaries.

    The MIT license is intentionally set up as an academic license, rather than something dreamed up by a malcontent dipstick living in a broom-cupboard.

    As “co-operation” goes, it seems to work very well indeed. Stress- and Lawyer-free, indeed. I offer, once again, OpenCV as a Poster Child for the MIT licence.

    You’re just being an imbecile again, aren’t you, Fifi?

    No it you being imbecile. As you stated MIT was created as an academic license not as a license idea for a standard.

    What happens in X11 history is horible. Closed source X11 servers added there own extensions lot of cases independent to each other this is why X11 has 4 ways of doing copy paste. So as bad it sounds a huge number of the screwed up parts of X11 directly relate to closed source developers not FOSS developers.

    Remember when X11 started all it had was the MIT license no charter no core body…. LGPL and GPL and other viral class works without a centeral body do seem todo better.

    Even opencv is showing closed source forks causing duplication in functionality that has to be emulated.

    DrLoser your poster child check on it in 20 years time and it might have developed issues like X11. MIT license alone for something that need to be standard is 100 percent not good enough. GPL/LGPL maybe but still a proper standard is better.

  46. LinuxGentlesir says:

    And he would not have even been able to integrate his Beast from white box crap parts had not commercial companies expended the money to fund the R&D for the original hardware that the Taiwanese and then the Chinese copied and commoditized.

    And surely if you go back far enough you can implicate abiogenesis.

  47. LinuxGentlesir says:

    These so called “professionals” are making hundreds of posts on your blog and apperently do so for free. Or are they astroturfers from Microsoft?

  48. oldfart says:

    “My rebuttal is equally simple. You, Robert, cannot and do not and never once have “made World Software and World Hardware without monopoly.””

    And he would not have even been able to integrate his Beast from white box crap parts had not commercial companies expended the money to fund the R&D for the original hardware that the Taiwanese and then the Chinese copied and commoditized.

  49. oldfart says:

    “Farting oldman… ”

    So tell me Queenie, was your boyfriend on the rag last night…?

    Or is that Dougie in drag?

  50. DrLoser says:

    Nonsense. I’ve been making stuff since I was a young lad.

    Precisely what that noble claim has to do with either World Software or World Hardware escapes me at the moment, I’m afraid.

    Your proposition is simple. “The world can and does make its own software and hardware without monopoly.”

    My rebuttal is equally simple. You, Robert, cannot and do not and never once have “made World Software and World Hardware without monopoly.”

    The fact that you can build a nifty granite kitchen work-top is neither here nor there. You could probably sell that skill, should you so choose.

    But you cannot sell your ability to build World Software or World Hardware. Mostly because that ability does not exist.

    Which is why people pay professionals to build the stuff for them.

  51. DrLoser wrote, “the fact is, you have never done either, have you?”

    Nonsense. I’ve been making stuff since I was a young lad. I’ve made a computer from components, not just that ATX stuff, but 40 pin microprocessors, and other ICs in sockets on a wire-wrap board. I built a lot of gadgets as a student of physics, stuff you could buy if you wanted to wait years to get it into a budget and have it shipped from elsewhere. It was faster and cheaper to build in those days. I still build stuff by working in wood and steel. I have no idea how much code I’ve written over the years but many years of my life was spent doing that, mostly for number-crunching and graphing the results. My Master’s Thesis was about building a gadget to locate and identify fast-moving charged particles.

  52. Adam Queen says:

    Let’s not forget farting oldman , that people will start to notice that That Other OS is slowing down and bloating as new releases of it comes out ( proof: http://blog.zorinaq.com/?e=74 ). Do you think people want that? As I say before, even M$ doesn’t want that , but they are too little too late to fix the sinking Titanic of That Other OS. You can already see this happening in M$’s cloud and desktop market share. These are facts oldfart , you can’t go against that.

  53. Adam Queen says:

    “The question is why does Robert Pogson feel he is entitled to dictate what a business sells its wares at?

    The commercial world does not owe Robert Pogson cheapskate anything!”

    Farting oldman take a look at what most users want: anything but M$. M$’s cloud market share (shown in one of Robert’s comments) and even desktop market share is only shrinking because of that. What does that tell you Farting oldman? It tells you that most people would agree with Robert. Way to call most people “cheapskates”!

  54. DrLoser says:

    kurkosdr X11 is not a particularly good example. X11 is a MIT licensed project. This turns out to be a fairly horible idea for cooperation.

    Bollocks, Fifi. By the way, there are two Rs in “horrible.” I live to serve.

    The MIT license is intentionally set up as an academic license, rather than something dreamed up by a malcontent dipstick living in a broom-cupboard.

    As “co-operation” goes, it seems to work very well indeed. Stress- and Lawyer-free, indeed. I offer, once again, OpenCV as a Poster Child for the MIT licence.

    You’re just being an imbecile again, aren’t you, Fifi?

  55. DrLoser says:

    The world can and does make its own software and hardware without monopoly.

    I wish you’d stop repeating yourself on this particular piece of soap-box drivel, Robert.

    Because, the fact is, you have never done either, have you?

    Creating your own recipe database does not count as World Software. And stitching Beast together does not count as World Hardware.

    The world can and does BUY its own software and hardware without monopoly.

    Realistically, Robert, if you are going to talk about “the world,” then this is what “the world” does. This sort of commercial activity is pretty much why a fiat currency exists.

  56. DrLoser says:

    I’ve often used esound to relay sound to thin clients. It was solid. I never had it fail.

    Failure is such an objective measure, isn’t it, Robert?

    It’s long past your bedtime as a commercial programmer for the likes of Ford, so you probably haven’t experienced the recent joys of video conferencing via, say, TeamViewer.

    I can confidently assert, having been through at least three commercial organisations using this modern stuff, that at least 10% of the average meeting is consumed by “tuning” the “thin client audio and video.” This stuff is still in the Bronze Age.

    And the fact that nobody in their right mind chooses to use esound for business purposes might alert you to the fairly obvious fact that FLOSS equivalents are not stuck in the Bronze Age.

    Unfortunately, they are stuck in the Stone Age. Now, where did I put my CLI? Cancel that Gnome window. Bring another one up! Check out the informative yet strangely inaccurate man page. Oh look, I need to apt-get some idiot library that was only updated yesterday!

    It’s all good.

  57. DrLoser says:

    The result, Intel E2-2650 got a bit more than twice the performance per chip but costs ~$1000 per chip.

    This is a fabulously interesting piece of analysis for those of us who wish to string together dozens of the things in order to build a supercomputer in our garden shed, Robert.

    On the other hand, it’s of no consequence whatsoever to the likes of me, thinking of building our own home server for, say, $800 max. And that would include a lot more than a CPU, obviously. Let’s go Hog Wild and choose two Intel work-horses on the assumption that I can find a suitable dual motherboard: the I5-4590.

    That’s going to cost me around $400, possibly plus 10% if I choose to accessorise with my own heat-sink stuff. Leaving me $400 for the case, the PSU, the mother board, the RAM and the disks.

    I think I’ll upgrade to $100 for a decent case. I can live with a small SSD for the boot disk. I can probably get 64GB at a knock-down price. I’d like something reliable for RAM, which means Crucial or similar, and that’s going to cost me … well, that’s not cheap right now, either. $200 for 8 gig, multiplied by two?

    Piecing all of this together, Robert, it will cost me something between $1200 and $1500 to build what I would consider to be a reasonable server. And if I wanted to go cheap on those costs?

    I don’t really think that buying a couple of inferior CPU chips and saving myself $100 in the process comes into the equation at all.

    Then again, I could probably repurpose my $250 home desktop as an alternative to Beast. So I’m probably talking to the wrong person here.

  58. oldfart wrote, “The commercial world does not owe Robert Pogson cheapskate anything!”

    I’m not demanding anything from Intel. I try to avoid their parts.

    The world does not owe any monopoly servitude. The world can and does make its own software and hardware without monopoly. The USA, oldfart’s home country, claims to be a champion of free enterprise. Why then do he and other USAians worship monopoly? It makes no sense. The world will move on and leave their archaic conceptions of IT in the dust.

    A quick fact-check:
    CERN did a benchmark of ARM v Intel in HPC. The result, Intel E2-2650 got a bit more than twice the performance per chip but costs ~$1000 per chip. The ARM A57 fits in a smartphone and costs less than $100. The Intel chip got about same performance/watt. So, ARM is competitive in HPC as far as CERN is concerned:
    “Our initial validation has demonstrated that APM X-Gene 1 Server-on-Chip ARMv8 64-bit solution is a relevant and potentially interesting platform for heterogeneous high-density computing. In the absence of platform specific optimizations in the ARMv8 64-bit GCC compiler used, APM X-Gene 1 shows excellent promise that the APM X-Gene hardware will be a valid competitor to Intel Xeon in term of power efficiency as the software evolves. However, Intel Xeon Phi is a completely different category of product.
    As APM X-Gene 2 is being sampled right now, built on the TMSC 28nm process, we look forward to extending our work to include it into our comparison.”

    Intel is having to compete on price/performance just as M$ is now. That’s a refreshing change.

  59. oldfart says:

    “The question should have been, “Why should customers pay that much?”.”

    The question is why does Robert Pogson feel he is entitled to dictate what a business sells its wares at?

    The commercial world does not owe Robert Pogson cheapskate anything!

  60. oiaohm wrote, “I guess you never used artsd or esound”.

    I’ve often used esound to relay sound to thin clients. It was solid. I never had it fail.

  61. Deaf Spy wrote, “Intel routinely charges more than 2.5 times the cost of production of chips

    “And why shouldn’t they?”

    God was reported to be content with a tax of only 10%. AMD and ARMed devices have a much lower margin so a customer is closer to getting that for which he/she pays. The question should have been, “Why should customers pay that much?”.

    tithe: ” A tenth; the tenth part of anything; specifically, the tenth part of the increase arising from the profits of land and stock, allotted to the clergy for their support, as in England, or devoted to religious or charitable uses.”

  62. Adam Queen wrote, “Backwards compatibility is not an advantage they have anymore.”

    It may be an advantage clinging to some shrinking slice of the pie but there’s no growth in it. M$ will have to work for a living in the cloud and when users see the price of M$’s service on their billing they will have a reason to make choices. That wasn’t so in the Dark Ages of monopoly. Indeed, backwards compatibility will sooner or later hold M$ and M$’s customers back.

    See Amazon Increasingly Duels Microsoft, Google in Cloud

    See Amazon still dominates the $16 billion cloud market

  63. Adam Queen says:

    Do not forget kurkosdr that we can’t stick with M$’s OS either as their boat is sinking as the bloat and performance is worsening. kurkosdr M$ themselves said they’re OS is dying too , and when you look at Linux , they’re constantly improving from before since they don’t need to satisfy customers with old software. Instead Linux is moving to the cloud so that customers can update with the package manager while cloud services constantly move forward. What has M$ got? Nothing. Backwards compatibility is not an advantage they have anymore.

  64. Adam Queen says:

    kurkosdr you don’t understand that old programs That Other OS is backward compatible with do not live forever. Once these old programs die out , GNU/Linux is a better and free alternative. I’m not as expert as Robert or oiaohm , but services and cloud computing doesn’t need backwards compatibility , and Linux is dominating that. That Other OS is not because backwards compatbility is holding them back, and Backwards compatibility only makes you more vulnerable to attacks.

  65. Deaf Spy says:

    Intel routinely charges more than 2.5 times the cost of production of chips

    And why shouldn’t they?

  66. oiaohm says:

    kurkosdr X11 is not a particularly good example. X11 is a MIT licensed project. This turns out to be a fairly horible idea for cooperation.

    PulseAudio is not as big as disaster as the items it replaced. Kurdosdr I guess you never used artsd or esound.

    Something to remember Linux kernel has had very limited IPC options.

    kurkosdr problem with your arguement is that ISV have been releasing on Linux since before the year 2000 without major issues. These are applications like Maya and Cad software of course these have been careful enough to ship with there own runtimes.

    Sorry Kurkosdr most of you arguments have been bogus. Main problem is the bogus has caused ISV’s to be unable to find the correct solutions.

  67. kurkosdr, not understanding the reason for electronic IT, wrote, “Desktop Linux COULD have been the future if it wasn’t for the poisonus nature of the FSF culture. The one that pisses on ISVs and IHVs by ruining backward and forward compatibility, just for the fun of it. “

    That argument is just backwards. The role of ISVs and IHVs in the economy is not to give the world what M$ wants but to give the world good IT at competitive prices. Wintel was all about maximizing the cost of IT, not competition. That’s what monopoly does. FLOSS eliminates monopoly and is the right way to do IT.

    e.g. Intel routinely charges more than 2.5 times the cost of production of chips. M$ routinely charges more than 2.5 times the cost of production of software.

    Meanwhile FLOSS charges the world what it costs to produce software.

  68. kurkosdr says:

    M$ fanboys just please stop , M$ themselves admitted that */Linux is the future , not their OS.

    Desktop Linux COULD have been the future if it wasn’t for the poisonus nature of the FSF culture. The one that pisses on ISVs and IHVs by ruining backward and forward compatibility, just for the fun of it.

    FOSSies don’t understand that the reason there are Linux successes like Firefox, XBMC and MPC-HC, is because they capitalize on existing stable, backward compatible standards (HTML, HTML5, MP4, MP3 Divx AVI etc).

    When FOSSies try to define their own standard, it ends up in disaster, like PulseAudio or modern X.org.

  69. dougman says:

    I say Win-Dohs is dead, the rise of Chromebooks is upon us!

    This is one school district that wishes they used Chromebooks.

    http://www.coinbuzz.com/2015/03/25/500-bitcoin-ransom-entire-school-district-shut-down/

  70. dougman says:

    LOL…I remember that article too funny.

  71. Adam Queen says:

    The future of */Linux is bright 🙂 . No more running on That Other unmaintainable OS. While That Other OS is is bogging down and bloating , officially said by M$ ( http://blog.zorinaq.com/?e=74 ) , */Linux is succeeding in desktop market share. M$ fanboys just please stop , M$ themselves admitted that */Linux is the future , not their OS.

  72. oiaohm says:

    DrLoser In the last 10 years. Redhat and Ubuntu has changed their Init system 3 times. Suse 4 times.

    Systemd is the first time the replacement init system is not 100 percent sysvinit compatible.

    “Systemd can be replaced.”

    But, realistically, won’t be.
    No DrLoser Systemd will either shape up or be replaced like all the priors have been. It will take 3 to 4 years to know what way its exactly going.

    Robert
    The design is flawed in that its way too complex to ever be secure. It mixes too many diverse functions together, like some desktop applications we don’t need. That’s not what you want controlling your whole system. I expect, but hope it doesn’t happen, that GNU/Linux will have “waves of malware” based on systemd thanks to this mess.
    You have this badly wrong. Linux has had exploit and malware problems.

    Linux kernel is complex 600+ exported functions. Sysvinit disregards all the security problems

    https://people.debian.org/~stapelberg/2013/06/09/systemd-bloat.html

    Robert Pogson turns out each application inside systemd is fairly cleanly designed. Yes systemd is a lot of tools in one package. Design the tools to correctly manage the Linux kernel is not exactly simple either.

  73. DrLoser says:

    To put it very bluntly, Robert, you are just not very good at this stuff, are you? In fact you never have been.

    And when it comes to any new idea, even if that new idea appears within the framework of Gnu/Linux and FLOSS, you’re not awfully good at working with this “innovation” stuff, are you? Despite your repeated observations that “innovation” is what FLOSS and Linux are all about.

    Never mind. We should all join you in the Manitoban Dark Ages. After all, it works for you.

    How could anybody ask for more?

  74. DrLoser says:

    The design is flawed in that its way too complex to ever be secure.

    The design is not “way too complex.”

    Just because you can’t deal with the rather simple ways of writing a declarative init system, Robert, does not make it “way too complex.”

    And, as a proponent of outward-facing FTP sites, you are in no position whatsoever to comment on “security.” In fact, I can’t even begin to imagine how a properly organised declarative init system could be accused of “inherent security issues.”

    No doubt, despite your pitiful attempts to make the thing work at all on Beast, you are in a position to contradict me with at least one “security violation” that has cropped up via your rigorous amateur at-home audit process.

    Well, I say, “no doubt.” I hereby forego Socratic Ignorance.

    You don’t stand the remotest chance of doing so, do you, Robert?

  75. DrLoser says:

    Apparently everyone in China has a relative who distributes CDs of that other OS…

    Oh, I’m sorry, Robert. I assumed that by “special relationship,” you were talking about a state-to-state thing. Such as the “special relationship” between the USA and the UK.

    But apparently you’re just talking about copying freebies. And you admit that China doesn’t have any more of a “special relationship” than, say, Vietnam and Uzbekhistan.

    What a thrilling “special relationship” that must be.

    On the other hand, if a billion Han have the free choice of pirating a copy of XP or else installing an equally zero-cost Linux Desktop …

    I’ll leave you to join the dots up there, Robert.

  76. DrLoser wrote, “the details of the “special relationship” between the PRC and Voldemort would be, Robert?”

    Apparently everyone in China has a relative who distributes CDs of that other OS… They still have 30% share for XP, for example. The only neighbour near that is Viet Nam at 33%. Even Uzbekistan only has 26%. There are only a few countries with more usage of XP:North Korea, Vietnam, Cuba, Eritrea, Cook Islands…China is indeed special that way. One of the world’s most advanced countries with a huge population and booming economy and XP…

  77. DrLoser wrote, “Your evidence for it being a “poor design” is that you can’t get it to work?” and “Systemd can be replaced.
    But, realistically, won’t be.”

    The design is flawed in that its way too complex to ever be secure. It mixes too many diverse functions together, like some desktop applications we don’t need. That’s not what you want controlling your whole system. I expect, but hope it doesn’t happen, that GNU/Linux will have “waves of malware” based on systemd thanks to this mess. In the meantime, I’ve figured out how to get it to work so I can ignore it thereafter. I have a feeling that Poettering won’t leave me alone. I expect RedHat will see the error of its ways and either fix or replace systemd. Their customers will not accept buggy software controlling their systems.

  78. DrLoser says:

    I still think it’s a poor design but not lethal.

    Your evidence for it being a “poor design” is that you can’t get it to work? That’s not really evidence, Robert. Ten years ago, you couldn’t get Windows Networking to work. Today, you can’t get the next generation of pretty much every Linux distro out there to work.

    Somehow I don’t think this is about commercial vs FLOSS software. It sounds more like unavoidable personal incompetence as a SysAdmin to me.

    Systemd can be replaced.

    But, realistically, won’t be.

    The only people earning real money out of corporate customers who need an init system that, say, matches what Microsoft could do in the 1990s are Red Hat.

    And if you think that either Red Hat or their commercial “competitors” (Ubuntu or SuSE) are going to abandon systemd just because you, the most important Winnipeg Citizen in the History of FLOSS, sez so …

    … You’re Demented.

  79. DrLoser says:

    And the details of the “special relationship” between the PRC and Voldemort would be, Robert?

  80. oiaohm says:

    Really you have to remember Microsoft releasing on Android has caused companies to reach into pocket and fund this.
    http://vmiklos.hu/blog/tiled-editing-part-4.html

    So there is now money to make a Libreoffice for Android version. Libreoffice connects to Sharepoint and Microsoft on-line services without having to pay a extra fee. So any extra income Microsoft is making from the Android version of MS Office is extremely limited.

    Libreoffice has the advantage of open source so governments will be able to audit and build their own android APK files.

    Gosh, too bad MS, err, I mean M$, just inked deals to preload their Office Apps on Android. So now they will profit even more from Android.
    Good question I don’t think the ODM’s will be paying Microsoft anything. It will become more that Microsoft will have to be like a crapware vendor and pay the ODM to be installed.

    Money can buy yourself shelf space in a market but Money on it own will not keep your customers. Why if you are buying shelf space and no one buys you are still out of that money.

  81. Agent_Smith wrote, “friendly fire nearly killed it. Or, will kill.”

    Systemd may eventually work. I still think it’s a poor design but not lethal. Systemd can be replaced. It’s a FLOSS environment, you see, no lock-in.

  82. Rshack wrote, of M$ and Android, “now they will profit even more from Android.”

    I don’t think M$’s office suite on Android/Linux has any particular competitive advantage. There’s not much use for it on a smartphone, for instance, unless docked. Tablets, maybe. Libreoffice will more or less be a competitive product at $0, so M$ will have to keep its price down and take revenue from “the cloud”. I’m all in favour of M$ competing on price/performance. They didn’t need to do that with their OS from the 1980s until a few years back. On Android/Linux, competition will be fierce on Day One. Today, we know there is no “natural monopoly” as M$ used to claim. Expect to see them issue a client for iOS and GNU/Linux too.

  83. Rshack says:

    Gosh, too bad MS, err, I mean M$, just inked deals to preload their Office Apps on Android. So now they will profit even more from Android.

    I don’t think this is the Linux revolution that Stallman had in mind.

  84. Agent_Smith says:

    Robert,

    I’m sorry to say that RH spoiled that. Systemd damaged Linux beyond repair. It could have been a winner, but, friendly fire nearly killed it. Or, will kill.

    Regards,

  85. DrLoser says:

    Most of the major countries have left ~1% share of page-views behind with the exception of China which has a special relationship with TOOS.

    And the details of the “special relationship” between the PRC and Voldemort would be, Robert?

    C’mon. Don’t leave us all in suspense. The future of a billion slaves awaits your magnificent wisdom!

Leave a Reply