Kubuntu Deployments: Woah! I didn't know this!

[woodsmoke]

Deployments

Kubuntu rollouts include the world's largest Linux desktop deployment, that includes more than 500,000 desktops in Brazil (in 42,000 schools of 4,000 cities).[43][44][45][46]

The software of the 14,800 Linux workspaces of Munich was switched to Kubuntu LTS 12.04 and KDE 4.11.[47][48]

The Taipei City government decided to replace Windows with a Kubuntu distribution on 10,000 PCs for schools.[49] [50]

The French Parliament announced in 2006 that they would switch over 1,000 workstations to Kubuntu by June 2007.[51][52]

Kubuntu by October 2007 is now used in all of the 1100 state-run schools in the Canary Islands.[53][54]

http://en.wikipedia.org/wiki/Kubuntu

woodsmoke

[DoYouKubuntu]

That's very impressive.

I'm always a bit perplexed when I see that Linux STILL supposedly only constitutes ~2% of desktop OSes. I just don't believe that--and that's based on my own experience, both in real life and on non-Linux forums, where the majority of people use Linux. Yet we're supposed to believe that, overall, Linux is only on ~2% of desktops?

[woodsmoke]

lol well consider the gazillions of desktops world wide lol But I quite agree! lol

[eggbert]

2006 and 2007, that's a long time ago, before kde 4.0 even. I want to know how it worked out for them and if they are still using Kubuntu.

[GreyGeek]

2006 and 2007, that's a long time ago, before kde 4.0 even. I want to know how it worked out for them and if they are still using Kubuntu.

It worked out very well for them. There are several recent articles about the conversion, which was completed about a year ago.

The "Linux has only 2%" mime is a myth, and always has been, but with the advent of smartphones that metric is meaningless. Including all digital devices Microsoft's market share is now about 20%: http://seattletimes.com/html/microso...more_than.html
The big killer is, of course, Android, which is based on the Linux kernel plus a lot of Google stuff tacked on. I say it is a killer because even Microsoft is selling smartphone with Android on them. http://au.ibtimes.com/articles/55720...smartphone.htm

Microsoft and Apple count retail sales and base their market shares on that dubious figure. While RedHat and SUSE sell a lot of Enterprise stuff most Linux installations are made from ISOs burned from free downloads and installed on one to many computers, some of which came with Windows preinstalled (as my Acer did), so while Window's tally as decreased by one and Linux's has increased by one, the bean counters don't count my swap, which means their tally is erroneous.

The bottom line is that Apple is (IMO) eating Microsoft's market share, as counted by the bean counters, but Linux (Ubuntu, Mint, RedHat, SUSE, Kubuntu, etc...) are gouging chunks out of both of them.

By the way, I am sure enjoying playing Steam/LinuxOS (Castle Story, Portal) on my Kubuntu installation with my grandson. Hires, lightening fast. When I run their server shell he can log in from his house and we can both play at the same time, with no lag or speed slow down. My grandson is running Kubuntu 14.04 too.

[DoYouKubuntu]

The "Linux has only 2%" mime is a myth, and always has been

That's what I've thought, too, but it's annoying to continually see that figure.

but with the advent of smartphones that metric is meaningless. Including all digital devices Microsoft's market share is now about 20%: http://seattletimes.com/html/microso...more_than.html
The big killer is, of course, Android, which is based on the Linux kernel plus a lot of Google stuff tacked on. I say it is a killer because even Microsoft is selling smartphone with Android on them. http://au.ibtimes.com/articles/55720...smartphone.htm

That's pretty funny, micro$oft basically throwing in the towel and jumping on the Android/*LINUX* bandwagon.

Microsoft and Apple count retail sales and base their market shares on that dubious figure. While RedHat and SUSE sell a lot of Enterprise stuff most Linux installations are made from ISOs burned from free downloads and installed on one to many computers, some of which came with Windows preinstalled (as my Acer did), so while Window's tally as decreased by one and Linux's has increased by one, the bean counters don't count my swap, which means their tally is erroneous.

Right. EVERY computer I've bought in the last 25-ish years came preinstalled with m$'s idea of an OS on it--which I promptly and satisfyingly removed so I could replace it with Linux. ALL of those computers got added to the "m$ has X market share!" numbers, despite the fact that they NEVER ran window$. It's all about sales figures, and since m$ has had a stranglehold on PC makers, right down to insisting that they ship with window$ OR ELSE, that's just how it was. I've really resented paying the 'micro$oft tax' on so many computers, but what invariably happens is I'll see a really great deal on a computer and decide to buy it--instead of building one from scratch, since I no longer have that much enthusiasm and/or desire--knowing full well that it comes with m$ preinstalled. I'm glad there are now companies selling Linux computers, but their prices can't match the deals I find on other computers. Besides, I LIKE installing my own Linux because I get to choose how to partition the drive in terms of sizes, how to set up the partitions in terms of its layout (/, /home, /data, etc.), and installing my preferred desktop environment right off the bat. Kubuntu, of course.

By the way, I am sure enjoying playing Steam/LinuxOS (Castle Story, Portal) on my Kubuntu installation with my grandson. Hires, lightening fast. When I run their server shell he can log in from his house and we can both play at the same time, with no lag or speed slow down. My grandson is running Kubuntu 14.04 too.

I LOVE us multi-generational Kubuntu people. I am so proud of my mom for having transitioned to Linux when she was in her 80s. Her ONLY complaint--after we got her "let me screw up my desktop royally!" problem under control--was, "it's too FAST!!!" Those are not words normally heard along with the word "window$" in the same sentence.

[eggbert]

It worked out very well for them. There are several recent articles about the conversion, which was completed about a year ago.

That's good to hear, I am glad it worked out well for them.

The "Linux has only 2%" mime is a myth, and always has been, but with the advent of smartphones that metric is meaningless. Including all digital devices Microsoft's market share is now about 20%: http://seattletimes.com/html/microso...more_than.html
The big killer is, of course, Android, which is based on the Linux kernel plus a lot of Google stuff tacked on. I say it is a killer because even Microsoft is selling smartphone with Android on them. http://au.ibtimes.com/articles/55720...smartphone.htm

I don't think the 2% market share number is a myth. If you count smart phone OS's like android as "Linux", then yes it's much greater than 2%. But the traditional Gnu/Linux desktop doesn't have much real world usage in any of the stats I've ever seen. I think wikipedia is a good indicator of actual real world platform usage, and Desktop Linux is barely hanging on with only Ubuntu and "Linux Other" both having under 1%.

http://stats.wikimedia.org/wikimedia...ingSystems.htm

[GreyGeek]

http://distrowatch.com/awstats/awstats.DistroWatch.com.osdetail.html

Your web metric is interesting. It mixes platforms though.

Windows	102,092 M	43.28%
iPhone	57,784 M	24.50%
Linux	28,865 M	12.24%
iPad	24,121 M	10.23%
Mac	12,796 M	5.43%
BlackBerry	874 M	0.37%
SymbianOS	66.1 M	0.03%
FreeBSD	28.9 M	0.01%
DoCoMo	14.4 M	0.01%
SunOS	5.7 M	0.00%
OpenBSD	685 k	0.00%

The problem is in correctly interpreting the "user agent compatibility mode" setting. Many non-Windows users set theirs to mimic IE. Or, used to. I don't pay any attention to that stuff any more. DistroWatch shows

Operating Systems

Windows 2,302,219 44.2 % Windows XP 284,139 5.4 % Windows Vista 494,836 9.5 % Windows (unknown version) 18,001 0.3 % Windows NT 1,182 0 % Windows ME 27 0 % Windows Vista (LongHorn) 52,326 1 % Windows Mobile 121 0 % Windows 98 875 0 % Windows 95 105 0 % Windows 7 1,438,862 27.6 % Windows 2003 9,082 0.1 % Windows 2000 2,580 0 % Windows 3.xx 83 0 % BSD 10,035 0.1 % OpenBSD 1,131 0 % NetBSD 318 0 % GNU/kFreeBSD 4 0 % FreeBSD 8,582 0.1 % Linux 2,325,426 44.6 % Zenwalk GNU Linux 6 0 % Vine Linux 44 0 % VectorLinux 3 0 % Ubuntu 101,355 1.9 % Suse 1,353 0 % Red Hat 248 0 % PCLinuxOS 552 0 % Mandriva (or Mandrake) 1,184 0 % Gentoo 464 0 % Fedora 774 0 % Debian 2,609 0 % Centos 318 0 % Google Android 124,634 2.3 % GNU Linux (Unknown or unspecified distribution) 2,091,882 40.2 % Macintosh 289,581 5.5 % Mac OS X 288,849 5.5 % Mac OS 732 0 % Others 275,991 5.3 % Unknown 249,856 4.8 % Unknown Unix system 15,517 0.2 % Java Mobile 4,334 0 % BlackBerry 1,760 0 % Sun Solaris 1,260 0 % Symbian OS 873 0 % OS/2 834 0 % Java 688 0 % Nintendo Wii 280 0 % Sony PlayStation 269 0 % Syllable 92 0 % RISC OS 89 0 % BeOS 70 0 % HP UX 44 0 % Irix 16 0 % Commodore 64 4 0 % iOS (iPhone/iPod/iPad/...) 4 0 % AmigaOS 1 0 %

The NetApplications website, which was the orgin of the 2% myth, was (is?) a rebrander of EXEs, meaning that only those interested in the Windows platform would hit that web site.

Wikipedia is a more generic website, but the metric still has problems. Does it count the same IP address more than once? Users of Linux tend to be more technically sophisticated and use fixed IP addresses, while the Joe and Sally Sixpack Windows & Mac users just let the ISP dispense a fresh IP address every time they boot up. What algorithms does it use to parse the user agent? Notice that the site pre-emptively removes certain agents and selectively reports others.

This report ignores all requests where agent information is missing, or contains any of the following: bot, crawl(er) or spider.Wikipedia: NT 5.0: Windows 2000, NT 5.1/5: XP/Srv 2003, NT 6.0: VISTA/Srv 2008, NT 6.1: Windows 7, NT 6.2: Windows 8/Srv 2012, NT6.3: Windows 8.1/Srv 2012 R2.
Wikipedia: OS X 10.4: Tiger, 10.5: Leopard, 10.6: Snow Leopard, 10.7: Lion, 10.8: Mountain Lion, 10.9: Mavericks.
Wikipedia: Linux distributions.

[oshunluvr]

Interesting information, but I will have to disagree about this conclusion:

Users of Linux tend to be more technically sophisticated and use fixed IP addresses, while the Joe and Sally Sixpack Windows & Mac users just let the ISP dispense a fresh IP address every time they boot up.

At least anywhere I know about, the ISP controls this option through costs. If you prefer a fixed IP, they charge you for it. My assumption would be that the answer to the "fixed or floating" question is primarily a non-factor for most people, regardless of OS. Those that need a fixed IP bad enough that they are willing to pay for it will do so, regardless of OS.

The assumption that Linux users in general are more tech-sophisticated is likely highly accurate, but I argue tech-sophistication alone isn't the determining factor regarding one's choice of IP type - need vs. cost is the primary reason. In addition, there are numerous tools (domain services are one example) available that let one overcome the effects of a floating IP. One could argue that a more sophisticated user might be less likely to have a fixed IP in an area where the cost outweighs the ability of the user to circumvent to effects of not having one.

I have a server and over the last couple of years my children have spread out around the country. Once we re-locate to NC, I will be setting up my server for them to access. Unless my ISP offers a free fixed IP, I will be using a different method to allow access to my server. I suspect there are many highly sophisticated users that have no need of a fixed IP and therefore more likely to pay for one or even bother getting one, even if it is free.

[eggbert]

Your web metric is interesting. It mixes platforms though.

Only under order of popularity does it group all variants of Linux. If you look under "Breakdown per platform for Mac and Linux", it even lists individual distros. The 12% you highlighted is Android. It groups Android as a Linux variant. If you look at the individual GNU/Linux distors, they all get less than 1%. Android is -not- GNU/Linux. It's a Linux Kernel running a Java VM.

The problem is in correctly interpreting the "user agent compatibility mode" setting. Many non-Windows users set theirs to mimic IE. Or, used to. I don't pay any attention to that stuff any more.

It would seem unlikely that many non-windows users would change the user agent string of their browser. Some might, but it's hard to imagine that it would be any significant percentage. Especially nowadays when there aren't many IE only sites like there used to be. What would even be the advantage of changing it?

DistroWatch shows

You'd kind of expect a site like distrowatch to have higher linux numbers. It is, after all, geared toward Linux users. It would be a bad metric to gauge overall linux usage.

The NetApplications website, which was the orgin of the 2% myth, was (is?) a rebrander of EXEs, meaning that only those interested in the Windows platform would hit that web site.

Wikipedia is a more generic website, but the metric still has problems. Does it count the same IP address more than once? Users of Linux tend to be more technically sophisticated and use fixed IP addresses, while the Joe and Sally Sixpack Windows & Mac users just let the ISP dispense a fresh IP address every time they boot up. What algorithms does it use to parse the user agent? Notice that the site pre-emptively removes certain agents and selectively reports others.

I still don't think these corner cases are enough to account for the small percentage Linux has. Even on sites which I am responsible for (Some get quite a bit of traffic), Linux usage is always less than 1% according to any web analytic we use.

I don't care if Linux has 1% usage, I will still use it, but I always see these claims of how there are so many Linux users and how the stats are so flawed, but it just does not seem to reflect reality (imho).

[GreyGeek]

... I still don't think these corner cases are enough to account for the small percentage Linux has. Even on sites which I am responsible for (Some get quite a bit of traffic), Linux usage is always less than 1% according to any web analytic we use.

What? Your own link gives Linux a 12.24% market share, which certainly isn't a "corner case" by any stretch of the term. Linux was measured at 4% in 2004 and 8% in 2008. It was predicted to reach 10% in 2010. By extrapolation, at the least, it has reached near 12% by 2014, a decline which the overal PC market is showing because of the adoption of the tablet and smartphone by Joe and Sally Sixpack. I'll settle for 12%. Microsoft overall is at 20% now and most models predict that Linux and Apple while split the lion's share with MS remaining lower than either of them.

[eggbert]

What? Your own link gives Linux a 12.24% market share, which certainly isn't a "corner case" by any stretch of the term. Linux was measured at 4% in 2004 and 8% in 2008. It was predicted to reach 10% in 2010. By extrapolation, at the least, it has reached near 12% by 2014, a decline which the overal PC market is showing because of the adoption of the tablet and smartphone by Joe and Sally Sixpack. I'll settle for 12%. Microsoft overall is at 20% now and most models predict that Linux and Apple while split the lion's share with MS remaining lower than either of them.

Please look at the stats again. The 12% is mostly all Android. If we're talking about traditional desktop GNU/Linux, then the < 1% still stands. If we're talking about "Linux kernel usage", then yes, I would agree, that's used everywhere. But Android, Chrome OS, etc is not GNU/Linux as it has been traditionally known. They use the Linux kernel, but otherwise they have little in common with typical desktop Linux distros like Ubuntu, Fedora, SUSE etc.

[GreyGeek]

Opps! My bad!

[DoYouKubuntu]

Only under order of popularity does it group all variants of Linux. If you look under "Breakdown per platform for Mac and Linux", it even lists individual distros. The 12% you highlighted is Android. It groups Android as a Linux variant. If you look at the individual GNU/Linux distors, they all get less than 1%. Android is -not- GNU/Linux. It's a Linux Kernel running a Java VM.

So it *IS* Linux. It's irrelevant [to me, anyway] what is run ON TOP of it, if it's still Linux at heart.

It would seem unlikely that many non-windows users would change the user agent string of their browser.

I remember when MANY Linux users did just that. And it's possible that some left it that way and/or continued doing it by rote with each new computer/browser. *I* never did. Any site that insisted on Internet Exploiter was NOT a site I'd want to visit and/or do business with. Can you imagine a bank insisting on the least secure browser on earth?! I understand that many used to do just that. Not mine!

Some might, but it's hard to imagine that it would be any significant percentage. Especially nowadays when there aren't many IE only sites like there used to be. What would even be the advantage of changing it?

True, TODAY there's not much sense to it or need for it.

I still don't think these corner cases are enough to account for the small percentage Linux has. Even on sites which I am responsible for (Some get quite a bit of traffic), Linux usage is always less than 1% according to any web analytic we use.

That's interesting, because on sites I'm responsible for Linux makes up a MUCH higher percentage of users, varying from 20-60% [not counting Android]. And these are non-Linux sites, they have nothing to do with Linux, other than running on it.

I don't care if Linux has 1% usage, I will still use it, but I always see these claims of how there are so many Linux users and how the stats are so flawed, but it just does not seem to reflect reality (imho).

See, I'm the exact opposite. It's specifically BECAUSE I know so many people who use Linux, both in real life and online--and I mean on non-Linux places, like TV show forums--that I believe the low percentage is incorrect. It simply doesn't make sense that in my little slice of the world the MAJORITY of people use Linux, yet in the world overall that number is <2%.

[woodsmoke]

well, when I see these discussions I always harken back to the bar POS computer at the Elks and the "Keno" computer. When rebooted, after a lightening strike, power outage, etc...one sees that the OS floats atop Ubuntu server.

The college is running all of the MS stuff, this next month "upgrading" to 8.1 atop Linux servers.

Surprisingly, the physical science department has a couple of Linux computers which run Linux apps, and....with the approval of the IT department.

I have had, in the last ten years, many students who run Linux laptops and also got "two year' type certifications in Linux.

All of the people who got the certs, got jobs locally, or out of the area.

And the last "in class" student with a lappy was spring semester '14.

So....don't know about the statistics, but this is the "heartland", not France....if Linux is penetrating here,there must be some validity to some of the statistics showing an increase.

Again, just anecdotal evidence.

woodsmoke