“from 12 August, public administrations looking for a new software solution must either use an application which they have already developed in-house, develop their own new program, use open source software, or any combination of these.”
I don’t know anything that FLOSS cannot do that really needs to get done so that should pretty well break down M$’s barrier to entry for applications, making M$’s OS an endangered species in Italian government sooner or later. I like it.
see Italy making way for open source – The H Open: News and Features.
17408
12734
206
3
2
23843 
11832 
11684 
4615 
4236 
1641 
198 
14 
2 
0 
0 
0 
Only when a technical and economical analysis demonstrates that it is not possible to obtain at a lower price an open source solution or to reuse a solution developed internally, then it is allowed to purchase a proprietary licence of use. (Source)
There you have it. Nothing will change.
“so that should pretty well break down M$’s barrier to entry for applications”
Spoken like someone with no actual development experience.
Do you even know what a barrier to entry is? It’s something that prevents you from entering a market.
Windows has the biggest marketshare, the simplest and most well integrated development tools, and an API and ABI that are stable enough to go back 15 years.
Those aren’t barriers!
Meanwhile, on Linux, I need to account for hundreds of distro inconsistencies (or just pick a popular distro and alienate the others). I need to account for an ever changing API/ABI meaning I need to make sure I recompile my app to make sure it didn’t break every time a minor kernel point release is issued. I need to account for multiple windowing systems (or just pick one if I don’t have infinite resources). And I need to account for the fact that the market is so small that only a fraction of the 1% of the market linux occupies may ever use my app.
Those are barriers!
But why listen to what an actual developer is saying when you could take the word of a retired high school teacher who taught kids how to crimp network cables.
TM lied three times when he wrote,
No, you don’t because a proper application runs at the user-level and rarely communicates directly with the OS. You are using libraries providing such abstraction are you not? A distro will have libraries that work with their OS and you don’t have to worry about it. That’s why many applications distribute the source code and the distros take care of all those details so you don’t have to and the users don’t have to. If you cannot distribute to distros in source code, that’s your problem not ours. Closed source software is its own problem, like square wheels.
No, you don’t. Just give it a web interface and let open standards take care of all that. More than 50% of new applications are now OS-independent. Perhaps you should modernize.
No, you don’t because a proper application runs at the user-level and rarely communicates directly with the OS.
Huh? The whole idea of an OS API is to eliminate a lot of code being carried in an application and to share it with other applications via the OS! Linux is world famous for busted APIs due to many folk tampering with everything and the difficulty of discipline in an environment where everyone is a volunteer.
If GNU/Linux were only ~1% why are the trolls/Astroturfers of M$ swarming around this blog?
Is that your proof?
Clarence Moon when you do take a close look at windows the same libraries that change ABI under Linux are shipped with Windows applications.
The Linux world famous for busted ABI’s is not in fact backed by data. There are libraries in Linux that are highly stable.
http://upstream-tracker.org/
The hard reality here is the volunteer arguement is bogus. ABI checking can be embedded into the libraries test-suite.
Clarence Moon what we see in wine is that you are bogus because even with Microsoft discipline fault detectable by ABI checking tools get inserted into Windows. So Microsoft is depend on their staff doing the right thing instead of science based testing.
Quality ABI does not depend on Humans todo the right thing. Instead has tests that detect when someone busts it. Since this is the case there is no major difference between volunteer and discipline environment. Only major difference is if the tests were performed or not.
So yes the data that discipline environment does not work to produce a Quality Stable ABI exists just happens to be Microsoft Windows.
Lot of direct x issues would be detected if Microsoft did have proper quality controls in the Microsoft ABI’s.
Clarence Moon the Linux kernel to user-space ABI is the most stable ABI of any in existence. Volunteer and produce quality in lots of projects. But there is a catch it also has stacks of test suites. These are the magic to ABI stability no good test suite no stablity. Next of all things is glibc. There has only be 2 breakage in all the time of Linux. Compare to Microsoft 7 and counting breakages of there libc equal libraries.
1 breakage was a complier change and one breakage was a feature removed on security grounds with glibc. So to use glibc provided from start to now funny is possible if your program is written in C. Some of the special handing for C++ was broken by complier change. Same has happened on Windows far more often.
If you want to be true its not because Linux projects are run by volunteer. Its the fact that libraries people choose to use don’t have ABI testing frameworks.
TM Repository a Linux standard base application release in 2000 will run on Linux 1995 and forwards. Stable ABI exists just not used.
The glibc was one of the first project outside the kernel to start running ABI testing frameworks.
Interesting enough ABI testing frameworks are mostly automatic.
There is a lot of research around Linux. http://oswatershed.org/
Exact tracking how old each distribution is retaliative to new releases.
Clarence Moon as far as I know Microsoft never released a single document how to make a ABI compatible library or check if a library is true ABI compatible. If it exists it would be interesting to add to my bookmarks.
http://ispras.linuxbase.org/index.php/ABI_compliance_checker There is a lot of Linux made documentation for it that gets used.
Clarence Moon when you look at the core c, c++ and .net libraries on windows I wish Microsoft Developers would stop tampering with everything.
The excuse that volunteer cannot make ABI stability has to die. The reality they have proven they can. The issue is getting projects to include ABI compliance checker in their test-suites. More are including it.
TM Repository as normal you are still stating something with no base in reality. I don’t believe for one min you have ever built a program for Linux. Truly targeting Linux. Even applications talking straight to kernel the userspace ABI basically does not change.
If it did every time you update your kernel you would have to update everything in the distribution above it. Reality this don’t happen. Linux distributions attempt skin kernel.org developers alive when they break the user-space ABI from kernel.
So the minor kernel point release issue is basically impossible for user space programs to be an issue since distributions will not let it happen.
The only reason minor kernel point releases are a issue is if someone is developing of all things a kernel driver.
Ubuntu has a habit of running experimental security frameworks. These result in valid ABI calls to kernel being rejected. Not ABI breakage. Security rejection. Same reason why particular programs when install on windows tell you to disable anti-virus.
Funny enough is lot of the Ubuntu security add-ons mostly only trip up over security flawed code in the first place. So your code was not quality workmanship??
Reality distribution don’t rebuild every program every point release. They don’t have to. Dependency mapping exists for a reason. This can produce a tool that will tell you if anything you application depends on being windows, Linux or OS X has changed that could possibly break it. Makes for far better installers.
This reduced quality assurance on all project. Yes the Linux Standard Base ABI checker once you know it is your best friend for developing on all major platforms with native code.
TM Repository and Clarence Moon. Get this threw your thick heads.
Evidence that binary compatibility exists on Linux is the existence of binary Linux Distributions.
If what TM Repository was claim was true every time the Linux kernel updated a point debian with its 30 000 packages would have to rebuild quite a few so they stay working.
This does not happen. You can chroot older versions of debian on current kernel because the kernel ABI to user space in ABI stable.
The reality Binary Distributions hate ABI change as much as the next person in fact worse. Problem is most Distributions hate having multi versions of libraries. So the ABI is stable just distribution don’t provide the framework for using older libraries.
There are a few libraries like boost that are design as cross platform libraries that are not ABI compatible because they were designed to be shipped with Applications.
Clarence Moon really do you think volunteer running distributions will like the idea of ABI changes. No way in hell will they like.
The ABI stablity thing is most bogus.
Dependency hell yes. Where the distribution removes a library you depend on. Not that the ABI of that library changes they simple go rid of that library.
Basically if the library you need is there you application works this is Linux. This is the way it has to be or binary distribution cannot work. So redhat, debian, suse, ubuntu would not exist if binary compatibility did not exist between kernel versions and even core libraries due to the fact they would be killed by work load.
“TM lied three times when he wrote,”
Keeping it childish with your editorial, I see.
“No, you don’t because a proper application runs at the user-level and rarely communicates directly with the OS.”
I see, how many applications have you written? Because needing to know whether or not there’s a network connection tends to be fairly important for most applications these days.
“No, you don’t. Just give it a web interface and let open standards take care of all that.”
So supply a custom bandaid solution rather than just use native controls? You do realize the astronomical amount of work that requires, reinventing the wheel, especially when accessibility and internationalization are taken into account. I’ve used plenty of frameworks which attempt to do this (PhoneGap, Titanium, Sencha, AIR, etc.) and they all have severe shortcomings, especially when you need native access. It ultimately wound up costing me less to write a separate iOS and Android app than use an html5 bridge.
“More than 50% of new applications are now OS-independent.”
Where are these statistics coming from? Please link me to the source.
“Perhaps you should modernize.”
Says the guy who still writes Pascal scripts and who’s only knowledge of Windows and OSX comes 2nd hand from Linux forums. To you, HTML5 isn’t useful technology, it’s ammunition in your holy war.
“This does not happen. You can chroot older versions of debian on current kernel because the kernel ABI to user space in ABI stable.”
I don’t have to trick Flash MX, a very old version, into running alongside Flash CS6. Same with Photoshop, Reason, etc. Hell, I can run IE5 on my Windows 7 machine if I wanted to without needing to do any chroot-fu.
Get it? On Windows, my users don’t have to think about what version of the kernel they’re running. Instead, they just install my apps and run them.
“And the ~1% thing again? Give it a break.”
I’m not the one saying 1%, Wikipedia is.
Those are their LIVE USAGE STATS. Are you telling me that people who use Linux on the desktop are not visiting wikipedia?
“If GNU/Linux were only ~1% why are the trolls/Astroturfers of M$ swarming around this blog?”
And I’m not trolling. I’m leveling actual complaints and issues I’ve had and asking you to accept that they exist. So far though, you’ve still got your hands firmly over your ears in the typical self-censoring of a fanatic. Ironic, given you’re a fanatic espousing freedom.
If you really loved Linux, you’d be very critical of it. You’re like a bad parent who refuses to believe their child is as horrible as all the other parents say he is. Rather than issuing some tough love, you coddle the kid and only encourage their bad behaviour.
(Plus, it’s bloody embarrassing to think you come from the same country as me. So much for Canadian tolerance, you’re a fascist as they come.)
This does not happen…
It happens all the time according to FLOSS developers. You are just a chipper, Mr. O, and would not recognize such real world things, apparently, but even the in-crowd of FLOSS sees the problem as a show-stopper. Check on some of the links, too.
TM Repository the answer comes back to the same thing.
Under Linux you don’t need to think about the kernel. This is you big mistake.
Its the libraries above it that break applications. TM Repository. chroot-fu does not change the kernel user-space exposed ABI. The kernel arguement is mostly bogus. The things from the kernel are the same reason why you would ship with a notice to disable anti-virus.
Using Linux Standard Base loader instead of default distribution means you now can ship binaries the same way as windows with there runtime parts.
IE5 will not install on Windows 7 without doing something like chroot-fu. Particularly this http://www.intechgrity.com/ie55-ie6-to-ie8-emulator-for-windows/ Otherwise you break windows 7 install if somehow you got IE 5 to install. For some reason it does not like its version of IE replaced by an older version.
Chrooting like stunt to run old applications is done under windows. Same with loader bending.
I have 4 copies if OpenOffice and LibreOffice installed at the same time different versions. All I did was tell the LSB packages from both projects to install in independent directories. So no chroot-fu required because they were packaged correctly.
Could graphical tools be made to make the chroot, loader bending stuff more transparent mostly likely. For the number of applications that it is minor.
TM Repository get this point. With ABI with Linux if the issue does not link to drivers there is no a issue. So userspace program failing due to kernel ABI basically does not happen. User-space program failing because security features are enabled in kernel and the program is doing something suspect does happen. Anti-viruses under windows cause particular program not to run as well. This is the normal migraine of security. The Linux kernel does a lot an anti-virus would normally do under Windows.
TM Repository you arguement has more holes than swiss cheese. I can hit from about 30 different directions. I am choosing different directions so I am not repeating myself as much.
The ABI arguments are that bogus they are simple to disprove.
There is only 2 major arguments that are somewhere near valid.
1) Dependency hell
2) Lack of kernel driver ABI
Everything else around ABI that trolls like you argue TM Repository are quickly proven false.
Dependency hell is what force use of chroot for using old binaries from a distribution.
Its what should force closed source software makers to use loader bending or Linux Standard Base loader.
Since a shim like system does existing in Linux and its lacking the front end framework to exploit it. This is how come you can change the loader and set a rpath on a elf binary. This might be case of go to distributions and say please make us a tool exploiting this. Doing this would overnight magically provide backwards compatibility to Linux to 1995.
Windows has a shim engine that detects particular programs and automatically applies changes. Remove the shims and lots and lots of programs instantly stop working.
One interface trigger on windows is you know compatibility mode settings. Same thing could in theory be done on Linux.
Shim engine is another way around binary incompatibility.
Kernel driver ABI one is a huge arguement. We need to see what happens after 3.7 Linux to see if there is any grounds to this arguement. Of course people miss that a lot of embedded devices use Linux userspace driver framework until they can get mainline. So there drivers are independent to kernel change completely.
TM Repository go back and read when you raised Loki games. The simple reality I disproved all this arguement back then. I really would say Robert Pogson would be in his right to delete you post if you raise it again.
If you want to talk about the real problem I have no problem talking about it. I will not lie or cover it up. We don’t need the lies.
Clarence Moon learn to read.
–http://tirania.org/blog/archive/2012/Aug-29.html–
From your link.
Miguel de Icaza
–We deprecated APIs, because there was a better way. We removed functionality because “that approach is broken”, for degrees of broken from “it is a security hole” all the way to “it does not conform to the new style we are using”.
We replaced core subsystems in the operating system, with poor transitions paths. We introduced compatibility layers that were not really compatible, nor were they maintained. When faced with “this does not work”, the community response was usually “you are doing it wrong”.–
This is what he did when he was running gnome.
Quote him does not help your arguement at all. He is the cause.
Linux Standard Base was designed to address this. Loader bending was designed to address this. Providing migration path no matter what people like Miguel de Icaza did.
The kernel never ever did what Miguel de Icaza did. Gnu libc never did what Miguel de Icaza did. KDE has never done what Miguel de Icaza did.
Reality he is talking about a minority.
Large percentage has not done what a few rogues have done.
He repeat made excuses why he did not have to provide a stable ABI.
–The efforts to standardize on a kernel and a set of core libraries were undermined by the Distro of the Day that held the position of power. If you are the top dog, you did not want to make any concessions that would help other distributions catch up with you. Being incompatible became a way of gaining market share.–
This has been his repeating excuse why he did not have to standardise.
Miguel de Icaza is one of the source problems.
Others developed solutions. What you are doing here Clarence Moon is talking to a murder and believing his excuses. So he gets to sleep in bed happy believing he was fine.
Clarence Moon quoting one rogue developer means nothing. There are quote from other developers over that post mostly disproving it. They did not want to be tarred with that brush of incompetence.
Clarence Moon if you want more proof that gnome has poor management go read the freedesktop.org mailing lists. Particularly his arguments why standardisation did not have to happen.
There is a far better developer to quote over this issue Clarence Moon.
http://people.gnome.org/~michael/blog/2012-09-10-desktop-linux.html
Michael Meeks he does not rose cover it and he does not say everything is bad. He also does serous-ally look at how windows does it and how linux and state exactly what he thinks is wrong. Even providing guidance how to go forwards.
Miguel de Icaza is completely the wrong person he has never really attempted to address it.
Gnome has had better backwards compatibility after Miguel de Icaza left. For some reason Microsoft gave Miguel de Icaza a MVP. Rewarding a person who does not believing in stable ABI’s is not a sane thing to be doing.
“Miguel de Icaza
–We deprecated APIs, because there was a better way. We removed functionality because “that approach is broken”, for degrees of broken from “it is a security hole” all the way to “it does not conform to the new style we are using”.”
Clearly, you don’t understand the man is being facetious when he says this.
He’s saying that this approach was bad! He’s saying there was a lot of activity, but nobody was directing it. He’s saying that everything became disorganized and monkey-patched. He’s saying “the road to hell is paved with good intentions”.
TM Repository he was the one doing that method.
–We introduced compatibility layers that were not really compatible, nor were they maintained.–
Here is a good question who was marked as maintainer of lots and lots of the compatibility layers in Gnome that were not maintained. None other than Miguel de Icaza himself. This is a case of a murder trying to blame others.
KDE did a lot of compatibility layers all have really been compatible and maintained. Same with a lot of other projects. Linux kernel for video for linux 1.0 to 2.0 there is a compatibility layer that works perfectly. Then the multi version handling in glibc what is another stack of perfectly maintained compatibility layer.
The reality with Miguel de Icaza he marked himself as maintainer on too much stuff. Result he did not have enough hours in the day to properly maintain it. He is a cause of the mess.
TM Repository
–He’s saying there was a lot of activity, but nobody was directing it.–
This is not correct the gnome disaster of compatibility is because 1 person tried to direct it all and the problem is too large for 1 human. So patched to the compatibility layers were not getting processed by him because he did not have the time.
He should have done more delegation. Like this person will be responsible to make sure backward compatibility here works and that is there only assigned role this is how most project that are not gnome work and how gnome started working after he left.
KDE, gnu libc, Linux kernel they all use delegation effectively.
Some of the problem with mono today is Miguel de Icaza failure to perform effective delegation.
If you have committed the crime you should accept the fallout. Not attempt to spread the blame on to others who did not do it.
Something marked as not having maintainer tells others who are interesting they can step up and be the maintainer. When something is marked as having a maintainer and it really does not this leaded to big problems.
Due to what Miguel de Icaza did today debian runs tracking to see if faults are being processed by upstream maintainers then if they are not go to check out if they are really there. Because actions like this cannot be tolerated.
This is why a lot of FOSS people when you quote Miguel de Icaza straight up get a disgusted response.
I am serous over this topic don’t quote Miguel de Icaza. A lot of what he is saying is more an attempt to hide he is a cause. He is say we instead I. Yes he really should be saying I a lot in that post.
TM Repository if you want a person to blame for some of your programming head aches on Linux it is Miguel de Icaza.
http://stats.wikimedia.org/wikimedia/squids/SquidReportOperatingSystems.htm
Wikipedia is run by tyrants who claim the authour of a book is not an authourity on the book…
Their article is inconsistent with their own web stats which show Linux at 6.81% of which 5.26% is Android/Linux and 1.55% is GNU/Linux. Their stats also show they oversample English traffic seriously. That would be USA, Canada, UK, Australia and some others. so ~1% is not even close.
NetApplications the source of much of these silly stats shows Google, alone, pulling all of USA by ~0.5% to GNU/Linux even though they only have ~10K PCs. Their sampling is clearly unrepresentative of global usage. The largest sale of PCs on Earth was GNU/Linux in India and NetApplications never showed a blip. NetApplications also shows Apple has more clients than it has sold.
TM wrote, ““More than 50% of new applications are now OS-independent.”
Where are these statistics coming from? Please link me to the source.”
Fig 1, page 10 of The Future of Computing: Indispensible or Unsustainable?
“Wikipedia is run by tyrants who claim the authour of a book is not an authourity on the book…”
LOL! So because Wikipedia supports a claim I make, and because you somehow think I’m a “Microsoft fan”, you’re going to turn around and say that Wikipedia, the most open, well known encyclopedia on the internet is “run by tyrants”? Are you going to claim that they are somehow “Microsoft fans” as well and are skewing these numbers?
“Fig 1, page 10 of The Future of Computing: Indispensible or Unsustainable?”
Great, but this is a brochure for their cloud computing initiative. Where are these numbers actually coming from? Also, this is talking about apps that aren’t “windows-only”, not “OS agnostic”. Meaning any app that has an OSX port is no longer a Windows-only application. I asked you to disambiguate and all you did was quote an equally biased source!
Tell me, where is the cloud version of Reason (you don’t even know what that is, I realize, but it’s music software)? Where’s the cloud version of Flash (most 2D animation studios in Canada rely on it to make cartoons, not ActionScript apps)? Where’s the Linux version of iTunes so I can buy content on the app store and sync my iphone?
I realize you don’t use any of these things, but it doesn’t mean the people around you don’t, you selfish goon! Your dubious claims are backed by even more dubious figures. I provide you with actual stats, you provide me with marketing material. You’re no better than Microsoft in those regards!
“Their article is inconsistent with their own web stats ”
It isn’t an article, it’s monthly stats.
“which show Linux at 6.81% of which 5.26% is Android/Linux and 1.55% is GNU/Linux.”
Android isn’t Linux, Dougman said so. It is especially not GNU (remember, it got hit by GPL violations)! It can no longer be built with pure GNU toolchains. You can’t get the source to every app on it; Not even the Google apps like Gmail, Google Maps or Google Earth. You can’t even get the source to the most recent versions. And let’s not forget that Android is wholly incompatible with upstream now because Google rewrote the archaic messaging system in the mainline kernel.
You don’t see anyone claiming that iOS is OSX even though much of iOS has OSX underpinnings. You don’t see anyone claiming mobile Safari and desktop Safari are the same thing. So don’t kid yourself. Android stats belong to Android, not Linux, and certainly not GNU/Linux.
TM Repository you cannot get the source code to every application that runs on Desktop Linux either.
–And let’s not forget that Android is wholly incompatible with upstream now because Google rewrote the archaic messaging system in the mainline kernel.–
Other than the fact that the binder IPC is now in staging so part of the default kernel now and has been for almost 6 months.
Android was incompatible with upstream. There is enough in the mainline kernel now to run Android. That statement is now false. Linux 3.3+ can run Android without applying the extra patches. Not perfectly but it runs.
TM Repository did you not read this link the last time.
http://elinux.org/Android_Mainlining_Project
It tells you clearly that binder is mainline.
Are you getting it yet you are using out of date information please stop.
TM Repository something to remember counterfeit nokia and iphones do happen. Counterfeit product being picked up does not stop those playing by rules from keeping on shipping product.
–Will Android Be Crushed for GPL Violations?–
Reality no. Because Nexus and many other phone and devices play perfectly by the GPL rules.
–Where’s the Linux version of iTunes so I can buy content on the app store and sync my iphone?–
http://www.gtkpod.org/wiki/Libgpod
To be correct you are refering to Amarok and a few other that have integrated music stores.
The projects are funded of all thing by the music they sell. This is where the idea that Linux people would truly pay for stuff if given the chance. Amarok only recently got a Windows for years they were getting kick backs from on-line music stores from what Linux users were buying.
The cloud based Audio editor serous-ally you need to learn a few things.
http://plucked.de/
TM Repository yes there is such a thing as a html 5 audio editor. Its not reason yet but its enough to prove a html based audio editor can be made.
Yes it not exactly cloud because the processing is done locally in the poor web browser with plucked. Cloud provided not cloud run is possible with audio editors. So reason software could be ported to html5.
I do expect html 5 online animation software to appear at some point rendering most of that flash stuff to the historic past.
TM Repository really if you had investigated the closest match you would have found some interesting things. Cloud providing of software is still fairly new. We are still learning what the limits will be.
TM wrote, “Android isn’t Linux”.
One picture being worth 1K words:
Android Architecture
As everyone can see, Android uses the Linux kernel as the basis.
As everyone can see, Android uses the Linux kernel as the basis.
So what? So does Amazon Kindle Fire. Both neither Fire, not Android are GNU anymore. And Amazon is never ever hints that Fire has any relation with Android.
Phenom wrote:
“And Amazon is never ever hints that Fire has any relation with Android.”
And PCLinuxOS never ever hints that it has any relation with Ubuntu… It’s just a given. All GNU/Linux distributions are related in some way. Some more than others. PCLinuxOS is actually a derivative of Mandriva, or whatever it’s called now. And Mandriva is a derivative of Fedora.
It’s purely a technical matter. What you call it, Phenom, isn’t going to change it. Only in your mind. There are special hospitals for people who have separate realities in their mind. Maybe you ought to check into one.
Pogson, breaking his own website, (so much for the advantages of open source) posted;
“And the ~1% thing again? Give it a break. If GNU/Linux were only ~1% why are the trolls/Astroturfers of M$ swarming around this blog?”
It is between 1% and 2% worldwide, according to most web counters. We’ve been over this before – they can’t all be wrong. You can drag out Sunnyvale, Brazil and India all you want, but;
1% to 2% worldwide.
Trolls? Astroturfers? Give me a break. Speaking for myself, I’m just sick of the lies and hypocrisy of the Linux fan-club, so I provide the facts or point out the dishonesty every now and again.
“Wikipedia is run by tyrants who claim the authour of a book is not an authourity on the book…”
And exactly how does this make their webstats incorrect?
The webstats they reproduce from “secondary sources” are just from the same old sources all the time. Their own webstats diverge greatly with the sources they use in the article. None of the stats are properly sampled. Even their own site is heavily English but they don’t break down the stats by region so we don’t know what it means in terms of share of OS usage. We do have good numbers from Apple because they file PCs shipped with MacOS with the SEC. They are a little above 5%. I have no doubt that GNU/Linux is higher than that because it was around 2003 and GNU/Linux has not slowed at all and there have been huge roll-outs of GNU/Linux desktops on every continent since then. Emerging markets mostly cannot afford Macs so that leaves GNU/Linux in the lead except the stats NetApplications reports report ~1% even in the poorest regions of the world. Cuba and Kenya’s stats are above 3% even though national policies promote FLOSS. The picture of ~1% is seriously warped and I know who’s interested in a picture warped that way.
Ted the 1-2 percent is coming from a non dependable source. We need better answers. Like we know more android devices are selling than iphone and ipad yet this is not reflected in the web numbers.
Now either android users are upgrading more often. Or maybe each goes after a different class of user who has different web habits. So iphone and ipad users are more web using. This would kind make sense thinking how expensive wireless broadband can be. The more expensive item turn up on-line more often might be a reality. In fact it could be both that Android users upgrade more often and have different web habits.
Without ground studies we don’t have a clue what is going on android and we have true sales numbers of android that are connected to phone networks. Linux desktop we don’t have even sales numbers to work from. So you have nothing to check the web numbers against to know if they are correct or even in ball park.
You would suspect Linux Desktop users and Android users to have stuff in common.
Each user group has different habits. Visit lots of FOSS software sites. Then take note how many have advertising. Interesting trend is most don’t. Then visit a lot of free closed source for windows and notice most do have advertising.
Most webstats come from monitoring advertising displays. FOSS users are going to see less advertising. This is why we need real numbers of how many are out there. If FOSS happens to be like 10 percent and web stats are only seeing 1-2. Advertising to FOSS users is not happening.
The government report points out countries that are poorer have greater interesting in FOSS they also have poor Internet connections so less likey to be on the Internet even if Linux is there.
Webstats are basically suspect only used because we don’t have better.
The picture of ~1% is seriously warped and I know who’s interested in a picture warped that way.
Rob ushers in the conspiracy theories. Yes, Ballmer has bought them all. Every! Last! One!
Only Rob P. knows the horrible truth.
I guess Microsoft has also successfully prevented FTL travel. They’re bad. Bad, bad, bad!
OK, so you’ve provided some reasons why Wikimedia’s webstats may be skewed away from Linux (but without any actual evidence). Now explain away all the others.
ALL the web-counters out there STILL show Linux as between 1-2% worldwide.
THEY STILL CAN’T ALL BE WRONG.
“We do have good numbers from Apple because they file PCs shipped with MacOS with the SEC. They are a little above 5%. I have no doubt that GNU/Linux is higher than that because it was around 2003 and GNU/Linux has not slowed at all and there have been huge roll-outs of GNU/Linux desktops on every continent since then.”
You have no proof whatsoever of the original premise – that Linux installs outnumbered Mac OS in 2003. So it follows that your conclusion is flawed. Please do NOT mention that slide from Ballmer; everything else he says/does is treated as a lie by you and your ilk, why should that slide be any different?
This is wishful thinking at best.
“The picture of ~1% is seriously warped and I know who’s interested in a picture warped that way.”
Spare me the pathetic conspiracy theories.
@oiaohm
“Ted the 1-2 percent is coming from a non dependable source.”
Of course it’s “non-dependable” – you disagree with it! And of course any web counter that shows Linux in a good position is gospel truth.
Ted demand evidence, “ALL the web-counters out there STILL show Linux as between 1-2% worldwide.
THEY STILL CAN’T ALL BE WRONG.”
without providing any, “You have no proof whatsoever of the original premise – that Linux installs outnumbered Mac OS in 2003.”
Shit. Read the web, twit.
Computerworld: “Linux captured the No. 2 spot as desktop operating system in 2003,” said IDC analyst Dan Kusnetzky in a recent interview.
Not all web stats report GNU/Linux as 1-2% as you can easily see by finding some:
A Brazilian site, http://www.areaseg.com.br is health-related subject matter not related to IT at all and the stats show 8.95% GNU/Linux. That’s cumulative since 2006 so it does not show current rates, just the share in the last six years. Since that year, the government has done much to promote GNU/Linux in the schools, government offices and by tarrifs.
Here’s another, a business that sells juice. It’s not much of a page, just an ad and a portal. That probably shows business usage at 9.5% GNU/Linux. So, where it the ~1% number for Brazil coming from? Don’t say NetApplications, because they don’t publish proper stats just meaningless numbers. Who are the clients visiting the sites and what are the sites?
Foul language and name-calling??
The incontrovertible signs you’ve lost the argument.
I take the 1-2% thing back – desktop Linux has dropped below “OTHER” worldwide. It’s no longer even worth its own line on the graph. Lumped in with Windows 9x, Windows 2000 and lowly Amiga. How the not-so-mighty have fallen.
Brazil again? Why not mention some larger markets, such as The United States? Is it because desktop Linux is below “Other” there too, even with Sunnyvale propping it up?? Or Europe? “Other” again. Gains in small markets will not affect the worldwide figures. Why not mention that the whole of South America is still only 1.1%? Gains in Brazil don’t even affect those of South America.
“Don’t say NetApplications, because they don’t publish proper stats just meaningless numbers. Who are the clients visiting the sites and what are the sites?”
I am reminded of what you so politely requested of me earlier. I will not, however, descend to your potty-mouthed and insulting level.
http://gs.statcounter.com/factsheet
“our stats are based on a geographically diverse sample of over 15 billion page views per month across more than 3 million global websites. This means we have large enough sample sizes in the majority of countries to be confident that our stats are broadly representative of real world figures” [emphasis mine]
3,000,000 websites.
Three and six zeroes.
THREE MILLION
For the hard of understanding, that’s a lot of websites. Rather too many to list, I’d say.
Ted wrote, “3,000,000 websites.
Three and six zeroes.
THREE MILLION”
Impressed by large numbers? I am not, particularly if they are meaningless. The first step in measuring anything is to know what you are measuring. NetApplications and others don’t tell us anything about what kind of sites those are. They could be M$-specific in large part, you know, “best seen at 800×600″ or such. Heaven knows M$ has tried to lock the world in in many ways.
Consider a university’s computer science department in Singapore. We know who will visit there, likely students, potential students and researchers. So, their stats will mean what those folks use.
Lo! and Behold! 15.7% GNU/Linux. How is that possible in Ted’s world where GNU/Linux is going nowhere? What are those students thinking? MacOS 8%. How can that be?
3,000,000 websites with advertising in most cases.
3 million sites does not mean you will get correct numbers. How many websites are there.
620,132,319 by last netcraft.
Hmm 3 million sites is a sampling from less than half a percent of all the web sites in existence.
193,533,160 are recorded active by netcraft. So at best 1.5 percent sampling of active sites. With such a small sample of sites the margin for error is huge Ted.
Like you would not call a election won when only 1.5 percent of the vote was counted.
So from so possible statistical error is huge. So the allow for possible error. 1-2 percent to 90 percent. Linux is somewhere in their. That is how big the error could be due to the sample size being way too small on web numbers.
The reality is web stats are worthless because they are errored in collection. They don’t collect enough data from enough different sources.
So Linux could have a 20 percent market share and since you sampled 3,000,000 million non FOSS sites you see 1 percent. Yes this is more than possible with how small the sample sizes are.
Robert Pogson his numbers are small there is the problem.
@Oiaohm I do take your point about sampling, it is likely to be culturally distorted as well. However 1.5% of the active sites is a large proportion of traffic if those sites are the most popular ones. I assume this is not the case as they will not tell us their methodology. As in the case of companies like Moody’s I suspect there is an element of telling the customer (MS and Apple) what they want to hear though it is obviously a different field.
Satipera Culturally distorted is particularly important when you remember FOSS users do have their own culture.
Large proportion of traffic. Lets take 1 percent of the USA population could be the upper-class.
Statcounter is one of the bigger ones doing this. Even they are not large enough that there numbers can be trusted.
It could be like saying almost no one speaks mandarin because you don’t have mandarin sites in your counters. We would know this was false straight way because we have ground surveys that cover complete populations over what language they use.
No country wide census has every asked the question what Computer OS do you use. Some have asked if you have a computer. Maybe at some point in the future we might get some areas of usable numbers if the question gets on a few country wide census forms.
In fact I don’t trust any web counter. Web counter only tells you what you customers are using that is about it. Statcounter and others are doing a lot of reaching claiming their numbers tell something about the market. The base facts are not there to support that claim. Statcounter can track a change in the population of users they monitor. That does not mean that population is a representative. Detect a change yes. Know what is truly going on not possible.
oiaohm wrote, “So from so possible statistical error is huge.”
Statistical error is not the problem. They are consistently biased towards business use. That’s why Google shows up with ~10K PCs to move California when school divisions with 100K GNU/Linux PCs count as nothing.
Robert Pogson its still statistical error.
If they are not monitoring the sites students will be visiting. So 100K machines in a school don’t exist along with many other groups.
Since there is not a enough coverage web numbers when you allow for the possible statistical error due to lack of data the number could be almost anything. In fact Microsoft could be losing and get the numbers the way they appear on web numbers. Basically worthless.
For those who quote Miguel de Icaza about the desktop.
http://www.zdnet.com/linus-torvalds-on-the-linux-desktops-popularity-problems-7000003641/
They nicely collated the quotes proving Miguel de Icaza was one of the problems and is attempting to blame others for his failure to manage.
Gnome illness that started under Miguel de Icaza rain. Its meeks who is tasked with sorting it out.
Really is sad the people Microsoft will issue a MVP to.
oiaohm wrote, “Really is sad the people Microsoft will issue a MVP to.”
It’s all part of M$’s technological evangelization. They will do whatever it takes to promote their stuff and denigrate others. Money can buy a lot of “X recommends That Ohter OS”.