I was going over an old Groklaw posting that got me thinking about Microsoft and it’s .NET Agenda. I decided to do more research and took a look at the Comes vs Microsoft documents again but from the perspective of the current Mono issues. After reading the information contained in the documents I am even more convinced that Mono and Moonlight do nothing but help Microsoft win against competing platforms. I will quote and comment on a few relevant sections I found useful to prove my point. These are old documents however I am looking at them with a different angle.
Evangelism Is War
Our mission is to establish Microsoft’s platforms as the de facto standards throughout the computer industry. Our enemies are the vendors of platforms that compete with ours: Netscape, Sun, IBM, Oracle, Lotus, etc. The field of battle is the software industry. Success is measured in shipping applications. Every line of code that is written to our standards is a small victory; every line of code that is written to any other standard, is a small defeat. Total victory, for DRG, is the universal adoption of our standards by developers, as this is an important step towards total victory for Microsoft itself: “A computer on every desk and in every home, running Microsoft software.”
Our weapons are psychological, economic, and political–not military. No one is forced to adopt our standards at the barrel of a gun. We can only convince, not compel. Those who adopt our standards do so as a rational decision to serve their own ends, whatever those may be. It is our job to ensure that those choosing an operating system are presented with an overwhelming abundance of evidence and reasoned argument in favor of our standards–so overwhelming that the choice of our standards seems obvious, or (ideally) that the developer is not even aware that a decision was faced, and a choice made.
We do this by understanding the barriers that might otherwise prevent the developer from adopting our standards, and removing them; by understanding the inducements that might facilitate the developer’s adoption of our standards, and providing them; by understanding the arguments of our competition, and countering them.
Many relevant points are brought up in this section. James Plamondon states (infamously) that “Every line of code that is written to our standards is a small victory; every line of code that is written to any other standard, is a small defeat.”
This would imply that by using the C# standard (ECMA 334 & 335), Microsoft receives a small victory. An example of defeat would be lines of code written in Java, C++ (Standardized, C++98 or C++03 for example), Python, etc.
The second paragraph explains that they want to use psychological, economic, and political weapons to convince people to adopt their standards. Often I hear a lot of business talk about wanting to use Microsoft because most of their customer base is Windows machines. Other excuses for sticking with Microsoft has been the FUD “Total Cost of Ownership” studies that falsely prove Microsoft is cheaper or that GNU/Linux is more expensive.
After that they give a plausible reason for us to assume Mono and Moonlight are approved to help Microsoft. It’s there in their own wording. People did not trust Microsoft not to sue over Mono so they had to issue the Community Promise.
“We do this by understanding the barriers that might otherwise prevent the developer from adopting our standards, and removing them;”
This sums it up quite nicely.
Platforms
What is a platform? A platform is any software service which an ISV’s code assumes is present in order for it to function properly. For example, Windows is a platform, because ISVs can write code that assumes the presence ofthe Windows DLLs, and which will not function properly in their absence.
What are Microsoft’s platforms? Once, the answer was simple: Windows (and before that, DOS). Now, Microsoft has many platforms, and the answer is more complicated. Windows is still one of Microsoft’s platforms. New platforms include Microsoft Office, Microsoft Back Office, Microsoft Exchange, The Microsoft Network, Microsoft Internet Explorer, DevOffice, and ActiveX. For each of these, ISVs can (or will be able to) write code that assumes the platform’s presence.
Turning a product into a platform perpetuates and broadens the success of the product. This is because writing code to support even one platform is hard; writing code to support multiple, alternative platforms is exponentially more difficult. This leads developers to choose between alternative platforms, and (ideally) choose one over all others. This choice, once made, is not easily changed; code written to one platform is only re-written to another with difficulty. In effect, once a developer chooses a platform, the developer is “locked in” to that platform.
This is what Microsoft considers a “platform”. This is not just Windows (or DOS) but also covers their programming languages and environments. They state that developers tend to pick one platform because its harder writing for multiple platforms. Picking one platform according to them leads to lock in which they seem to state as a positive. It is also a huge win for them if they can turn a product into a platform.
Critical Mass
“Critical mass” is a term borrowed from nuclear physics, in which it means “a sufficient mass of fissionable material to sustain a chain reaction,” in which a chain reaction is defined to be “a self-sustaining nuclear reaction yielding energy that causes further reactions of the same kind.”
“Self-sustaining” is the key phrase in this definition. Once started, the nuclear reaction naturally reinvests its energy in perpetuating and growing the reaction. Once the point of criticality is reached, no additional external energy or intervention is required to maintain the reaction.
So it is with evangelism. Once a platform has reached a certain degree of support in the industry, it becomes self-perpetuating (for reasons discussed below). Once this “critical mass” of support for a given platform has been achieved, new applications will tend to support it too, even in the absence of further evangelism.
Note that were are concerned with the “mass” of support by. Both the number and the importance of applications supporting a platform contribute to the overall mass of the platform’s support.
Mass in motion gives momentum, which can be defined as the amount of effort required over a given time to stop a body in motion. A sufficient mass of applications delivering support for a platform over a sufficiently short time, gives the platform sufficient momentum that it is, for all practical purposes, unstoppable.
Current economic theory suggests that the emergence of a dominant standard from a number of competing alternatives, is the result of “random events.” It is our job to ensure that there’s nothing random about it: we create the platform momentum that ensures Microsoft’s victory.
This explains the importance to Microsoft over the adoption of a platform. They need to reach critical mass via evangelism in order to get the machine rolling. Once they have enough applications and users, new applications will be written to support the platform as well even if they halt their evangelism.
C# platform development has been quite slow at reaching critical mass (and it hasn’t). This explains a lot about why Microsoft is acting so friendly towards anyone wanting to adopt C# and Silverlight. They need people to support these platforms in order for them to reach critical mass or they consider it a defeat. Mono on Linux, OSX, The iPhone, etc will help them reach critical mass. Talking about Silverlight and C# positivity will also help them reach critical mass.
Third-Party Applications and Business Solutions
Platforms, by themselves, are of little or no value to end users. Their value is derived from the applications and business solutions which take advantage of the platforms’ services in order to deliver real value to the customer. If we wish our platforms to be successful, we must ensure that our platforms are used by the applications and business solutions that customers want.
Third-party applications and business solutions are the measurement of the success of evangelism. The greater the mass of shipping applications that support our platforms, the more successful we have been.
The art of evangelism is concerned with increasing the mass of applications and business solutions that support our platforms, to the point of criticality.
This is more of the same stuff from above except more or less cementing facts. They added that end users don’t care about platforms other than the applications that can be run on those platforms. The more applications that support their platform the more customers they expect. This means that If people support the C# platform then this helps Microsoft gain more customers because their platform is seen to be well supported.
The Measure of Success
The success of evangelism is measured in the mass of shipping applications that support Microsoft’s platforms. Those applications that support a given platform, move it closer to critical mass, and thus count as successes; those that support a competing platform, move it away from critical mass, and thus count as defeats.
Microsoft has many platforms, and what appears to be a victory for one Microsoft platform may appear to be a defeat for another. If OLE is sucessful on the Macintosh (making Macintosh applications more powerful), is that a victory for OLE, or a defeat for Windows? If the Windows API is sucessful on Unix (making Unix applications less expensive to develop), is that a victory for the Win32 API or a defeat for Windows NT?
Diversionary tactics, holding action, and retreats may each seem contrary to the achievement of the overall objective when considered solely in their own terms, but taken in light of the overall conflict, may contribute to overall success. In the Chinese Civil War that followed World War II, Mao Tse Tung’s Army ran away from every battle, until they won the war. They knew that overall victory, not local victory, was the objective.
Thus it is imperative to measure each action in accordance with its contribution to overall, not just local, victory.
This is quite interesting on several fronts. They cement the fact that any application written for one of their platforms will help it reach critical mass. This means that F-Spot, Banshee and Tomboy all help Microsoft get its platform to reach critical mass. In doing so Microsoft can make more customers and in effect, more money.
Further on Microsoft even explains that the Win32 API on Unix is a win. OLE on Macintosh is also a win for them. Why wouldn’t the C# Platform on Unix also be considered a win? Microsoft explains these are diversionary tactics which appear to contradict their goals but not their overall victory. Even if they back away and retreat from a battle they expect to win in the long run. This would imply that even though standardizing and making a community promise for C# would appear to be their downfall, it only helps them in the end because more people will be developing applications for their platform.
Platform Example: Win32
* Win32 on Windows
o Windows NT, Windows 32s, Windows 95
* Win32 API on Unix
o Sun, WABI
o Bristol, MainSoft
* Win32 API on Mac
o Visual C++ for Macintosh
* Result: Win32 is the universal API
o We win!
This part only further reinforces their idea that their Win32 API on any operating system is a win for them. It further implies that they use the same logic in regards to C# being universal, and being considered a victory for them.
As a result of all this information I suggest free software and open source developers stick to alternatives to C# and other Microsoft platforms. There are options such as Java for those who need garbage collection. C++ and C for those who want a little more power and Python or Perl for scripting. Other alternatives also exist and are FAR too numerous to list but should be easy enough to find on Google. One I might recommend keeping an eye on is Google Go, which is quite new, and offers the benefits of dynamic languages with the speed of compiled code.
[...] This post was mentioned on Twitter by Steve Michaels, Ender. Ender said: Evangelism is War – The Mono/Moonlight Agenda – http://bit.ly/7xQwun – Read and retweet
[...]
[...] of our readers, Brandon, has written a long post to explain Microsoft’s “Mono/Moonlight Agenda”. To quote just a portion (it is a very detailed analysis): Many relevant points are brought up in [...]
Is this from the Protocols of the Elders of Redmond?
Definitely.
Ah, again and again and again people ‘discover’ these MS-documents and never realize that generations of conspiracy-minded folks have discovered these before they did and laid them out to the public before they did.
So my word to the author of this article: You are being a bore; we _knew_ this article before you did. EVERYONE knows. This is OLD. And the idea that Microsoft adheres to their now-not-so-secret-anymore plan that they laid out decades ago is ludicrous. Stupid. Gah. Get out more often.
I wouldn’t bother anwering at all but I think ppl like you are giving the Linux community a bad name. We aren’t all conspiracy-buffs who live in the constant fear of Microsoft taking away their favorite toy. Or even their freedom. Good god, all the illegal copies of MS software that are in use are a much more real problem to MS than Linux could ever be. Again, I feel like, ugh; man, get out in the fresh air!
The very term ‘conspiracy theory’ deserves a Godwin like designation. Anyone who uses it in any context is by default opting out of any legitimate discussion of facts and their logical interpretation. The term ‘conspiracy theory’ is both empty and self referential in its semiotics. It is a signifier that only displays the users total disregard for lucid and productive dialectic : either they don’t comprehend the matter or they choose not to. The term ‘conspiracy theory’ is toxic, sterile and ultimately only of use in a juvenile context.
stop confusing “conspiracy theory” with institutional analysis. stop it. stop it right now! bad dog!
Institutional analysis is that part of the social sciences which studies how institutions—i.e., structures and mechanisms of social order and cooperation governing the behavior of two or more individuals—behave and function according to both empirical rules (informal rules-in-use and norms) and also theoretical rules (formal rules and law). This field deals with how individuals and groups construct institutions, how institutions function in practice, and the effects of institutions on society.[1]
Since institutional analysis is focused on the systematic study of people’s collective behaviour in institutions, its ability to explain major political, social, or historical events is sometimes contrasted with the use of conspiracy theory to explain such events, since the latter focuses on explaining such events by a secret, and often deceptive, plot by a covert coalition of small numbers of powerful or influential individuals rather than by the systematic, regular, publicly documented behaviour of the institutions.[2][3]
http://en.wikipedia.org/wiki/Institutional_analysis
it’s like showing a dog a card trick….
This is by all means a conspiracy-theory because it is purely self-referential.
Get over it! Microsoft doesn’t need support from starving open-source developers nor from our lousy 1% market-base to reach a ‘critical mass’! Any standard that MS adopts has reached critical mass per se since MS is – listen closely – the monopolist in home-operating systems.
Thus .NET has reached critical mass simply by its adoption by Microsoft itself. No need for convincing US that it is good.
With Silverlight its a bit different – but here the opponent is not us, the open-source community, but ADOBE. And again, MS doesn’t need us to win the war, but big marketing-channels, like TV-stations, the Olympic Games, the World Series.
Get over it folks, we’re just not that important.
You are not either informed enough or intelligent enough in regards to this issue to own any kind of respectable opinion.
1%… in server Linux is nearly 50%. In netbooks it is close to 30%. Desktop doesn’t matter so much, it is about 5%. Smart Phones? 70% have a linux kernel. Windows Mobile is the most dismal product in the entire mobile industry with rapidly declining market share and profitability. Redmond laid off 5,000 employees in 2009. MS is borrowing money for the 1st time in its history. 1997 called, it says you have its opinions….it wants them back.
correction 70% for smart phones is linux + bsd (iphone).
Please check your logic; MS doesn’t need the 1% of people who use Linux on their desktop in order to market Silverlight on phones and …servers? Silverlight/Moonlight on the iPhone; now THAT is a different story. We (as desktop linux users) are utterly, utterly unimportant to Microsoft. So PLEASE finally get over your paranoia.
If you want to be paranoid, be against Silverlight on the iPhone or something. Or even better, don’t even buy one of these things. Oh, and don’t use Google. Evil, too.
Chris/EET, you might want to check out Netcraft.
Novell is leveraging Mono for use with enterprise servers.
GNU/Linux has far greater than 1% market share.
Any application written to their standards is a win, and it’s not paranoia when it is stated as fact by Microsoft. Old plans or not, they existed and I doubt Microsoft will do anything but advance and add more strategies to make those plans work better for them. Including hiring people to post comments on blog sites.
You are a moron who is immune to easily obtainable, documented facts.
yes, B. I think you have a paid astroturfer/corporate troll here. He is as predictable as all the rest, and spouts the same bullshit.
In my contact with MS-conspiracy-theorists it has always been true that you folks never brought any solid countering arguments except ‘you are a moron’ or ‘you are a paid MS-minion’ .
Argument: Microsoft just doesn’t _need_ you (‘you’ as in ‘desktop-linux users’). Neither as developers, nor as users. Investment in winning you over or destroying you doesn’t make any economical sense.
Counter-argument: ‘____’ (nada, none, nothing)
Apple IS a threat to Microsoft. Yet they play that competition very fairly; heck, they even saved Apple from bancruptcy once. Adobe is a threat to MS. And most importanty, Google is a threat to MS. Linux. Is. Not.
Perhaps you’ll be interested to know that Ballmer has said “All open source dev should happen on Windows”. Mono is just a step towards this goal.
Also claiming Google is evil is something that Microsoft loves to talk about. Even though they are even more guilty of committing evil.
The Register
The meaning of
is unknown to you?
No but perhaps you should stick to the facts, get some evidence and actually come up with a rational argument.
Well, you are one to talk!
Let me repeat myself:
Argument: Microsoft just doesn’t _need_ you (‘you’ as in ‘desktop-linux users’). Neither as developers, nor as users. Investment in winning you over or destroying you doesn’t make any economical sense.
Your counter-argument: Ballmer has said “All open source dev should happen on Windows”. Mono is just a step towards this goal.
That is no rational counter-argument. That Ballmer expressed his (understandable) wish that all open-source development should happen on Windows doesn’t make it any more plausible that Microsoft should devote its considerable resources to undermine Linux.
Some say, all open-source development should happen on Linux – but does that make an Anti-Windows conspiracy? I think not.
And BTW, you have insinuated that I said something about Google which I said not (irony-tag-smiley). Don’t try to casually slur that over with ‘No but perhaps…’.
Unfortunately, this conversation seems to end as another fruitless attempts to get any remnants of reason out of an argument with conspiracy-theorists.
I’ve given you actual facts and source material. All you’ve given me is your incorrect opinion based on invalid facts which you can’t source.
However I understand it would be against your contract to source the astroturfer’s manual.
Also I’ve stated that Microsoft wants Mono because it will make its platform reach critical mass. This isn’t about destroying Linux.
As always, following the traditional scheme of conspiracy-theorists, you have now – tadah! – declared me a paid minion. Well, hurrah.
And, as always, you repeat arguments which I rebuted before: .NET reached critical mass the day it was born because Windows is on 90+ percent of desktops out there. Our meagre percentage of usership (be it 1% or 5% really doesn’t matter) doesn’t make a difference to that.
Fight amongst yourselves people.. but hey pwnage… thanks for the posting… for some of us its the first time we actually saw this side of M$ and will forever change our future decisions. I didn’t believe your clippings could be legit as it seems so Facist… so I did some checking around myself and found this stuff is all very real.
As a Christian I see it like this:
Microsoft are the Pharasees…
Apple are the Saducees…
Open Source is the only “Free” way offering freedom for all…
As Jesus said (and I apply this to everything in life) “Ye shall know them by their fruits”.
Thanks again… off I go to make a donation to Free Software…
…and in case that you say ‘Mono’ but mean ‘Moonlight’, I already said that major redistribution channels are important for MS’ battle against Adobe, not Linux. Whether we can watch what, say, NBC or whonot transmits in glorious, animated, colorful bloat is completely irrelevant for the success of either Flash or Silverlight. The decisisive step for MS is to convince the NBC to actually _do_ it.
I hope you don’t chicken out and approve these comments forthwith.
1% Linux Market Share = 100% Dishonesty
By Carla Schroder on May 5, 2009 9:45 PM | Permalink | Comments (24) | TrackBacks (1)
A PC desktop market share analysis of questionable reliability is being eagerly swallowed by Redmond’s tame battalions of “tech journalists.” Folks, if you’re getting paid to be gasbags who are not required to publish anything of substance you’re doing fine. If you’re being paid to be real journalists, you’re not succeeding. It’s like you all have Alzheimer’s and have forgotten the previous stories that you dutifully parroted stating completely different figures, like “Linux desktop market share to reach 6% in 2007.”
Our very own GreyGeek came up with some different reports:
“Matt Assay said it was at 2.02%
ZDNet reported on Feb 24th, 2004 that the 2003 Linux desktop market share hit 3.2% and expected it to hit 6% by 2007.
In 2005 they reported that the 2004 saw the Linux desktop at 4%.
I believe that the all the ZDNet figures were spot on. If anything, the Linux desktop market share has continued to increase and is probably currently at 8-10% and rising. Dell and the other PC OEMs wouldn’t have invested in selling Linux pre-installed if it appealed only to less than 1% of the desktop market.
It is quite obvious that NetApplications latest “report” is merely Microsoft’s continuing attempt to control the news about Linux’s success in replacing Windows on the desktop….”
http://blog.linuxtoday.com/blog/2009/05/1-linux-market.html
Chris,
Nothing mentioned is a “conspiracy theory”, you might want to go with “conspiracy fact”. These are all documented facts and based upon the research I’ve done I’ve reached certain logical conclusions. Nothing you have mentioned can be proven as fact, you’re spouting the conspiracy theories.
Can you please source some proof of your correlation between .NET reaching critical mass and the success of the Windows desktop in general? Not everything Microsoft releases for Windows becomes mainstream. They have no market share of the anti virus market for example despite having a free as in beer anti virus product. Silverlight is also not mainstream, and Adobe is killing it. I would also support Adobe over Microsoft, and Free Software over both. HTML5 is going to make Silverlight and Flash obsolete.
Can you show me any site out there that shows C# beating C, C++, Java or any mainstream language at all? Just because they released it for Windows doesn’t mean every windows user now has .NET and are running applications based upon it. A lot of developers are still coding C++ on Windows. XP didn’t even come with .NET for example and ATI had to bundle the redistributable with their driver due to all the complaints about missing files. This isn’t the case with Vista or Win7 but it does not mean every piece of software written since Vista and Win7 came out was based on anything from .NET
I would be very surprised if everyone who took the effort to learn C would voluntarily stop coding in that language and start using C#… For commercial software though, .NET means faster and less costly application-development – perfom a seach on “based on .NET” and similar and you’ll see.
Thus all talk about .NET needing to reach ‘critical mass’ is purely academical; .NET as a framework is in place. All it needed was a broad (potential) user-base. Everyone who uses Windows will find nothing odd about installing .NET from MS should another application he/she wants to install demand it. Not that it needs a sentient decision; such applications simply bundle .NET runtimes or pull them in.
About market-shares – here is a rather solid one from last year; I hope you find it inconspicuous as it reports a declining Windows market-share:
http://www.tomshardware.com/news/Windows-Mac-OS-X,5813.html
Excerpt:
Windows: 90.89%,
Mac OS X: 7.94%
Linux: 0.80%
Yes, of course these figures can have changed since that – but certainly not radically.
Yeah, I just read that GreyGreek’s foaming-at-the-mouth comment to the statistics and – except for hot air – there’s nothing in his ramblings that casts a shadow of doubt on the figures. How blind-sided must one be to accuse MS of somehow paying (again, this accusation of his has no base in reality) for study that reflects the decline of Windows usage and the rise of Mac OS X?
Solid? Is this a joke? You posted a link to Toms Hardware, which also doesn’t source it’s data. If you look up the company “Net Applications” for example they cannot provide data or source material as to where their findings actually come from. They provide no explanation. Is this web browser hits to their site or another site? Flawed. Is it based on units sold? flawed, as Linux is mostly used without cost. Is it based on North America, the world? No clue. I have no idea where they got the data that says Linux only holds 0.80% market share.
Bruce Byfield has this to say on Net Applications:
Why are you ignoring Netcraft? Netcraft is actually reputable and industry standard for determining server market share. The server market does matter as Novell has a Mono server offering which ties in to Apache. This would also boost Microsoft’s platform. No desktop required.
You’re also ignoring my request for a correlation between .NET or C# being critical mass right now due to it being released for Windows by Microsoft. You also can’t show me any sort of evidence which shows .NET and C# being dominant platforms. This is what I asked for. Your 1% linux market share arguement is invalid until some sort of data is shown. I could make a pie chart in Kspread that shows what I want people to think too.
When rational argument fails, ad hominem attacks occur on the authors.
Why is it that the same people who call others astroturfers yell blue murder as soon as anything as daring as’blind-sidedness’ touches the outer regions of their aura?
> Why are you ignoring Netcraft? Netcraft is actually reputable
> and industry standard for determining server market share.
Because we are not talking server market when we are talking about how little it matters whether you use or don’t use Mono and Moonlight on your desktop. Linux market-share on servers might as well be 100% and it still wouldn’t hurt .NET’s success on the (as little as you may like to hear it) predominantly Windows-desktop.
My senile father inlaw of age 71 is using Mandriva linux, and he loves it. It doesn’t give him the problems with malware windows did, and KDE is so similar to his XP UI that he required no retraining. He still calls me up and asks for his account password once in a while, and it is his first name…
This “Chris” character is either a fraud or a maniac or both. Ignore his ranting.
The future of computing is in smart phones and tablets, and they are not going to be using a MS operating system ; they will be using Android Linux.
Also the origin of the data used in the statistics of that Net Applications company is by far not as hazy as you make it sound; see http://marketshare.hitslink.com/ for an explanation. Ye who searches, shall find.
Forget about all elaborate schemes for a moment and just imagine Microsoft without any support from Linux users or Linux companies. That’s what you’re asking for, right? See? It doesn’t change a whole lot for the big picture. Microsoft can perfectly well thrive without our support and even with our stern opposition. All that politicking and conspiracy-theorizing simply does not matter.
I really don’t see how the market share can only be 1%. Ubuntu has over 10 million users, Fedora has over 12 million users. I’ve never seen hard numbers on SuSE, Slackware, or any of the other myriad numbers of distros out there.
Our marketshare has to be somewhere in the neighborhood of 25 or 30 million users. Thats gotta put us closer to 10%.
Microsoft themselves stated Linux’s market share is bigger than Apple’s. http://www.osnews.com/story/21035/Ballmer_Linux_Bigger_Competitor_than_Apple.
Andrey, Stever Ballmer himself said that the PC market is about 300 million machines (http://www.osnews.com/story/21035/Ballmer_Linux_Bigger_Competitor_than_Apple), which puts known Linux PC deployments at around 10% (3% Ubuntu, 4% Fedora, and Suse is likely around 3-4% as well). I’d call 10% the lower limit, as Linux (according to Ballmer) has greater desktop market share than Apple, and Apple has 8-9% market share.
“We collect data from the browsers of site visitors to our exclusive on-demand network of live stats customers. The data is compiled from approximately 160 million visitors per month.”
Well Net Applications is something I’ve never heard of so its not received hits from my 5 linux boxes and I’m sure its missing traffic from quite a few people. This would be like using Distrowatch to try and determine the most popular linux distribution by hits.
Actually Sandala, Fedora has 20 million installations. This is irrefutable because they track this. However it does not show the number of users who did NOT submit their hardware profile. Ubuntu probably has even more than you suggest if Fedora has 20 million.
And Chris, i’m not just talking about the desktop market. Read what I am saying: Mono is available for the server, so servers are relevant! Also, I’ve not personally attacked you.
Net Apps are doing representative studies, that’s the best you get in statistics.
Ubuntu measures their install base in just that – installs! How many Ubuntu installations have you performed? Every single one of them is counted in those 10 million!
Personally, I make a rough estimate of 1 Linux install per week (VMs also count). 10 million might sound a nice figure but you must take that with a BIG grain of salt. I’d say, 2-5 Mio Linux users _in_total_ would be realistic.
If you don’t believe me, do a simple representative study yourself: Go out in the street and ask every person you meet what OS they use. I would be VERY surprised if even every 50th person uses Linux. Mac OS, yes of course. Linux, no.
Net Apps study only counts people visiting the Net Apps website. Hardly scientific at ALL and does not represent world wide usage in general. IDC is even saying you are wrong. I also find it ironic you say servers don’t matter but yet people who use Net Apps are running their own web server or website. Consistency?
If I did this I would have data based upon my home city, not the entire world in general. For example, China is going to have it’s own Linux distribution and it has a much higher population than North America. What would that do to global market share? A LOT. Look at Brazil, they migrated their schools over to Linux. How does that effect global market share? A lot. I would say those two countries alone account for over 1% of the linux global market share simply by using Logic.
I personally know a lot of Chinese, Koreans, Japanese. They all use (almost always pirated) copies of the latest Windows. So, no Linux joy there. But, reading on, I see that you are entirely on the wrong track anyway; you seem to think that we are only talking American usage stats here.
admin said: 2009.12.28 14:14
> Net Apps study only counts people visiting the Net Apps website.
NO, oh; no. No I see that you completely misunderstand what Net Apps are doing. Net Apps process accesses to over 40,000 websites world-wide, weighted by country, according to the number of internet-users. This gets them by far more accurate results than any other poll I know of.
Sorry, only German-language links:
http://www.heise.de/newsticker/meldung/Internetstatistik-Rasche-Akzeptanz-fuer-Windows-7-856130.html
http://www.heise.de/newsticker/meldung/Net-Applications-aendert-Gewichtung-bei-Browserstatistiken-749521.html
Chris, at most it only represents sites that use Net Apps. I don’t use them. I have many web sites out there. 40,000 sites isn’t a good sample for world wide linux market share dude.
I would really love to see a google survey as it would be far more accurate.
Please research the difference between a survey (in which you voluntarily decide to participate and thus persons with an agenda weigh disproportionally strong) and sampling which – if done right – yields much more accurate results.
No, you can’t top a representative study that monitored/sampled accesses to 40,000 websites world-wide and weighs them according to those countries’ computer-literate population by ‘a google survey’.
But statistics is science, and I might be asking too much of you to follow me here.
Also, the sites Net Apps monitors are not a random sample; plus they also are weighed according to their relevance.
And no one in the entire industry agrees with Net Apps, apparently not even Steve Ballmer, who stated that Linux has more desktop market share than Apple. The real money is in server, and MS knows this. You can not get away with running a business with pirated server software, and they charge per CPU. The BSA will cut your balls off if you try it. 50% of the webservers that create the internet as we understand it are linux servers. For a long time MS had to use Solaris to run hotmail, as Windows Server couldn’t handle it.
Look at this fool, at one moment he is all ‘go out on the street’ which is wildly unscientific. Now he is trying to abuse statistics, when he doesn’t even have data to source! What this twit knows about “science” comes from wikipedia, and no further, if that.
Yes, yes; excuse me for being so ‘ad-hominem’ at people loosely connected to you when you yourselves treat me with such diligent courtesy although I happen to disagree with your opinions…
5% son, 5% as I stated… that is conservative.
http://www.w3schools.com/browsers/browsers_os.asp
Linux went from 1.25 percent in May of 2007 to 2.02 percent in March of 2008. That is 61.6 percent increase in market share in nine months. [Put another way,] that is 82 percent annual growth in installed computers.
These numbers don’t match up with similarly gathered data from Hitslink, which has Linux market share still hovering around 0.61 percent market share. But even Hitslink shows strong growth over time: 41 percent-plus growth since May 2007.