Slashdot Mirror
Debunking Linux-Windows Market Share Myths
bc90021 writes "Nicholas Petreley has a great article over at LinuxWorld explaining why it seems that Windows has such a high market share when 40% of developers are focusing on Linux. From the summary: "There are dozens of reasons why people have underestimated how quickly Linux has been grabbing Windows' market share. Windows starts out with a false boost and maintains its illusory market share even as it gets replaced by Linux. In 2004, don't be surprised when Linux overtakes Windows to become the main focus for developers.""
90% of the end users wonder "what is Linux ?". To them - Windows is the computer.
But that doesn't count, since many browsers are configured to lie about what they are to work around stupid JavaScript/Website constraints.
95.7% of statistics are meaningless.
"Little does he know, but there is no 'I' in 'Idiot'!"
How is that an effective way to check? As a network admin I have 12 Linux servers and a Windows machine as my workstation. Guess which one racks up all the web stats on Google.
Keep in mind most Linux machines are servers and most people don't browse the internet with their servers.
Zeitgeist, we can assume, overwhelmingly refelects desktops, not servers. The article's author doesn't make to too clear, but it sounds like he's (mostly) talking about servers.
I'm also, I have to say, doubtful that any browser-sniffing gives an accurate picture of what people out there are using, because so many people set Opera et al (on any OS) to report itself as IE for Windows. Personally I think that's a terrible idea -- if I find a site that refuses to work with my preferred setup (Mozilla on OS X) I figure, well, what the hell, I didn't really need to look at that site anyway -- but an awful lot of people do it.
The correlation between ignorance of statistics and using "correlation is not causation" as an argument is close to 1.
Apple has a quarterly SALES pentration/market share of 3% to 4% but has an installed base somewhere around 11%.
Of course we all know what "Mac People" like to point out about the RISC processor being 40% faster than an x86 and in most cases 75-90% than a Celeron. Who knows about the Centrino. (What a poor name to choose - "trino" anything sounds miniscule)
It's the same way in the Linux community. Most versions of Linux run faster than Windows on the same hardware. (true in some cases on Mac hardware than OS X/OS 9 as well) There is a significantly higher number of 'nix users than M$ would like us to believe. I don't know the member numbers at Sourceforge & Slashdot. Not all are 'nix users but it is significantly high. That alone is large enough base. M$ wants everyone to belive that only mainstream/mass advertising companies (like themselves) have market share. They like to take advantage of the public psyche.
Yell & scream & rant & rave... it's no use... you need a shaaaave ~ Bugs Bunny
I agree that Linux is taking over Windows shares, however an article like this proves nothing. There is no statistical information, what is the population size? What type of companies were part of this "study". How were the companies included, by picking them, by them asking, or just a random sampling.
Sure it's great to think Linux will eventual kill windows, I just don't see any proof of that yet.
So the margin of error is at least in the 5-10% area? That sounds quite large, for a survey that purports to take in a wide range of developers. Methinks the author is taking an overly optimistic view of the subject matter - but that's not really surprising...
Stop by my site where I write about ERP systems & more
That carried them to a point where the leverage point was effectively "if you control the desktop you can win the servers (killing off novell, banyan ...). This in turn led to "if you control the desktop and departmental servers, you can muscle into the enterprise" -- fortunately (imo) they've had less luck - despite considerable effort - in killing off Unix/Linux/BSD.
So far at least. I don't think this games very predictable, (and the LW article is *very* thin on data, but there certainly is a deep groundswell of good things happening in OSS, and virtually all big-iron oriented code now targets Linux along with Unix).
However it's perfectly reasonable that as developers move (back) to *nix, eventually the market will follow.
(Remembering Grace Slick of the Starship singing about 'egg-snatchers' -- dunno the Borg's a big target, and elephants are best eaten a bite at a time)
Linux is Linux, if One need clarify their dist: <Dist>/GNU Linux
bsds are of course just BSD
Windows is still king when it comes to desktops. Not only due to a huge market share but also FUD (not all unfair) towards Linux from common users.
If we define the market as computers as in CPUs, I'd say that neither Linux nor Windows wins or comes even close. There must be far more 8051 controllers out there running a hand coded snippet of control code than there are 8086 derivates.
The GNU/Linux movement shouldn't say that we have beaten Windows, let's relax. Rather we should say we can beat Windows, let's work as much as possible producing quality software.
I have to take this article with a pinch of salt - I know it's hardly empirical evidence, but almost every developer I know is not installing Linux over Windows, rather they're dual-booting their systems to run both Linux and Windows. Maybe this will change in the long run, but I doubt the swing will have been made by 2004.
I'd love to believe what he says, but it doesn't quite ring true from my own personal experience.
Life is like a sewer; what you get out of it depends on what you put into it...
Developing in a windows environment, even with something like cygwin or Visual Studio.NET just plain sucks compared to actually being in linux.
How is this? Sorry but having coded for many years using Borland's tools in Windows I found it very very difficult to adapt to Linux development: no context-sensitive help, no organized documentation (yes, lots of documentation, but no "central" organized index which means a research job for a fucking function declaration), no intellisense, no autocompletion, and having to resort to home-brewed makefiles is just a pain in the ass.
Could you please explain which tools are you using for development, so I can use them too and make my life easier? :-)
I can imagine a future when although a vast majority of people are using Linux, Windows still shows up in the "statistics" as the most popular OS, just because it is shipped by default with most PCs. As long as the Microsoft spin doctors could keep the myth going, manufacturers would still ship PCs with Windows pre-installed by default, thus reinforcing the "statistics". In effect, Microsoft could still claim the "Microsoft tax" even if nobody was using their product.
If I seem short sighted, it is because I stand on the shoulders of midgets
I think the games and the gaming communit, especially with the overwhelming mass of MMORPG's and the ilk, make up a fairly significant portion of software development budgets globally. For me at least, this has been a major valley in me switching to 'nix 100%. There isn't one single game I play (yes, I'm a gamer ;P) that's available in any flavor of 'nix. A server dedicated server-side products, yes. Clients (ie: actual game), no.
So a magazine called 'LinuxWorld' has an article on how Linux is going to overtake the most popular OS on the market! Wow! That's really impressive...*sarcasm*
No offense to Linux, but 'MacWorld' and 'PCWorld' magazines probably could write the same article about their systems. Can't trust a media produced to promote things.
"I only speak the truth"
Karma: null(Mostly affected by an unassigned variable)
Ironically, if MS gets what they want, and the XBox2 is a roaring success, there's no reason that we won't finally see the long proclaimed death of PC gaming, or at least the marginalization of it. Consoles need a little work, too, but nothing's stopping them from getting a keyboard and mouse needed to finish the control paradigm from where it stands now.
If no one's putting out games for the PC, why not just buy a console and then wipe Windows off your box? That's a good question, and Microsoft might just show us the answer soon. It's sure as hell more economical than running the infamous PC upgrade treadmill.
-Erwos
Plausible conjecture should not be misrepresented as proof positive.
Aren't we missing something very important here - client vs. server breakdown? The article seems to assume that it's all about clients, but I just can't believe 40% of developers are doing linux client work. Servers, of course, are another thing altogether.
And of course we all know 1 server box (or 1 cluster) can keep several developers happily employed for years, maintaining & enhancing some line of business app/web application/etc.
So it seems the relationship between the number of developers and number of installs is a pretty weak one.
Read reviews of shopping cart software
Just like to remind evreybody that this is from LinuxWorld. Not exactally a bastion of unbiased reporting when it comes to operating systems.
Mod point free since 2001
nicholas petreley is a moron. Let me prove my case.
He said:
1. Evans Data Corporation hired me to help out with a research report focused on Linux developers.
2. Of the developers surveyed, more than 50 percent who now develop primarily for Linux used to develop primarily for Windows. Only 30 percent used to develop for some other Unix or Unix derivative. In case you missed it, the operative phrase is "used to." In other words, this is not a prediction of an emerging trend. It is cold, hard information about what has already transpired, and it withstands the scrutiny of a jeweler's eyepiece.
Now put one and two together and what do you get?
A statistics that is just plain wrong. By only focusing on linux developers, you cannot get accurate statistics.
For example, assume we have.
Microsoft developers 1000 people
Linux developers 10 people
of which
5 used to develop primary for windows
3 used to develop for other unixes
2 otherwise.
How can you prove anything by only doing a survey on linux developers? To get accurate info, you need to either include all developers or take a random sample of all developers.
Rolling out 1,000 desktops requires virtually no developer input. Rolling out a unified health and social care workflow system (which is what I'm working on now) takes a lot of developer man hours - but when it's finished it will sit on one (Linux) server (and be accessed by hundreds of desktops, most of which will almost certainly run Windows).
This does not matter
We are not playing a numbers game. We don't need to take over the world. The fact that most users still prefer to use something else on their desktops doesn't make Linux a bad operating system, or a failure, or anything like that. Linux is very successful in a lot of niches. If it ultimately becomes more widely used than Windows, well, that will be interesting; but it won't make Linux any better (or Windows any worse).
I'm old enough to remember when discussions on Slashdot were well informed.
"40 percent of developers"? How should I understand this figure? Do you count as one each person who just happens to develop ANYTHING, even some obscure freeware of no importance? "40 percent of developers" sounds like a shampoo commercial to me ("your hair will be 50 percent more healthy").
The irony here is that Windows gets an unfair market-share boost because it is inferior to Linux and requires more installations to do the same work.
While I wholly agree with Nicholas on most of the article, this line doesn't seem to help the community. One of the stereotypes of linux users is that we think it is better than everything else in every application. This is simply not the case.
I recently helped my grandmother purchase a computer. Her budget was large enough to get an iMac, so I suggested it, and she purchased it. I made sure she got the extended warranty from Apple because it includes phone support. She would not dial the number no matter how much I suggested it.
She liked to get face to face help, and everyone she knows uses Windows. They couldn't help her with her Mac, even though it's the simplest thing to use. In this situation, I bought the iMac from her and replaced it with a Windows PC, and she is now satisfied.
Every system has its place, and ignoring this fact will reflect badly on the Linux community until we realize it.
In the long run, we're all dead.
The data gathered doesn't answer this question, but I have to wonder.
MS has a well known habit destroying any successful developer of Windows software, either through theft in the guise of a buyout or making a knock off product then welding it into the OS.
While the Free Software community is known for making Free versions of proprietary software (also a potential liability from a proprietary developer's standpoint), it is at least not a total wipeout. They at least know that the OS won't be changed to lock their app out.
The more MS expands, the clearer it will become to developers that Windows development is a good way to get stomped on.
Not how the linux developers don't support the cause:
1. They usually download it for free rather than buy a distro.
2. They wiped the hard drive free of windows, meaning they didn't buy from a white box Linux vendor.
With friends like this who needs enemies.
So far the only Linux machine that has accessed my site is my own - the site has been up for months and is quiet - about 20 - 30 visitors a day, but the results are still very disappointing.
The server world is very different - I am hacking away at perl for apache on Linux right now.
One word: Java.
I haven't got any hard data, but I'd be very surprised if Java didn't represent a large chunk of those 40% of developers. Basically it means that whenever developers get the choice of platform (not for deployment, but for the actual development) 40% will choose a Linux environment.
And there are huge swathes of commercial Java apps. Most if not all have been made to work on Linux.
At least, that's what I use Linux for; and most of the other people I know who use Linux on their workstations do it because they program in Java.
-S
I do consulting, my goal is to line my pockets with the most amount of money possible.
:-)
I give the customers two options.
1. A windows system with a box delegated for each task and the required MS licenses.
2. A single custom Linux box.
I then explain Linux has no fees associated with it, but it's more expensive to setup. If they go the MS route they will need to pay a few thousand in license costs, it will cost less to initially setup, but more over it's lifetime to maintain, and to reach the same performance levels I will need multiple Windows Machines (don't even talk to me about running exchange with anything else).
My client wins by spending less, and I win by making more.
These are server instances though, I have only ever done a single Linux desktop at a company, and that was for a limited use workstation in a remote part of the country, it can not break unexpectedly.
This is only me, I know the legions of cloned MCSEs outnumber me, so for every Client I flip to Linux they retain many more on the MS side.
Run Linux and profit
Any company which wants to stay in business knows that it's the Customers who drive the business. Why do you think Motorola is such a crappy company? It's run by engineers! All they do is create, create, create, but never bother to ask "does anyone want this crap?" I have Linux at home, but that doesn't mean I'm blind ... customers drive the market, not developers. Convence John Q. that Linux is superior, and then everyone will use it.
Everybody here is like "I'll take that article with a grain of salt", or "Statistics don't mean anything".
It seems a lot of people missed the point. That article focused less of stats and more on ways they can be interpreted.
The author is trying to present some alternative ways of interpreting the usual old Win vs. Linux stats that we hear so much. He isn't trying to carve anything in stone. So many of you engineering folks see things in black or white, but this article says there is a grey area. You can look at the grey aread two different ways and it can seem black or white depending on how you look at it. He's not trying to tell us that it's all black or that it's all white.
He makes a great point that there are key ideas in those stats that are often overlooked and are seldom addressed by the Linux community, and never addressed by the Win community(Win preinstalled, Lin more efficient requiring less installs, unsold copies of Win at Best Buy, etc). This is a very insightful article, and while it may be mostly speculation and interpretation, the one fact that we do know is that the Linux developer/user base is growing and will continue to grow(so let's not get hung up on how fast, etc).
Servers are invisible. Linux could run on every server in the world and remain essentially unknown to desktop users.
No matter how good Linux is from a technical pespective, it won't threaten Microsoft on the desktop until Linux developers offer consumers a non-ideological incentive to go to all the hassle and risk of abandoning Windows. I think it will take something along these lines:
1) Create Linux applications that do compelling and unique things that Microsoft apps don't (being "as good as" Office won't cut it for most people, any more than Texturized Soy Protein outsells real beef);
2) Slap a $39.95 price sticker on those apps, write good documentation, stuff 'em in cute shrinkwrapped boxes and get them onto store shelves;
3) Create a "wedge" of non-geek, non-techie "real" people using Linux and exploit their existence to the hilt. Convince ordinary people that Linux is for them. (See Apple's "Switch" campaign.)
4. Build Linux installation software that allows a new user to stick a CD in the slot, click "Go", walk away, and come back 15 minutes later to find Windows gone, Linux up and running, and all his Windows-created data and files preserved and migrated to the right Linux apps.
Good luck, and have fun.
-- Slashdot: When Public Access TV Says "No"
I was reading this article and thinking to myself two seperate thoughts: "Well, that's odd...", and "Uh huh... who cares?".
I work for a non-IT consulting company. Me and the team of ~20 developers here write software for engineers to use in petroleum engineering consulting. All of the software, 100% of it, it developed for Windows. I look around and see 0% Linux developers, and 100% Windows developers. But, alright, obviously my survey is baised. However, I only know one single developer here who has ever used Linux, or any operating system other than Windows. The two of us both have our shiny Powerbooks sitting next to our desktop computers while we work, for e-mail and web browsing and the occasional graphics work. I think the statistic that only 50% of developers use Windows is rather odd... since 95% of users are using Windows. Are there huge fields of programmers who develop cross-platform software and trust that it will work in Windows without testing it? Or develop server-side software only, which never sees a user?
Secondly, who cares. I look at a project like Mozilla, and I can see that a lot of the developers are on non-Windows OSes. But I think the majority of even Mozilla users are Windows users. I advocate Mozilla to my friends and family, installing it on computers and replacing IE/OutlookE, and everyone is happy. They're using Mozilla and Windows, and I think this is highly common. FTL even replaced the 'INTERNET' icon on his grandmother's computer with Mozilla, and I believe the only comment she ever had was about the cute dragon. Developers may be using non-MS platforms, but they're still developing for users who are in Windows. Right? Or is half the world using Linux on their desktop, and I'm in some la-la land?
I'm also not seeing anything quantitative other than a claim that he's spoken with market researchers. I'd like to see some names of the people surveyed or at least information about what the surveyed developers were/are doing in general.
I smell an unsubstantiated claim!
// -- http://www.BRAD-X.com/ --
A large number of users buy a computer with windows xp installed. I've been looking recently and can't find a cheaper "mainstream" retailer that sells linux boxes cheaper than windows. In fact I just bought a laptop. Dell had the cheapest one that I liked, and it came with xp. Finding anything comparable with just linux on it I found I would be paying out the ass to get linux over windows.
Sometimes windows is cheaper, and definitely easier to find on new computers...plain and simple
I think what Petreley says about developers programming for the server might be close to being correct-- Linux is taking market share from Microsoft, and I think it could extend to Windows server installations as well as Unix. I don't think that Linux is really having an impact on the desktop yet. The other thing he doesn't mention is whether this developer survey is just the USA or whether it's international-- it seems quite probable that Linux has more mind/market share in Europe in particular, than it does in the USA.
My question is, why would someone go to Circuit City or Best Buy to purchase a machine that has Windows pre-installed, then take it home and install Linux? Yes, yes, you can create a dual boot system (not easily), but it would seem to be easier to buy a "naked" PC from an online builder and save yourself the Microsoft "tax", then install Linux. I realize that MS is trying to clamp down on OEMs, but naked machines are still available out there. I bought one just to assure that I could reinstall Windows 2000 (which I got elsewhere) if I had to.
Petreley has a reputation as cheerleading for Java, too. I am surprised he didn't tell us how many of those Linux developers are using Java. I think there's a trend toward Linux, but it is developing less dramatically than Petreley would have us believe.
Always look on the briight side of life! (whistle, whistle)
Nicholas Petreley worked with Evans Data Corporation on a survey which asked Linux developers about there preferences, and he learned that 50% of them were no longer developing Windows software.
Somehow from these figures, Petreley concluded that Windows has signifigantly less developer mindshare than Linux.
In other news...
99.9%(with a margin of error of 5%) of respondents at a KKK rally were against affirmative action programs, from this we can clearly conclude that tea in China is selling at 50 cent per pound.
"Consider this: most linux users have a static IP but a large fraction of MS users will have a dynamic IP. So if they are counting unique IPs it will have a heavy windows bias. "
First, that's an unreasonable assumption. Second, I rather suspect that google.com is calculating statistics on all hits, not just unique visitors. There's no reason to do otherwise, at least not from a statistical sampling point of view.
"Proxies. Again, more linux users could be behind a proxy (a few hundred linux users at my univ go through a single proxy) than windows users "
Many Windows machines are behind proxies. We have over 5,000 machines at my company behind proxy.
Don't confuse your experiences at University with the real world... they aren't remotely close.
"Third, some factors similar to those described in the article could be at work (linux more efficient ==> less linux servers for same job). Maybe linux users are more efficient googlers? I think this is unlikely, but still a possibiility. "
Well from what I've seen Linux users are less likely to be interested in facts, so maybe they don't care enough to go looking up answers to questions.
(Oh yeah, that was sarcasm)
"Fourth, it doesn't agree with my webserver stats (i.e, counting the hits I get from google searches). "
Yes, clearly your anecdotal evidence is more accurate than massive statistical sampling.
(BTW, that was also sarcasm)
"but on the whole it makes me think that some combination of the factors above may be at work in decreasing the perceived share of linux."
I've always said... Don't let facts get in the way of a good argument.
If you want to believe Linux is really popular, and mass adoption is just around the corner, be my guest.
completely disregarding the notion of "user preference,"
User preference is highly overrated. Users will often prefer an inferior and inefficient way of working just because that's the way they were shown first.
How often have you gotten that annoying "Would you like to submit this to Microsoft" box?
It take more faith to believe in evolution than it takes to believe in God
"Man" pages?
HAHAHAHAHAHAHA
The dickheads that write those things don't fucking realize that not everyone wants to read documentation that requires a phd in engineering to understand.
Of the developers surveyed, more than 50 percent who now develop primarily for Linux used to develop primarily for Windows. Only 30 percent used to develop for some other Unix or Unix derivative.
But since Linux is a posix-compliant UNIX variant, many apps developed for it can be ported straight to Unix flavors by simply recompiling. Windows is a whole different ballgame (unless you develop with the Wine libraries for portability).
Drivers need a bit of porting, but are still 'way closer to Unix than Windows, and apps that use Lunix-specific features will need some tuning (or just not go if the whole POINT is to interact with the Linux feature).
But when a developer switches from Unix to Windows his work is likely still available in the Unix world. (Perhaps moreso, if he's GPLing it now.) Those that switch from Windows to Linux are pretty much GONE on the Windows side of the world.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Well, he may have written what you would call "another gnome sucks" article, but for a lot of people that article made a lot of sense. He actually penned the feelings a lot of users have about Gnome. Please, this is not a flame war about Gnome/KDE, but about how Linux spreads and how difficult it is to get exact figure on how many people actually use Linux. And here I agree with him. The nature of Linux makes it hard to count. The nature og how MS spreads also makes it hard to get a realistic count.
I would say it was a good article with a lot fo valid points. MS is loosing market and fast too.
If you mod me down, I *will* introduce you to my sister!
Actually, I have a laptop that I instlled Windows XP on about a half a year ago to give it a try (I am a 100% Linux user at home, only use Windows at work). The laptop has been up for as long as 90 days at a time and has only had to go down because I wanted to shut it down. It could probably run longer than the 90 days I've seen so far. I have had several applications crash, but no OS crash to date. However, I will also say that Linux has been just as stable for me. My record uptime so far is 299 days. I lost the uptime because of a power failure and didn't have a UPS on the system. Current uptime on that system is 51 days. It runs as my PDC, WINS server, internal and external DNS servers (running to instances of named), SAMBA File and Print server and mail server all on a Celeron 400 with 64 Megs of RAM. Same thing with this box as the XP Pro box, no OS crashes, only application crashes every so often. At this point, I'd say that a custom installed XP and a custom installed Linux done by someone who KNOWS what they are doing are equally stable. The major benefits with the Linux box are price and licensing. Since I build my own Linux from scratch I don't have to pay for the OS distribution or licensing. I didn't have to pay for XP Pro for my laptop either since it's not my laptop, but my employers. So, if you've had crappy experiences with either OS, chances are you haven't researched the problem enough to resolve it. There is no substantial benefit of going with Windows for me, so I stick with Linux. It does everything I need it to. Inlucding video capture, video editing, and games. With RedHat 8.0 the stakes just got higher too.
-"...bad old ideas look confusingly fresh when they are packaged as technology" - Jaron Lanier (Digital Maoism on Edge.o
"The problem with Linux is that, it's not as pretty as windows. And mose users just want something that's really pretty when they turn it on. And quite honestly if your computer already comes with XP on it, there is no point to downgrade to a Linux distro."
If you believe Windows is "prettier" than Linux, you should get yourself a copy of RH 8.0 and take about 5 minutes to play with themes. Take a sec to go download some more, and notice how your entire interface can look like Mac, Windows, OS/2, chrome, wood, sparkly colors, etc. All of the folks that I know who use Linux (fellow developers, friends, even my folks at home) think it looks much better than Windows. Linux can't be called ugly.
And regarding your "downgrade" to Linux comment, if you're referring to program compatibility, I can't argue. If you're talking about features, speed, etc., you need to use linux [again].
I'm a developer by trade and a sysadmin in my spare time. I read all sorts of documents and files, I do graphics and organize all of my data, check weather, etc. The only thing I need Windows for is games. And if you think this is only possible because I'm a developer, I'll refer to you my parents, college friends, non-programmer co-workers, etc. for references.
~Dalcius
Rome wasn't burnt in a day.