Slashdot Mirror
Ask Nicholas Petreley About Linux Usage Statistics
This Slashdot discusssion, about a story Nick wrote, is already going (and heated). I did a NewsForge interview with a SuSE rep who quotes an IDC study that says Linux desktop use will double by 2004. Sounds nice, but how reliable are all these statistics? Nick's been studying Linux use in depth lately, so let's ask him directly what all of these numbers mean, if anything, and how IDC, Evans Data, and other analysts get and massage them. We'll post Nick's answers to 10 of the highest-moderated questions as soon as he gets them back to us.
Do you place any credibilty in the tendency for certain analysts to derive things like a "mindshare index" from arguably disparate sources?
The difference between stupidity and genius is that genius has its limits.
The truth of the matter is, linux users these days install it themselves. even if i recieved a preinstalled linux box, i would reinstall, repartition, etc. Plus everyone likes a different distro.
I will admit for used machines I am a bit more lax...I booted into a box i was given once with a floppy and changed the password for root, and kept running the Mandrake 5.1 install it had because he had done a beautiful job tweaking his samba settings. If they offered such services as good service tweaking etc. instead of a default install, I think they would see better results selling linux desktops.
In SOVIET RUSSIA... erm...NSA AMERICA, the Internet logs onto YOU!
(And if someone mods me "Interesting", I'll shoot myself.) ...
Thing is though, how are you going to shoot yourself twice?
Or maybe this is one of those ". . . and I bet you won't read this out. . " radio/newspaper letters.
I think I'll give a try, if someone mods this as "Insightful" I'll eat this small puppy, deliciously garnished with a sprig of parsley and a squeeze of lemon.
I don't get this -- a relatively flamebait-ish writer from a relatively flamebait-ish Linux publication writes an article claiming that Linux desktop share must be high because of a survey focused entirely on developers, and because, well, you can't prove it's NOT! and we're lining up to ask him detailed questions about his analysis?
Look, the Google numbers speak for themselves. If people want to tell themselves that >90% of Linux desktop users are faking their browser ID strings, then bless 'em.
What I'm listening to now on Pandora...
What will it take to get Adobe to release their flagship product Photoshop for Linux? Many users will feel that Linux has arrived when apps such as Photoshop ship for Linux.
Apart from the free (speech/beer whatever) source/ closed source propoganda what has linux to offer a average customer (read non-geek) that windows doesn't ?
:)
Well, I don't really need any more than that. And once educated on the issue, a lot of non-geeks don't either. There are a lot of people who base their purchasing decisions on exactly that kind of thing. You think only geeks are privacy-conscious? Try telling people that WinXP makes you agree to let MS search and alter your hard drive, and see how much they value MS's promise to only use that for voluntary updates.
If its all about choice, then what will prevent customers from choosing windows in future, even if hypothetically speaking the customer is able to buy a no pre-installed OS pc.?
Assume you can get whatever operating system you want pre-installed, otherwise that makes the question kinda pointless (people will pick Windows because they can get it pre-installed). But nothing will -prevent- them from picking windows. Certainly people will. But once there -is- choice, then what's to prevent them from picking Linux?
I know that linux distros include much more than a bare bone OS, but do the customers really care ?
I think so. I think they'll care when they realize that the $100 for the XP upgrade doesn't come with Office. And people who think FreeCell and Minesweeper are pretty cool go apeshit over the games that come with Linux distros.
Basically, they might not care -now-, because they don't know. People aren't used to getting like 3 disks crammed with applications with their operating system, but once they are, doesn't it seem like they'd like that?
One of THE MOST important things for linux to succeed in desktop is to get the OEM manifactures out from the microsoft's clutches.
I agree. The only advantages MS has are by virtue of its entrenchment -- consumer software and hardware vendors target Windows first. Until then, the folks who'd run Linux "except (I need to use Photoshop | UT2003 runs like ass on my Raedon | my scanner won't work, period )" will be beyond reach.
The enemies of Democracy are
Nicholas, given that the vast, vast majority of developers write software for internal corporate use and B2B applications, what possible significance can your 40% number have for the consumer desktop? Windows dominated the consumer desktop market before it penetrated a corporate market dominated by Unix, VMS, OS/360, etc., not the other way around. No one seriously disputes the rapid growth of Linux in the enterprise, but it seems to me that the corporate server market -- even the corporate desktop market -- has very little influence on what my grandmother or my daughter use on their desktops.
Proud member of the Weirdo-American community.
What is the biggest hurdle, in your opinion, for Linux to be on everyone's desktop?
Yes, yes, yes, I am going to be marked as a troll. Sigh.
Ask yourself this question: "Why should Linux be on everyone's desktop?" The usual reasons are (1) that it's open source and (2) that it's more secure than Windows. Those are legitimate points; many other such arguments are more vague and harder to objectively back-up. But even though those points are good, is this a legitimate basis for making the Linux kernel + X11 + KDE + GNU tools a desktop standard? It's still a configuration nightmare, it's still a very complex system, it's still rooted in old tech and very slow to move forward (witness any discussion about getting rid of X11, for example). At the same time, lightweight mobile systems have getting very popular and are making do without most of the baggage that Linux and Windows bring along. At some point, something like "Palm on your desktop" isn't going to be such a bad idea.
I'm not anti-Linux or Windows by any means, but they're both lumbering behemoths from another age. Maybe they're good for some uses, but as a desktop standard? It's time to move on.
Hi--
/. community pokes holes in studies like this when they come out of Windows, shouldn't the /. community hold data to higher standards, even if it does support their cause?
I actually have a bit of a problem with the survey. From where did they recruity the sources? To quote from the article:
Representative Samples
When conducting demand side primary research it becomes important to recruit the participants (or samples) from sources that are as unbiased as possible. During the five years that EDC has been recruiting developers to participate in surveys this ideal has continuously been foremost in our efforts. Consequently, though we have used over 100 different individual sources for recruiting, the following principles have always been and will always be applied:
No vendor lists have ever been used in EDC subscription surveys and none have ever been added to the panel
No platform specific lists have ever been used in any EDC general subscription surveys and none have ever been added to the general panel*
No language specific lists have ever been used in any EDC subscriptions surveys and none have ever been added to the panel
In this way we provide the most eclectic and unbiased sample available anywhere. With thousands of developers chosen in a deliberately unbiased way from a wide variety of neutral lists, our data truly provides in-depth looks at representative samples of the developer population.
*Note: our Linux Development survey does use lists targeted for the Linux platform, however all developers recruited for that survey are kept in a separate database and are not used in any surveys other than Linux specific ones.
Clearly, it says that they use Linux specific developer lists, which indicates that this is not the broader community at all, but a very specific set of Linux developers (of the size and scope of which we have no idea). My question is this: Given how much the
I am located in Orlando Florida and have been looking for work over the last six months or so. I have been checking job ads within the Titusville, Melbourne, Orlando, and Tampa metro areas. This excludes Miami and Jacksonville metros.
I have seen a few ads requesting GNU/Linux experience. Only one or two mentioned FreeBSD, which is a real shame. A lot request Solaris, AIX, and HPUX. By far, the most significant requests are for Microsoft systems administrators though (75% or greater) You have to consider that this is not Silicon Valley over here. There are a lot of government contractors, call centers, real estate business, health care, and banking. Almost all of the companies that requested GNU/Linux experience were small or had been in business for less than 20 years.
I would say that out of the last 100 unique job posts that I have seen, 10 have requested GNU/Linux experience. One even requested that you had to submit your resume in Open Office format, which I think is a great way to weed out some of the fools.
One ad requested Debian. The rest were Red Hat. I do not recall requests for any other distributions.
It is of worthy note that Largo Florida is part of the Tampa metro area. You may remember some stories on Slashdot about how the local government of Largo uses GNU/Linux. And yes, I have seen several requests for GNU/Linux over in the Tampa metro area. A few in Orlando, and few to none in Titusville and Melbourne.
Hi to anyone in Orlando who go to hear John Hall speak about a week ago! I was there, and was the one that stole the last chocolate cookie! Ha ha!
In fact, half of the techs in my office use linux, and the other half use Opera on Win2K... but everyone identifies themselves as IE, just to get around stupid browser sniffers.
Sounds like it's time to make the spoofing relative to the site or the page. Broadcast a reliable presence indicator for Linux for as many sites as possible, but if a site is known to be broken, then spoof. In fact, if your browser followed up every spoof with a second invisible request with a browser named "Linux browser using Microsoft tag because your site is broken!", it might inform the site operators. Hey, I could dream, right?
[