The Pen is mightier than the FUD
|
| Author | Content |
|---|---|
| pogson Jul 19, 2007 7:50 AM EDT |
"Besides Gartner and IDC telling us that FOSS is booming instead of declining, Rob gets many other facts wrong." Is this true? It was true around 2004, but IDC/Gartner have quit releasing such news lately. Web stats show a flat line for Linux, too. I am not sure what is wrong with the stats but they seem contrary to what I have personally observed, constantly increasing activity in GNU/Linux. When I ask Google for hits on -IDC linux- I get stuff from 2004 that is encouraging. Have they made a deal with M$? Gartner gives dismal predictions from 2005. w3schools stats are around 3% and I believe Linux on the desktop passed 3% years ago. see http://www.w3schools.com/browsers/browsers_os.asp I would expect Linux to be more than 5% of the desktop seats by now. This is frustrating when trying to reply to MS-fanbois/trolls on several sites and it gives Enderle room to wiggle. Linux is winning on all fronts all battles except the FUD which seems to spring up from every hole in the Earth. |
| Sander_Marechal Jul 19, 2007 12:38 PM EDT |
Gartner 2007: http://www.gartner.com/DisplayDocument?id=509681
Quoting:Hype Cycle for Linux, 2007. The Linux operating system is moving into what we classify as its third generation (earlier versions were basic infrastructure and midtier Web/applications) to play a major server role in data centers. Its progress in the PC sector will continue. IDC December 2006 http://searchenterpriselinux.techtarget.com/originalContent/... Quoting:As part of a list of predictions for 2007, research firm IDC said Microsoft's client operating system anti-piracy efforts will backfire in 2007. Instead of stamping out software piracy, the campaign will drive customers toward a Linux desktop, said the Framingham, Mass.-based research firm. So far, it's come true :-) IDC March 2007 The Linux desktop market is growing Quoting:Linux still goes into only a tiny proportion of the desktop and laptop PCs sold. But in a recent report, market researcher IDC said licenses of both free and purchased versions of Linux software going into PCs world-wide rose 20.8% in 2006 over the previous year and forecast that licenses will increase 30% this year over last. That compares with 10.5% growth in 2004, according to IDC. |
| pogson Jul 19, 2007 3:40 PM EDT |
We have had "growth" like that for years but the market share is still in the range 1-3% according to web stats. What's with that? With this kind of growth we should have Linux on 6% or more of desktops. Maybe that is a comment on web stats. I guess I will have to teach for a few more years to boost the numbers... |
| tuxchick Jul 19, 2007 3:59 PM EDT |
The numbers are bogus. None of the Famous Analysts even acknowledge the large numbers of installations from downloads or re-purposed windozes and Mac PCs, nor counting the numbers of purchased commercial Linuxes that are installed by the users, and I doubt they're trying very hard to count the numbers of Linux preinstalls either. |
| tracyanne Jul 19, 2007 5:36 PM EDT |
Quoting:The numbers are bogus I've believed that to be the case for some time now. A quick and dirty calculation, based on the number of installers I've done that I doubt get counted by anyone, tells me that the stats are likely 5 to 10 times higher than the analysts would state. |
| Sander_Marechal Jul 19, 2007 9:07 PM EDT |
Quoting:w3schools stats are around 3% and I believe Linux on the desktop passed 3% years ago. w3schools stats are flawed for this comparison. Web design and web development is dominated by Windows because sites must always be checked out to work with IE (IEs4Linux works, sorta, but IE7 is a problem). Also, photoshop and dreamweaver are a web designers most used tools. Neither has a Linux version and here Wine cannot help. So there's a disproportionate amount of Windows users on w3schools. It's very hard to find a good site to look at it's webstats. I think that Google is one of the few that would not have disproportionate stats, but the don't show their stats unfortunately. |
| pogson Jul 20, 2007 8:32 PM EDT |
Sander_Marechal wrote:"Web design and web development is dominated by Windows because sites must always be checked out to work with IE (IEs4Linux works, sorta, but IE7 is a problem)." Thanks, I did not realize that. My world is Linuxian where folks use that site as a standard. I thought everyone did. I wonder if Linus would go for a once-a-year pinger in the kernel to phone home and be counted? |
| jdixon Jul 21, 2007 4:48 AM EDT |
> I wonder if Linus would go for a once-a-year pinger in the kernel to phone home and be counted? Almost certainly not. And even if he did, that won't help. Most home users (whether dialup, cable, or DSL) have dynamic IP addresses. |
| Abe Jul 21, 2007 4:05 PM EDT |
Quoting:I wonder if Linus would go for a once-a-year pinger in the kernel to phone home and be counted? It doesn't have to be in the kernel and it could be a user initiated open source tool. Quoting:that won't help. Most home users (whether dialup, cable, or DSL) have dynamic IP addresses. True, but the MAC address would be pretty accurate although not perfect. |
| tracyanne Jul 21, 2007 4:06 PM EDT |
Probably is most of us actually bothered to register our install with the distributor, there would be a better set of stats. |
| azerthoth Jul 21, 2007 4:37 PM EDT |
While not a big fan of ANY call home for any reason the use of dynamic IP would have zero effect on the ability to do so. The MAC address doesnt translate past the gateway/router/modem so is absolutely useless unless it is polled and encapsulated in the call home message. Not to mention several distro's already default ship with a call home function, Debian and Ubuntu I know for certain. It can easily be disabled and some do so without even realizing that they have. The biggest drawbacks to the idea would be: A: Convincing every single distribution out there that it was in their users best interest to violate their privacy even this much then convincing the users themselves to upgrade so that the patch was there. B: Kernel level, convincing every single user and business to upgrade their kernel or to push the individual distributions to release a new kernel. Using Debian as an example that could take as much as 3 years. |
| Sander_Marechal Jul 22, 2007 5:45 AM EDT |
Quoting:Not to mention several distro's already default ship with a call home function, Debian and Ubuntu I know for certain. You're talking about popcorn. It tracks what packages you have installed and reports that periodically to a central server. Debian an Ubuntu ask at install time if you want to install popcorn or not. IIRC the default is "no". |
| azerthoth Jul 22, 2007 6:58 AM EDT |
Actually I have never been asked, even in the cases with Debian where I have done a debootstrap install. |
| Abe Jul 22, 2007 9:34 AM EDT |
Quoting:The MAC address doesnt translate past the gateway/router/modem so is absolutely useless unless it is polled and encapsulated in the call home message. My fault, I didn't make it clear enough and you didn't grasp my note "user initiated open source tool". I wasn't talking about sniffing or poking. I was talking about a tool, with user consent, that collects and send, what ever necessary information, to a central repository for statistical purposes only. Totally anonymous. A standard tool that can be and could be adopted by all distros. The idea of getting fairly accurate stats about FOSS usage need to and will happen sooner or later. Repositories do have good information about what is being downloaded, whether they have any other information that is being downloaded without users knowing about it! I can't say. In any case, does it matter to users if hardware information was downloaded along without any, or association with user personal information? I don't think so. I believe this scenario is appropriate and doesn't not infringe on any privacy. |
Posting in this forum is limited to members of the group: [ForumMods, SITEADMINS, MEMBERS.]
Becoming a member of LXer is easy and free. Join Us!