I don't believe that 25% of Betanews readers are Linux PC users

Twenty five percent. Well, 25.82 percent to be exact, so closer to 26. That's the number of respondents claiming to be Linux PC users in my poll: "How would you identify yourself as a computer user?" The number is seemingly credible: 2,014 votes from 94 countries -- 988 from the United States. There are 1,858 unique IP addresses, which along with cookies blocking repeat votes suggests the results are largely clean of Linux enthusiasts stuffing the ballot box, so to speak. Something else: After the poll reached the 25 percent Linux PC threshold, it stayed there even as votes accumulated -- that's another indicator the results are good. But I don't believe them, certainly not that only 60 percent of respondents are Windows PC users.

I do believe that Betanews has among its readers very enthusiastic and credible Linux PC users; some of them will share their stories in this post. But the Windows community is much larger, something reflected by other metrics, such as comments or download forums. Additionally, the poll results are unqualified, meaning I don't know who the respondents are. Therefore, I must consider the data to be unreliable.

It would be oh-so easy to make a sensational story out of the data, emphasizing the lower number of people identifying themselves as Windows PC users or the surprisingly high number of Linux PC users. Instead, ahead of this post, I wrote on my blog: You Can't Trust Polls or Surveys." The misuse of poll or survey data is rampant, and major blogs or news sites -- and OMG Facebook -- post new polls and results every day. You can justify just about anything with a simple poll and loose interpretation of the data. It's a rampant -- and from personal perspective as a journalist, disgusting -- problem.

That's not to say all this asking people questions is useless. I asked my poll question in a way to deliberately gauge sentiment -- what kind of PC user people feel they are rather than which operating systems (likely more than one) they use on a daily basis. I do believe that the poll in conjunction with reader response in comments and e-mail reflects something important about Linux. There is likely a larger minority of Linux PC users than broader business or consumer research indicates. Based on the consistency of one data point throughout the entire poll, I would comfortably assert that more Betanews readers identify themselves as Linux PC users than Macheads (12.02 percent, by the poll).

Ian Brunton: Linux Newbie and Happy About It

Who are these Linux PC users and why? That's a question I posed in post "Do more Betanews readers use Linux PCs than Macs?"

Ian Brunton isn't "professional in IT in any sense," but a "hobbyist programmer. I've used Windows most of my life, from 3.1
through 7, skipping lightly over Vista." Ten years ago, as part of a summer job, he learned Perl and dabbled in Linux. "At that point, desktop Linux was rather clunky." Last year, Brunton gave "Linux another shot, with, Ubuntu 9.04. This went quite well until the automatic upgrade to 9.10 gave me some very disturbing error messages and did not work well. I migrated back to XP, and shortly thereafter bought a new desktop with Windows 7." Linux still wasn't good enough. Or was it? After additional experimentation, Brunton "switched to OpenSuse 11.3 in August 2010, and have stuck with it."

Brunton is exactly the kind of customer Microsoft shouldn't want to lose -- he switched not from XP or Vista but Windows 7. His problem is longstanding: "I always disliked the default interface," something Windows 7 didn't change. Not surprisingly, among his reasons for choosing Linux: "The ability to customise it is probably the biggest. I can have all manner of different themes without the least bit of trouble. On Windows, getting a dark theme required hacking the system, because for some reason Microsoft won't let you install one, and dark themes are what I find most comfortable. OpenSuse 11.3 came with one ready to go."

The "other big reason" surprised me: "Command-line interface." Windows most certainly has one of those, as does Mac OS X. "I've found Microsoft's version clunky, stunted, and counter-intuitive," he explains. Another reason: "Programming is simple. Since my primary language is Perl, a good command line is important to me. Perl on Windows never seemed to work quite right. I also like the bash shell as a scripting language." Brunton gives other reasons, and of one of them I'm not surprised at all: "I appreciate the repository approach to software, which automatically takes care of dependencies and conflicts with minimal effort on my part." Even Windows enthusiasts complain about the registry and multiple DLLs.

I asked Brunton about Linux drivers, as this is a common complaint about Linux (or at least something Microsoft evangelists like to call out): "I have not had any problems with drivers. Both Ubuntu and OpenSuse worked  just fine with my hardware 'out of the box.' This was a pleasant surprise. I installed Ubuntu on a small Compaq, and OpenSuse on a powerful new Dell."

Does he have any other reasons for choosing Linux. A big one. "Finally, I love the philosophy behind Linux. Note that I have not actually saved any money by using Linux, since while it's free I bought the computer and its licensed copy of Windows first," he further explains. "I decided I wanted to support the philosophy of GNU/Linux and open-source software." Brunton emphasizes: The bottom line is that I find GNU/Linux more compatible with my (non-professional) needs and desires than Windows, and a lot of fun. With GNU/Linux I can have exactly what I want, not just something I can live with."

Jean Hominal: Geek, Programmer and Linux User

"I identify myself as a geek and a programmer. I am also a Linux user," Jean Hominal explains. Hominal gives four reasons why Linux, which I fully quote:

1. There is a range of excellent software that I use every day (GIMP, Git, Apache, PostgreSQL, Tomboy) that works better with Linux than with Windows. Of course, there is selection bias here -- I have chosen that software because it was excellent and ran very well on my computer.

2. The variability of Linux distributions, and the open-source nature of most of Linux software, ensures that software packaged with a decent Linux distribution will be of better quality than nearly all third-party software for Windows. Variability is important because it forces developers to establish and document the assumptions (required APIs, libraries, administrator rights) that their program will need to function. Open-source is important because it ensures that problems can be easily fixed once discovered, and dodgy software can be exposed.

3. My package manager makes keeping software up-to-date a lot easier than Windows' updater.

4. The laptop I am currently using was taken from my brother, who had installed Windows 7 on it (it was pre-installed with Vista). As he has since used that copy of W7 on another machine, and as I have been unable to reinstall Vista using the recovery partition of the laptop, my only choice to get Vista legally would have been to send a letter to HP and pay them so that they can send me a DVD of Vista. No.

Hominal identifies two drawbacks: Can't play Blu-ray discs, and limited game support; Brunton also identified the latter as a problem.

Martin Lindhe and Other Betanews Readers: Why Linux?

Martin Lindhe is a long-time betanews reader  -- "7-8 years now?" -- he doesn't remember exactly. About 4 years ago he "switched from Windows to Linux and now I run only Linux on everything: My media center PC, my laptop, my home stationary PC and my work PC." Like Brunton and Hominal, Lindhe is a programmer. He first tried Linux, Mandrake, around 1997, but it wasn't good enough. Since making the switch, "I have never looked back."

Like Brunton and Hominal, Lindhe finds Linux's customizability to be appealing. "Linux gives me so much more features and abilities than I ever had with a Windows system, and since I like tinkering with my system it allows for great stuff." Lindhe switches to the evangelist role: "These days Linux is working quite well even for end users; and I have a few non-tech friends who made the switch to linux also. If most of what you do with a laptop takes place in a web browser session, the underlying OS matters less."

I asked Lindhe if he has ever used Mac OS X. "No, I never really considered using Mac OS due to the pricing. Last time I really used it was before version X, and that was quite crappy experience. I do use my second iPhone all the time though, so I do like their stuff. However, I sync it in Linux with Rhythmbox, which gives me many features not even available in iTunes, such as playing songs directly from the device, mounting the full file system to change sound files etc." I also asked Brunton and Hominal about Mac OS X. Like Lindhe, they haven't used modern Macintosh.

As I write there are 114 comments to the aforementioned post asking the Linux PC versus Mac question. The responses above are from e-mails. I've left out most of the best Linux comment responses, because the commenters aren't identified. Among the few readers who are identified, Brian Masinick proclaims: "I prefer GNU/Linux based software." Linux appeals to him in part because of a long personal history using Unix.

Bobby Russ asks and answers: "Do you identify yourself as a Linux user? Yes." He pounds out more questions with answers:

Which Linux version do you run? Linux Mint 9 and Ubuntu 10.4 LTS. What are the primary benefits you see from Linux over Windows? Quick start up, No start-up cost to install, Very little chance of dealing with Trojans, adware, spyware and other malware. Used to run all Windows-based machines, but it seems they are more and more bloatware. The cost of the OS is no longer worth the Windows environment. Linux is fast and free. Security updates come as quickly if not quicker.

In closing, there is a reasonable question to ask: Could I be wrong about that 25 percent? I think not, based on aforementioned reasons and data from other sources. For example, according to NetApplications, which measures operating system usage share by Website access, year to date: Windows, 91.46 percent, Mac OS 5.13 percent and Linux 0.98 percent. If you have evidence that the percentage of Linux users is much higher -- at least among the readers here -- please present it in comments or e-mail joewilcox at gmail dot com.

[Editor's Note: Spelling of Ian Brunton's last name corrected.]