Why People Like Linux

GNU/Linux has a rapidly growing user base according to our polls. Today, we posted articles from Belfast, India, Wales, Croatia, Glasgow, South Africa, Sri Lanka, the UK and Massachusetts. On other days, we see articles from many other countries. We have readers from 165 countries and more often than not our US readership runs only about 50%. Is price the only reason for the interest? Unlike the early days when Linux provided a learning platform for people who couldn't afford UNIX, price seems the least important reason. What about the Microsoft alternative? Again, we saw that as barely a factor. People just like Linux.

People say that change is the only constant in the universe. A young man with a brain tumor once told his children that everything in the universe is on its way to somewhere else. I used these statements as analogies to the phenomenon called GNU/Linux or Linux depending on your leanings.

The things about Linux that one could characterise a year ago have changed. I remember how wonderful I regarded the 2.035 kernel, Samba, Apache and "C" in 1998. We didn't have KDE and Gnome but that didn't matter.

So, here's the most current findings I have in order of importance. They differ greatly from the reasons Ransom Love gave at a Linux Business Conference in Austin six years ago when the major reasons people used Linux included: File and Print serving, an Internet Platform, Business Applications and Low-cost alternative for LANS.

• An enjoyable user experience
• Challenging and fun to learn
• A powerful operating system
• Development platform and available tools
• Exciting to watch
• Versatility and variety of available options
• Community support and applications
• Web services
• Microsoft alternative
• Stability and Security

The top ten reasons have changed dramatically since kernel 2.035 when Linux users numbered approximately 2 million. We didn't have a top ten, we only had a top four.

I found our surveys interesting because few people listed the enterprise as a main reason to use Linux. Linux as a server barely rated in the list of reasons. I attribute that to our readership demographics and in those that chose to respond on other sites such as digg.com, O'Reilly media, Slashdot and other Linux news sites that carried our survey.

I did a follow up with people I know in large organizations. Several Linux advocates felt uneasy responding due to policies inside their companies. One Fortune 15 company, for example, has a pilot going which the company designated as classified. Even with the high adoption rates within various organizations, only key personnel could have answered the survey questions. That seemed to skew the results.

Some interesting observations

The Apple Macintosh users also had little knowledge of Linux but indicated extreme prejudice against it. I also observed that Mac users fell into more vertical categories such as graphics, music and film editing and journalism. It appears Apple has lost its traditional education market.

Windows users voiced dissatisfaction with Windows and said they primarily "put up" with Windows to use Microsoft Office productivity tools. Primary use of Microsoft's productivity suite followed a pattern I anticipated. Of those surveyed approximately 80% used Microsoft Word, 10% used Excel and 3% used Access and the remainder used a variety of other tools such as Project.

Windows users overwhelmingly used Microsoft Outlook, Outlook Express and Internet Explorer while Media Player has wide spread use among home users and modest use in organizations. Windows users were subject to panic disorder, anxiety attacks and depression consistent with those in the general population.

How accurate is the survey?

Of the people surveyed, the results are surely accurate. But, I would not say it represents an accurate representations of all IT users. For example, we did not compare satisfaction among users of Red Hat on IBM mainframes compared to modest cost hardware with Windows 2003 Server. We also didn't survey the Red Hat women's groups around the country that have absolutely nothing to do with computers.

Also, I would not count on the survey as a projection of the growth of Linux. We had a tiny percent of users from Brazil, China, Korea, Japan and the Middle East. Those appear to have higher growth rates of Linux than the US or Europe.

I would also attribute a high error rate due to hanging chads. Many of the participants had more than one reason for using Linux. So, we could not tell for whom they voted. We double counted those.

I enjoyed reading the posts and disqualifying non-Linux trolls. Thanks to Herschel Cohen for his help in tabulating the results.