Explosion of Interest in GNU/Linux at Distrowatch.com

Distrowatch is a popular site. Anyone who wants to compare distros or quickly check what various distros are doing should waste a bit of time there.

My findings today?

Amazing. The last week saw the number of distros with 1000 or more hits per day increase from 9 to 15 compared to the last month. What happened is that the total hits/day for the top 100 distros increased 9%, but the hits/day for the top 15 increased 20%. This means the world is becoming more focussed on the popular distros. A consolidation of distros is happening.

This is as it should be. The world can make its own software but the world does not need hundreds of distros. Eventually we should see a decrease in the number of distros with only the most popular and the most valuable special-purpose distros surviving.

Also unavoidable is the conclusion that interest in GNU/Linux is still growing rapidly. All this FUD about the death of GNU/Linux on the desktop is rather silly. What business would not be celebrating 10-20% growth in activity per month?

About Robert Pogson

I am a retired teacher in Canada. I taught in the subject areas where I have worked for almost forty years: maths, physics, chemistry and computers. I love hunting, fishing, picking berries and mushrooms, too.
This entry was posted in technology and tagged , , . Bookmark the permalink.

14 Responses to Explosion of Interest in GNU/Linux at Distrowatch.com

  1. “a lot more data from a lot more sources” does nothing to answer the fundamental question of usage. Only a scientific survey will come close to satisfying that and web stats are nothing scientific. There’s no control of the web sites, the languages or subject matter and no way to overcome NAT without peeking at MAC addresses. Even that is unlikely to be reliable enough because of proxies/firewalls etc. I do plan to set up a “honeypot” of IT-neutral stuff on the web to collect some webstats with a known unbiased website. I will publish the results in a year or so after the data has been collected and analyzed.

  2. twitter says:

    Looking at the trends for the year, I see an increased and broadened interest for the year. The top ranking distro, mint, lost about 20% but all the other top 20 gained 10 to 20% when comparing yearly vs monthly averages. For the 800 hits lost by mint, there are about 2,000 gained for the next 19 most popular distros. Outside of the top 20, the changes per distro are noisier, but the sums should be reliable for the top 100 and these show a rise of about 3% from 40,200 to 41,400 between 1 year and one month averages.

    I happen to think a large number of distributions is healthy, as long as they each have something special to offer users.

  3. Yonah says:

    Hmm…

    “light-hearted way of measuring the popularity of Linux distributions”

    “should not be used to measure the market share of distributions”

    “…nothing more.”

    Seems you’re doing something you’ve been told you shouldn’t do. You need a lot more data from a lot more sources before you can make a “definite” declaration.

  4. kozmcrae says:

    Crow wrote:

    “On the other hand, could the total of hits show people general interest in Linux?”

    When people first get interested in Linux they often visit DistroWatch and similar sites. That is what I did before I downloaded and installed my first copy of Linux. An increase in traffic on DistroWatch definitely means an increase in interest in Linux. It most likely also means an increase in Linux users.

  5. The explanatory notes from Distrowatch:
    “The DistroWatch Page Hit Ranking statistics have attracted plenty of attention and feedback over the years. Originally, each distribution-specific page was pure HTML with a third-party counter at the bottom to monitor interest of visitors. In May 2004 the site switched from publicly viewable third-party counters to internal counters. This was prompted by a continuous abuse of the counters by a handful of undisciplined individuals who had confused DistroWatch with a poll station. The counters are no longer displayed on the individual distributions pages, but all visits are logged. Only one hit per IP address per day is counted.

    The DistroWatch Page Hit Ranking statistics are a light-hearted way of measuring the popularity of Linux distributions and other free operating systems among the visitors of this website. They correlate neither to usage nor to quality and should not be used to measure the market share of distributions. They simply show the number of times a distribution page on DistroWatch.com was accessed each day, nothing more.”

  6. Crow says:

    Mr. Pogdson, I use to read your publications with interest and maybe you can tell me what do you think about this.
    To this moment I don’t understand what the numbers at Distrowatch means: is not the number of users, is not the preference for a distro, is not the use of a distro. It seems to me that it means curiosity about a distro which is easy to promote using some hype and a frequent release cycle.
    On the other hand, could the total of hits show people general interest in Linux? have you seen changes over the months/years?
    Thank you for your time
    I think the data could have a better use than the confrontation between distros.

  7. That Exploit Guy says:

    @Robert Pogson

    ‘Of course there is. A lot of distros are getting a lot more hits per day this week than last month. How else can you describe the data? The whole point is that the average has changed dramatically.’

    By the same logic, I could also say that Americans generally care more/less about the weather than ever before (depending on which slice of the data you look at), except that would obviously be an absurd conclusion to draw.

    TEG’s observations are only correct for statistical variation. This is a real shift, not a variation. SQRT(1000) is about 32. That’s 3%. We see a 20% shift.

    Ugh… Try again.

  8. NetApplications is seeing something similar:

    week of September 30, 2012, USA GNU/Linux =1.91%

    month of September 2012, USA GNU/Linux= 1.59%

    What big roll-out of GNU/Linux happened last month?

    NetApps:
    California, September, 8.81%
    week of September 30, 10.46%

    Yipee! It’s Mountain View, CA which went from 10% for September to 52% for the week of September 30. Did Google expand? Did a neighbour take up GNU/Linux?

  9. TEG wrote, “Without knowing at least how those 30 data points are distributed, there is simply no way to draw any of that conclusion you have put in your post.”

    Of course there is. A lot of distros are getting a lot more hits per day this week than last month. How else can you describe the data? The whole point is that the average has changed dramatically.

    TEG’s observations are only correct for statistical variation. This is a real shift, not a variation. SQRT(1000) is about 32. That’s 3%. We see a 20% shift, like Debian GNU/Linux going from 1300 to 1600. So, TEG is comparing apples and oranges or raising straw men, again…

  10. MK says:

    October/November and April/May are release seasons, and so, naturally, interest is up and clicks are up, in other words, it’s business as usual. Does it tell us anything at all about the rapid growth of interest? I think not.

  11. eug says:

    Opinions (by Jesse Smith)

    What Linux needs is an operating system

    Anyone who has spent any significant amount of time in the Linux community has heard the questions raised: “Why are there so many distributions? Why can’t the various projects get together and make one unified distribution?” Of course this is never going to happen, nor should it. One of the greatest benefits of open source software is that it gives people the ability to use their computers the way they wish to and modify their systems to work they way they want. This has given the Linux community a great degree of flexibility, letting it run anywhere, from extremely low-resource systems to super computers, from stable servers to cutting-edge desktops to niche administration tools like Clonezilla and KNOPPIX. The ideal of one unified distribution sounds nice to newcomers overwhelmed by choice and to third-party developers, but it won’t fly in a community where freedom, empowerment and flexibility are priorities. One might as well ask why the human race has so many languages when having one would seem so much more simple.
    (…)

    http://distrowatch.com/weekly.php?issue=20121008

  12. That Exploit Guy says:

    ‘The conclusions I made in TFA are the same and I was well aware of the terms. That’s why I chose to compare both 30 days and 7 days period.’

    That still doesn’t make any sense. You are comparing the average of 30 data points with the average of 7 data points from those 30 points. Without knowing at least how those 30 data points are distributed, there is simply no way to draw any of that conclusion you have put in your post.

    To illustrate, let:


    a = the average page hit of the first week of the month
    = 1000
    b = the average page hit of the second week of the month
    = 1010
    c = the average page hit of the third week of the month
    = 950
    d = the average page hit of the fourth week of the month
    = 1000

    The average page hit of all four weeks = (a + b + c + d) / 4
    = 990

    See the problem here? Despite a = d, the average page hit of all four weeks can still be smaller than d. The conclusion you draw from those data simply makes no sense.

  13. TEG wrote, “The page counts in the chart are “average number of hits per day over the stated period of time, not numbers of hits of the day.

    Welcome to the new low of mrpogson.com, my fellow readers!”

    Welcome to the new low in comments from TEG, my readers. The conclusions I made in TFA are the same and I was well aware of the terms. That’s why I chose to compare both 30 days and 7 days period. The 7 day period averages both the weekend and the weekday to smooth out spikes on the weekend. It was still a remarkable increase in the past week. Compare with my previous report, where only ten distros made it above 1K hits per day.

  14. That Exploit Guy says:

    It seems that you don’t just post useless statistics – you also misread your useless statistics as well.

    The page counts in the chart are “average number of hits per day over the stated period of time, not numbers of hits of the day.

    Welcome to the new low of mrpogson.com, my fellow readers!

Leave a Reply