Ubuntu: Just how popular is it?
November 3rd, 2007
Update: We updated this article recently to include the latest statistics. View it here.
There is no doubt that Ubuntu’s popularity has grown dramatically over the past few years, but just how popular is Ubuntu? How many people have ever heard of Ubuntu? How many people visit the Ubuntu site each month? How many people have tried Ubuntu, and more importantly, how many people are actually using it?
According to Canonical’s official press release for Gutsy Gibbon, Ubuntu has a “strong and growing user base of over 6 million people.” Where Canonical got this number is not clear, and they have provided no evidence to back up this claim. Nobody really knows how many people are using Ubuntu, but we found some interesting statistics online that show Ubuntu’s popularity is growing. From these statistics, it looks like Ubuntu has become far more popular than any other Linux distribution.
Note: this article is in no way a scientific study of Ubuntu’s popularity, it is just a collection of interesting stats from around the net. Have fun with it!
So, where can we look online to judge Ubuntu’s popularity?
1. DistroWatch.com
Traditionally, people have turned to DistroWatch.com’s ranking of different versions of Linux to judge a distro’s popularity. This is simply a ranking of the average number of hits per day that each distro’s page gets on the DistroWatch.com site. This ranking system is obviously not a very accurate representation of a Linux distro’s popularity, but it is the generally accepted by the community as an indication of what distros are most popular. Ubuntu has been at the top of this list for some time, and only in the last six months has been surpassed by PCLinuxOS.
DistroWatch.com’s most popular linux distributions for the past 6 months
- PCLinuxOS
- Ubuntu
- openSUSE
- Fedora
- Sabayon
- Mint
- Debian
- Mandriva
- MEPIS
- Damn Small
Does this mean that PCLinuxOS is now more popular that Ubuntu? As the following stats show, this is hardly the case.
2. Website popularity
There are several companies that specialize in ranking websites. None of these sites are perfect and many people discount them all together. We thought it would be interesting to see how some of the top Linux distribution websites stack up in these net rankings. Here are current rankings for the top 10 distros (from the Distrowatch list). When a distro has a corporate sponsor, we’ve included their website ranking in parentheses.
Netcraft Rankings
- www.ubuntu.com: 1,649 (www.canonical.com: 88,013)
- www.debian.org: 1,719
- fedoraproject.org: 4,314 (www.redhat.com: 1,273)
- www.OpenSUSE.org: 4,622 (www.novell.com: 630)
- www.mandriva.com: 7,691
- www.mepis.org: 8,021
- www.damnsmalllinux.org: 8,605
- www.pclinuxos.com 11,144
- www.sabayonlinux.org: 28,549
- www.linuxmint.com: 41,331
Alexa Rankings
- www.ubuntu.com: 2,445 (canonical.com: 119,849)
- www.debian.org: 3,499
- www.OpenSUSE.org: 7,878 (novell.com: 9,154)
- fedoraproject.org: 11,127 (redhat.com: 7,089)
- www.mandriva.com: 18,497
- www.damnsmalllinux.org: 49,544
- www.pclinuxos.com: 57,390
- www.sabayonlinux.org: 72,331
- www.linuxmint.com: 69,753
- www.mepis.org: 82,654
Update: As someone pointed out on Digg, Gentoo’s website should probably be included in this list. We didn’t include Gentoo in our original numbers because we used the Distrowatch.com top 10 as our starting point (you have to start somewhere). To be fair, let it be noted that Gentoo’s Netcraft ranking is 859 and their Alexa ranking is 8,919 which would place gentoo.com in 1st place and 4th place respectively.
Blogging Trends
Sites like Technorati and BlogPulse allow you to track how often people are blogging about a certain topic. We compared Ubuntu’s blog buzz to other top Linux distros and found that people are writing about Ubuntu far more than any other Linux distro.
BlogPulse
We took the top 3 Linux distros from the above website rankings and compared them on BlogPulse. As you can see, people are blogging about Ubuntu far more often. There is also a large surge in blog posts about Ubuntu surrounding the recent October 18th release of Gutsy Gibbon.
Technorati
Technorati shows very similar results when comparing Ubuntu, Debian and Fedora over the last 30 days (note the scale difference in these graphs).
| Ubuntu | Debian | Fedora |
 |
 |
 |
Google Trends
Perhaps the most interesting statistics come from Google Trends. This tool allows you to compare different terms and see how often people search for them. The tool is far from perfect and is still in Google’s “labs”, but it does give some interesting insight into how often people search for different Linux distros. Of course, there is a lot of room for error as someone could be searching for the philosophy of Ubuntu, for a new Fedora hat, or for The Red Hat Society.
First we compare our top 3 distros from above. You can clearly see that from the second half of 2006, there are far more searches for Ubuntu than for Debian and Fedora (including Fedora Core, and Red Hat searches).
Next we compare Ubuntu to the rest of the top 10 distros combined (including variations of the distro names such as Red Hat). As you can see, for the last few months, Ubuntu has been searched for more often than all the other top 10 combined.
We found another interesting trend when comparing Ubuntu to Linux, Unix, FreeBSD and Solaris. Clearly Linux has the lead, but Ubuntu is not far behind.
Finally, just to be fair, we compare Ubuntu, Linux, Mac, and Windows.
DIGG
Then there’s always Digg.com. How many times have these popular Linux distros made it to the front page of Digg in the last 6 months (stories containing the distro’s name in the title)? We did some searching and found the following information very interesting.
- Ubuntu: 163
- Fedora: 10
- Mandriva: 8
- SUSE: 8
- Debian: 6
- PCLinuxOS: 3
- MEPIS: 1
- Sabayon: 1
- Mint: 0
- Damn Small: 0
Conclusion
So what do all these statistics really mean? Honestly, probably nothing. They can’t tell us anything about real Ubuntu usage. However, they do seem to show that Ubuntu has managed to gain a large portion of the Linux mind share, at least amongst the tech community. We enjoyed uncovering and compiling these stats, but please remember that it’s just a bunch of unscientific data, it’s just for fun. We hope you enjoyed this article, now back to your favorite OS (whatever that might be).
November 3rd, 2007 at 6:07 pm
The distrowatch ranking of distros means absolutely nothing outside distrowatch.. and the distrowatch folks have stated that time and time again. I know of lots of people who go to the Distrwatch page on a daily basis and click on a specific distro’s link (PCLinuxOS) just to raise the hit count for that distro. That rating system is flawed and I feel that it should not even be mentioned. I wish they would take that silly rating off their site and leave it off.. it’s misleading to a lot of people.
November 3rd, 2007 at 6:19 pm
1) Blogs:
“Sites like Technorati and BlogPulse allow you to track how often people are blogging about a certain topic.”
Are you counting the negative blog entires as well as the positive blog entries? I have seen blogs that state Ubuntu is the worst distro some people have used.
2) Number of installs:
I install and use many distros when a new version is released. But, most of them are uninstalled within days because due to bugs and the like. So gauging popularity by number of installs is not accurate either.
3) Searching Google (and other search engines) does not equate to distro popularity:
Many times I search for a distro, read about it, and decide it’s not for me.. so that distro never gets installed on my machine – but I did search for it.
4) Digg.com:
I have used more than 28 distros at one time or another, but I don’t have a Digg.com account, so gauging popularity that way is also inaccurate.
Conclusion:
There’s really no accurate way to gauge a distro’s popularity because one man’s trash is another man’s treasure. Use the distro that works for you and stop worrying about the numbers.
November 3rd, 2007 at 6:40 pm
In the Alexa rankings the third and fourth place seem to be swapped.
And imo there is a huge difference between “having heard about” or even “tried” and actual usage. Still haven’t seem any resilient number about actual usage/installations.
The “6 million users” number of Canonical seems to be made up: http://spevack.livejournal.com/31016.html
November 3rd, 2007 at 6:54 pm
Great posting. I assume you’re probably trying to deflect flaming comments by saying all these numbers are just for fun, but I think together they give the clearest and most balanced view possible of Ubuntu adoption. Anyone familiar with webstats will know that you can’t really measure what you want to know directly (ie. do people like my site?), so those folks (like myself) are accustomed to inferring from a number of related indicators (like blog activity, search string activity…).
Personally, I think the Google Trends data is the strongest. Why do people enter search stings with Ubuntu in them? I think it’s because they’re interested in Ubuntu and at the core of all that search activity is people actually using the OS. It’s not a 1:1 relationship, obviously, but it’s best used to compare OSs to get a indication of the relative popularity of an OS. Like you said, we can’t figure out the actual number of users, but we can determine the relative number of users, and in the end that’s what everyone is really interested in: rankings.
Just to add to the confusion of stats. Mozilla recently stated that half of those who download Firefox don’t end up actually using it. I wonder how many people download Ubuntu to test it out and don’t actually end up using it either. Or triple or quadruple boot?
November 3rd, 2007 at 7:20 pm
With all I have read about linux and Ubuntu, I would like to try Ubuntu,
however I can’t get past a certain requirement by windows xp, so I can’t do it
on my own and I need help but no one, so far will help me. i have asked for help but no one I have asked ever responds to my request. Oh well, xp works ok.
Can this be done remotely? I’m surprised that no one will help me. too bad so sad.
Thanks,—Doug Rosbury
November 3rd, 2007 at 7:24 pm
why not add things like size of forums/mailing lists.
also there are the popcon rankings for debian and ubuntu, not sure if other distros have similar systems. http://popcon.ubuntu.com/ http://popcon.debian.org/
November 3rd, 2007 at 8:04 pm
The size of forums was published in a recent issue of Linux Format Magazine.
Here was the results:
———————
Ubuntu: 376,200
Gentoo: 109,000
Fedora: 100,000
Mandriva: 58,800
OpenSuse: 52,000
Kubuntu: 21,000
Freespire: 18,600
PCLinuxOS: 11,700
Arch Linux: 10,600
Sabayon: 8,500
Puppy: 7,000
Mint and Zenwalk: 3,400
November 3rd, 2007 at 10:49 pm
I have been using Ubuntu for some time. Very interesting stats – I am happy to see it making headway.
More impressive than Ubuntu’s gains is your post. I am really impressed with your use of stats tools and how you put the post together. All in all – well said.
November 3rd, 2007 at 10:51 pm
Ok Doug, I’ll bite. What’s the issue that’s holding you to Windows XP that is preventing you trying Ubuntu?
private message me on the Ubuntu forums with your repsonse.
November 3rd, 2007 at 11:21 pm
Doug, same here. You can contact me via the Ubuntuforums private messaging system, as Bil-E-daKid says, username p a p a t r p t 8 9 (without the spaces)
November 4th, 2007 at 12:41 am
No mention of Dell and their Ubuntu option?
November 4th, 2007 at 2:43 am
I would think that the count would be more accurate if you counted UbuntuForums.org hits instead of Ubuntu.com hits. There is more registered users at the forums than the main site. I know I visit the forums more than the main site. Also does this include the *buntu derivatives(Kubuntu,Xbuntu,etc)?
November 4th, 2007 at 2:46 am
I did a similar experiment looking at some statistics of websites and a/emule
http://grigio.org/quanti_sono_utenti_linux_statistiche_ottobre
Normal Linux users are lesser than 1%, but among “techie” users, Linux beat Mac.
November 4th, 2007 at 2:51 am
Ubuntu provides automatic software updates, which should theoretically provide a good estimate of how many people are using their OS regularly. I don’t know if that’s how they got their numbers, but it makes sense to me. There’s also a package you can install to allow them to use your package list to help determine package popularity, if you want them to do that.
November 4th, 2007 at 2:54 am
Well it seems to me the best way to calculate usage is with the patch/update downloads. Again not totally accurate as I guess some people won’t keep there machines patched, but I tend to think it would be about the most accurate way we have. just my 2 cents worth.
November 4th, 2007 at 4:26 am
Comparing google trends with the word ‘windows’ is misleading. Most windows are made of glass.
November 4th, 2007 at 5:42 am
I agree with 14 & 15. The easiest way to discern a distro’s popularity is updates. Of course this depends on two factors, A) the distro in question has an update system, B) the person using said distro has an internet connection. I think it’s fairly accurate since most distros that do have an automatic update system have it set to connect with the server to look for updates automatically, regardless if the person chooses to update or not. This would show “at least” that x-amount of people used said distro at least once.
Keep your stick on the ice…
Landor
November 4th, 2007 at 7:23 pm
Nice statistics. It is indeed hard to draw conclusions, but alltogether these numbers say something about the popularity of the different distro’s! (though this does not imply correlations with quality, look for example at the popularity of the distro called windows…..)
November 5th, 2007 at 8:56 am
Parachute pants were popular, but still quite lame.
I’ve tried several versions of Bloatbuntu, and I have yet to be impressed.
November 6th, 2007 at 5:37 pm
STOP! Hammer time!
November 6th, 2007 at 10:18 pm
The most helpful marker would be the number of discreet computers accessing the ubuntu repositories over a short span of time (2-4 weeks), which would catch people who only update occasionally, without double-counting folks who have reinstalled (maybe changing the identification of their computer).
I don’t know if any distros really make this possible, since it is a potential privacy breach.
July 29th, 2008 at 8:47 pm
Just found this on digg, Ubuntu is the most well known distro after maybe redhat