New KDE Plasma NM Sees Its First Release

[Luke_Wolf]

Ubuntu forums have 1,8 million users, Mint has less than 100 000, FedoraForum.org has 140 000 and openSUSE forums have less than 80 000k.

Circumstantial; Shows usage of Forums not usage of Distro. Large possibility that usage is boosted by Users from other distros including derivatives such as mint.

Ubuntu Google+ community has 340k likes while Gnome has 23k , Fedora 24k, openSUSE 34k, Linux Mint 14k and so on.

Invalid, likes show popularity but not usage. Furthermore with Google+ the Ubuntu google + community is shown prominently on the suggested communities pages box if you show interest in Linux, other distros not so much. Possibility they paid to win.

Ubuntu reddit page has 38k readers, Mint has 4k, Gnome 2,4k, Fedora 3k, openSUSE under 1000.

Circumstantial, Reddit is 4chan's little brother and thus this statistic shows the level of Hipster interest in Ubuntu, but again this doesn't translate directly into users.

Ubuntu Facebook page has 850k likes, Linux Mint has 40k, Fedora 38k, openSUSE 20k, Gnome 40k, KDE 40k...

Invalid. see Google+

According to Google Trends Ubuntu is immensely more popular than any other GNU/Linux distribution (multiple times more than all others combined).

Invalid, incomplete statistic. Google Trends breaks down by keyword but not what other terms were attached to that keyword. While this may be appropriate for things like bands this makes it invalid for declaring anything when it comes to software, and thus we don't know if this is people searching for news, people actually looking to use Ubuntu, or people having trouble with Ubuntu. Furthermore another possibility is that they're going to the Ubuntu forum looking for help with their problems while not using Ubuntu. This statistic is simply invalid due to it's incompleteness

Alexa shows similar trend in ammount of page hits. In the global rank Ubuntu is 1559th, Fedora is 8000th, Linux Mint 7500th, Gnome 18000th, openSUSE 10000th... (their respecive home pages).

Circumstantial, home page hits do not directly translate into number of users. Particularly given that it's an easy platform to get to the forum or other services they provide.

Ubuntu is being sold in both India and China in hundreds of stores.

Invalid, Ubuntu is available in hundreds of stores in India and China however you did not give us statistics on units sold. Furthermore even if you gave us those statistics it would be circumstantial because we don't know if the users would end up wiping and installing their own distro of choice.

...and about Steam:



It can be confirmed from the Steam hardware survey too. You see the trend here?

Invalid, Valve distro statistics are invalid for up to a year or two after they launched Steam for Linux due to them seeding their userbase by pushing Ubuntu on launch. There are other reasons that Steam isn't a particularly good statistic right now but for the time being while valid to find the steam user linux breakdown, it *cannot* be used to say anything about the larger ecosystem.

Quite frankly there really isn't a good way to do this online right now, valve will be in the future but not right now. If you really want legitimate data what you need to do is get a bunch of people to have feet on the ground and go around asking people what OS they use and if Linux what distro, the other alternative is for all distros to start including a hardware and OS reporting software that reports to a central database that tabulates that data, Fedora used to do this but nobody else joined in.

[monraaf]

^no group hates amazon integration more than slashdot yet they are sticking with canonical

[Luke_Wolf]

^no group hates amazon integration more than slashdot yet they are sticking with canonical

Internet based opinion polls are completely invalid, due to the Reddit/4Chan Effect.

[dee.]

FWIW, I believe it to be (coincidentally) probably mostly accurate to say that Ubuntu has around 1/3 share of the desktop Linux home user market. It probably has the largest share of any single distro, but certainly isn't used more than other distros counted together. (Plurality, not majority.)

Forum usage statistics are a very invalid method of counting user amounts, as can be shown by a very simple question: How many people have accounts on Windows forums? Compare that to the amount of windows users...

Also, I have an account on Ubuntu forums, despite the fact that I haven't used Ubuntu for quite a while. I'm willing to be that many people who use buntu-derivatives (Mint, Kubuntu, Xubuntu, etc.) also have accounts on Ubuntu forums. They're, after all, a good resource for troubleshooting for anyone using a buntu-derivative OS. Sadly, that might change soon, as Ubuntu differentiates itself more and more from the rest of Linux OS's...

There really isn't an accurate way to measure distro popularity, as there isn't a really accurate way to measure OS usage stats in general. We can make ballpark guesstimates, but not much beyond that.

[Teho]

Watch your language! BTW you are just shooting the messenger.

If you actually stuck to the facts you there wouldn't be a problem, unfortunately that's not the case for the already mentioned reasons.

I can't take the honour for KDE going down. The contributor lost interest in KDE's bloated buggy code, and that's it!

...and what's that supposed to prove? The Gnome Ohloh.net repository tracks stuff like GTK+, glib and Clutter that are provided by Qt for KDE. It also tracks stuff like GStreamer that's also used by KDE/Qt. Also if you actually had even the most basic critical thinking skills you would notice the suspiciouly fast rise in contributions and very sudden drop to the previous level on activity in KDE. Unless that data is verifiable by other sources you should at very least question it and not draw some clearly false conclusions like "KDE is dying". Also when software projects reach maturity their contributions usually drop. Also I don't see any evidence for your claim that it hasn't anything to do with supposedly "bloated" or "buggy" code.

Quite frankly there really isn't a good way to do this online right now...

Yeah and for that reason I brough up no less than eight points that all point to same direction. Like I said people are free to believe what ever they want but I really would like to see some even somewhat notable statistics that prove anything but the utter dominance by Ubuntu. Ohloh.net and DistroWatch being among the worst possible sources imaginable (very specific, small and technical audience). I guess you could also follow the (mainstream) news sites ie. The Verge and Engadget (who almost solely report about Ubuntu), twitter buzz, blogs (most of which focus on Ubuntu)... in the end doing the research takes bit too much time for my liking. I personally would love to see some other distribution pass Ubuntu but I don't see that happening because there doesn't seem to be any distribution anywhere close to Ubuntu's popularity and brand recognition.

[TheBlackCat]

Watch your language! BTW you are just shooting the messenger. I can't take the honour for KDE going down. The contributor lost interest in KDE's bloated buggy code, and that's it!

Still spreading lies I see. For everyone's information, this only tracks the master branch of git repositories. A lot of KDE development right now is going on in frameworks 5-related branches, which will not show up in those numbers until they are merged later this year or early next year. So the numbers are simply useless, they miss a large part of the developers and development going on right now.

Honton knows this, but continues to dishonestly pretend not to.

So Honton is a proven liar. Just put him/her on your ignore list and save everyone else the trouble of having to see the quotes.

[mgraesslin]

For everyone's information, this only tracks the master branch of git repositories. A lot of KDE development right now is going on in frameworks 5-related branches, which will not show up in those numbers until they are merged later this year or early next year.

Not entirely true. I just checked on ohloh.net and for kdelibs the frameworks branch is also tracked. For other repositories it isn't. E.g. kde-workspace is only tracking master and thus for example the 160 commits to kwin last month missing.

In general the numbers from ohloh.net have to be taken with a salt of grain. New repositories are missing. E.g. kscreen is missing, bodega is missing, the gsoc project I'm mentoring is missing, etc. etc. It's not auto-synced with KDE's project files.

Also the drop down in the stats a few years ago is rather pointless. It's when KDE started to switch from svn to git. As a matter of fact people used the svn repository also as a "scratch" area. With git that completely goes away. Creating a local repo is rather trivial and even if you want to backup it on KDE infrastructure you just create a scratch repository which will never make it to ohloh.net.

Number of commits is also pointless as in SVN times the number of commits was about twice the number of what we have in git as we did backports. Nowadays we do merges. So the number goes down obviously.

So in summary: awesome graph! Unfortunately not telling what it looks like.