Now we are over 10 million Ubuntu users!

[artir]

According to Matt Zimmerman's blog: http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/ .


The latest I had heard about the number of user was that we were 8 millions in 2006.

[Johnsie]

How do they count this? I have machines running on two ip addresses at home and one at work. Does that make me three users?

[LookTJ]

How do they count this? I have machines running on two ip addresses at home and one at work. Does that make me three users?
Are you counting LAN-wise or WAN-wise.

If you are counting LAN-wise, that makes you two users.

If you are counting WAN-wise, that makes three.

[SomeGuyDude]

Yeah I'm kinda curious how that's tabulated. 10 million seems oddly high.

[Giant Speck]

Who cares if the number is accurate? It just means more people not using *******!

lolololololololololololololololol!!!!!1!!!one

[Sealbhach]

I saw the number 8 million being bandied about last October:

http://www.internetnews.com/software/article.php/3780651

Who knows?

.

[mxboy15u]

Canonical says over 8 million users.

[MaxIBoy]

Yeah I'm kinda curious how that's tabulated. 10 million seems oddly high.Seems a bit low to me. I was thinking something along the lines of 15 million.

[gnomeuser]

I saw the number 8 million being bandied about last October:

http://www.internetnews.com/software/article.php/3780651

Who knows?

.

Sadly not backed by data, either time and number.

I hesitate at calling it dishonest, but it is a consistent lapse in Canonicals behaviour to leave out how they arrive at this number and what degree of accuracy they put on it.

To those who say it's impossible, it simply isn't (http://fedoraproject.org/wiki/Statistics)

Now if 10 million is a good estimate and data can be produced, it would be interesting to see how they were distributed. How many users on LTS releases e.g. would be an interesting percentage to know for ISVs and just in general to know how many users can be expected to upgrade. Another thing would be when in the release cycle people are upgrading, on the day/week of release or later. We could learn a lot of this data.

[LowSky]

Who cares if the number is accurate? It just means more people not using *******!


actually a large percentage are still dualbooting, so that kinda doesn't help the numbers game



we can’t be certain how many people are using Ubuntu globally, but estimate the number to be over ten million based on Internet traffic

Well if it based on the info sent by the Firefox browser than maybe they can get a decent estimate number.

[Giant Speck]

actually a large percentage are still dualbooting, so that kinda doesn't help the numbers game

My post was a sarcastic joke, evidenced by the "lolololololololololololololol!!!!!1!!one" following it.

[SomeGuyDude]

If we're talking worldwide, I suppose I can believe it. It just doesn't seem painfully likely.

Also worth pointing out the high number of Linux users who have a bunch of distros just to play with them. Like I'd be considered an Ubuntu stat because I puttered around with Jaunty but I don't really "use" it.

[ZarathustraDK]

How would you count the amount of Ubuntu-users accurately?

I suppose you could put some kind of phone-home mechanism into the installers, but I don't think it'd fly well with the community, too Windows'ish.

Then you could make a signup-page, but I doubt everyone would sign up.

How DO you do this kind of thing anyway?

[Namtabmai]

How DO you do this kind of thing anyway?

One way is to monitor the number of downloads of a core package, something like libc which everyone needs regardless of their DE. Of course on top of that you have to try and weed out any false positives.

[gnomeuser]

How would you count the amount of Ubuntu-users accurately?

I suppose you could put some kind of phone-home mechanism into the installers, but I don't think it'd fly well with the community, too Windows'ish.

Then you could make a signup-page, but I doubt everyone would sign up.

How DO you do this kind of thing anyway?

Like this (http://fedoraproject.org/wiki/Statistics), it is not spot on accurate but gives you a very valid best estimate.

[Wiebelhaus]

Didn't distro watch clock like 30 million downloads of Ubuntu or something..

[benj1]

well you can count ship-it's and direct downloads and i suppose request figures from companies that post discs for a fee. i don't know about bit torrent is that measurable????
i would be more accurate using an LTS as you can be fairly certain everyone will be upgrading.
as for people that just download to check it out, im fairly certain microsoft counts each preinstallation in its figures even if people subsequently install something else over the top.
i do agree though that it is very hard to accuratly measure total usage, probably the most accurate would be a good old fashioned door to door survey, although that would take money.

[BuffaloX]

I think 8 million in 2006 was the estimated total number of Linux users.
Today that has doubled.

Linux market share according to Net Applications:
http://marketshare.hitslink.com/operating-system-market-share.aspx?qprid=8

Internet world stats Total:
http://www.internetworldstats.com/stats.htm

1,596,270,108 * 0.90% = 14,366,430.

But Net Applications show the lowest stat for Linux, while w3schools show the highest at 4%.

w3schools:
http://www.w3schools.com/browsers/browsers_os.asp

1,596,270,108 * 4.00% = 63,850,804

This is just those connected to the internet.

[Methuselah]

Remember that the w3cschools stat is for that site only.

[Wiebelhaus]

Someone needs to start a far reaching multi-language campaign to register simple information from Linux users in a non evasive way.

It's the least that could be asked for downloading completely free software.

[Sandsound]

Someone needs to start a far reaching multi-language campaign...

I think this site is attempting something similar, but I haven't seen much activity yet : http://statix.sourceforge.net/Home.html

btw. In my family, only 10% use Windows, the rest of us use Ubuntu or Debian ;-)