looking for "market share" analysis of SCMs.

looking for "market share" analysis of SCMs.

[Ittay Dror]

Hi,

I'm looking for a survey that analyses the market share of the major 
SCMs (including git of course). If someone knows about something like 
that, can you please point me to it? (optimally, such a survey will take 
into account how large the projects are)

Thank you,
Ittay

-- 
Ittay Dror <ittayd@tikalk.com>
Tikal <http://www.tikalk.com>
Tikal Project <http://tikal.sourceforge.net>

Re: looking for "market share" analysis of SCMs.

[Thomas Koch]

Am Dienstag 20 Mai 2008 10:57:38 schrieb Ittay Dror:
> Hi,
>
> I'm looking for a survey that analyses the market share of the major
> SCMs (including git of course). If someone knows about something like
> that, can you please point me to it? (optimally, such a survey will take
> into account how large the projects are)
>
> Thank you,
> Ittay

http://wiki.debian.org/Statistics
http://upsilon.cc/~zack/stuff/vcs-usage/

-- 
Thomas Koch, Software Developer

Young Media Concepts GmbH
Sonnenstr. 4
CH-8280 Kreuzlingen
Switzerland

Tel    +41 (0)71 / 508 24 86
Fax    +41 (0)71 / 560 53 89
Mobile +49 (0)170 / 753 89 16
Web    www.ymc.ch

Re: looking for "market share" analysis of SCMs.

[Teemu Likonen]

Ittay Dror wrote (2008-05-20 11:57 +0300):

> I'm looking for a survey that analyses the market share of the major
> SCMs (including git of course). If someone knows about something like
> that, can you please point me to it? (optimally, such a survey will
> take  into account how large the projects are)

Some kind of "market share analysis" is Debian popularity contest which
gathers automatically information about what packages are installed on
people's Debian systems. Perhaps you can get some useful information
from it. I constructed a link which compares the number-of-installs of
some major SCM/VCS systems:

http://people.debian.org/~igloo/popcon-graphs/index.php?packages=darcs%2Cgit-core%2Cmercurial%2Cbzr%2Csubversion%2C+cvs&show_installed=on&want_legend=on&want_ticks=on&from_date=2003-10-01&to_date=&hlght_date=&date_fmt=%25Y-%25m&beenhere=1

Re: looking for "market share" analysis of SCMs.

[Sverre Rabbelier]

On Tue, May 20, 2008 at 11:32 AM, Teemu Likonen <tlikonen@iki.fi> wrote:
> from it. I constructed a link which compares the number-of-installs of
> some major SCM/VCS systems:
>
> http://people.debian.org/~igloo/popcon-graphs/index.php?packages=darcs%2Cgit-core%2Cmercurial%2Cbzr%2Csubversion%2C+cvs&show_installed=on&want_legend=on&want_ticks=on&from_date=2003-10-01&to_date=&hlght_date=&date_fmt=%25Y-%25m&beenhere=1

A more interesting one I think is http://tinyurl.com/3lf4mn as it
shows the history better, that is, it better shows CVS and SVN
decreasing in popularity (it also shows the point where SVN became
more popular than CVS). From this graph you can also see that git is
on it's way to become more popular than CVS!

http://tinyurl.com/3lf4mn


-- 
Cheers,

Sverre Rabbelier

Re: looking for "market share" analysis of SCMs.

[Daniel Barkalow]

On Tue, 20 May 2008, Sverre Rabbelier wrote:

> On Tue, May 20, 2008 at 11:32 AM, Teemu Likonen <tlikonen@iki.fi> wrote:
> > from it. I constructed a link which compares the number-of-installs of
> > some major SCM/VCS systems:
> >
> > http://people.debian.org/~igloo/popcon-graphs/index.php?packages=darcs%2Cgit-core%2Cmercurial%2Cbzr%2Csubversion%2C+cvs&show_installed=on&want_legend=on&want_ticks=on&from_date=2003-10-01&to_date=&hlght_date=&date_fmt=%25Y-%25m&beenhere=1
> 
> A more interesting one I think is http://tinyurl.com/3lf4mn as it
> shows the history better, that is, it better shows CVS and SVN
> decreasing in popularity (it also shows the point where SVN became
> more popular than CVS). From this graph you can also see that git is
> on it's way to become more popular than CVS!
> 
> http://tinyurl.com/3lf4mn

Doing it as a percentage is kind of odd; it demonstrates that using 
version control is decreasing in popularity as more and more people who 
don't use version control submit votes. All of the version control systems 
show an upward trend in number of users, but most of them are not growing 
as much as the market is shrinking. Unfortunately, there doesn't seem to 
be a way to get the people who use each of them as a percentage of people 
who use any of them, which is what you want for "market share".

	-Daniel
*This .sig left intentionally blank*

Re: looking for "market share" analysis of SCMs.

[Sverre Rabbelier]

On Wed, May 21, 2008 at 12:07 AM, Daniel Barkalow <barkalow@iabervon.org> wrote:
> Doing it as a percentage is kind of odd; it demonstrates that using
> version control is decreasing in popularity as more and more people who
> don't use version control submit votes. All of the version control systems
> show an upward trend in number of users, but most of them are not growing
> as much as the market is shrinking. Unfortunately, there doesn't seem to
> be a way to get the people who use each of them as a percentage of people
> who use any of them, which is what you want for "market share".

Ah, you are right, I didn't analyse the graph enough, I thought
percent would show what you described above, obviously it doesn't
though as the values don't add up to 100%. Even so, I think this is a
better graph:
http://tinyurl.com/4hu2cn
Since it shows users that regularly use the package, instead of user
that have it installed. Considering that these days distro's have a
lot of stuff pre-installed if you go with the "coders packages" the
usage votes give a better picture than looking at who has it installed
(which includes people that don't use the package at all). That graph
shows that in the end of 2006 SVN became more popular than CVS,
whereas the one from Teemu (http://tinyurl.com/4vpqzg) does not show
that, probalby because a lot of people have it installed, but don't
use it.
It would be interesting if we could link the grow spurts from
http://tinyurl.com/4hu2cn to programming-related events, something
that might have caused people to suddenly use a VCS a lot more (for
example the one in CVS's line from 05-2007 till 07-2007).

-- 
Cheers,

Sverre Rabbelier

Re: looking for "market share" analysis of SCMs.

[Jakub Narebski]

Ittay Dror <ittayd@tikalk.com> writes:

> I'm looking for a survey that analyses the market share of the major
> SCMs (including git of course). If someone knows about something like
> that, can you please point me to it? (optimally, such a survey will
> take into account how large the projects are)

It's not exactly "market share", and answers are obviously severly
biased, but you can take a look at answers and _comments_ to question
"10. What other SCMs did/do you use?" in Git User's Survey 2007:
http://git.or.cz/gitwiki/GitSurvey2007#head-4ed59e7615e2dc4f4baa8fa0beccde52c93789f8
(comments mention Ohloh stacks and Debian popularity contents,
popcon).

You can also read the following articles on Elijahs' blog:
  http://blogs.gnome.org/newren/2007/11/17/adoption-of-various-vcses/
  http://blogs.gnome.org/newren/2008/03/01/happenings-in-the-vcs-world/

HTH
-- 
Jakub Narebski
Poland
ShadeHawk on #git

Re: looking for "market share" analysis of SCMs.

[Daniel Barkalow]

On Wed, 21 May 2008, Sverre Rabbelier wrote:

> On Wed, May 21, 2008 at 12:07 AM, Daniel Barkalow <barkalow@iabervon.org> wrote:
> > Doing it as a percentage is kind of odd; it demonstrates that using
> > version control is decreasing in popularity as more and more people who
> > don't use version control submit votes. All of the version control systems
> > show an upward trend in number of users, but most of them are not growing
> > as much as the market is shrinking. Unfortunately, there doesn't seem to
> > be a way to get the people who use each of them as a percentage of people
> > who use any of them, which is what you want for "market share".
> 
> Ah, you are right, I didn't analyse the graph enough, I thought
> percent would show what you described above, obviously it doesn't
> though as the values don't add up to 100%. Even so, I think this is a
> better graph:
> http://tinyurl.com/4hu2cn

Yeah, that's the one I thought was informative.

> Since it shows users that regularly use the package, instead of user
> that have it installed. Considering that these days distro's have a
> lot of stuff pre-installed if you go with the "coders packages" the
> usage votes give a better picture than looking at who has it installed
> (which includes people that don't use the package at all). That graph
> shows that in the end of 2006 SVN became more popular than CVS,
> whereas the one from Teemu (http://tinyurl.com/4vpqzg) does not show
> that, probalby because a lot of people have it installed, but don't
> use it.

Yeah, also a lot of people have CVS installed from when they used it, but 
they not use something different, but haven't built new computers without 
it.

> It would be interesting if we could link the grow spurts from
> http://tinyurl.com/4hu2cn to programming-related events, something
> that might have caused people to suddenly use a VCS a lot more (for
> example the one in CVS's line from 05-2007 till 07-2007).

I think that around 05-2007, a lot more people started voting, largely 
non-coders, but also coders. In the percentage, there's a sharp drop where 
that sharp rise is.

	-Daniel
*This .sig left intentionally blank*

Re: looking for "market share" analysis of SCMs.

[Linus Torvalds]

On Wed, 21 May 2008, Sverre Rabbelier wrote:
>
> Even so, I think this is a better graph:
>    http://tinyurl.com/4hu2cn

Woo-woo!

Add RCS to the list (because real mean use the real thing(tm)), and see 
how git passed RCS late last year.

		Linus

Re: looking for "market share" analysis of SCMs.

[Martin Langhoff]

On Wed, May 21, 2008 at 4:06 PM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> On Wed, 21 May 2008, Sverre Rabbelier wrote:
>> Even so, I think this is a better graph:
>>    http://tinyurl.com/4hu2cn
>
> Woo-woo!
>
> Add RCS to the list (because real mean use the real thing(tm)), and see
> how git passed RCS late last year.

Heh. And trim it to start on 2005-7-12 - the date of the first debian
package for it. Makes the graph more interesting.

Do I confess I added tla to the list too? Some things in my past I
rather not talk about...

cheers,


m
-- 
 martin.langhoff@gmail.com
 martin@laptop.org -- School Server Architect
 - ask interesting questions
 - don't get distracted with shiny stuff - working code first
 - http://wiki.laptop.org/go/User:Martinlanghoff

Re: looking for "market share" analysis of SCMs.

[Sverre Rabbelier]

On Wed, May 21, 2008 at 6:06 AM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> Woo-woo!
>
> Add RCS to the list (because real mean use the real thing(tm)), and see
> how git passed RCS late last year.

I have to admit I never heard of RCS before you mentioned it, but sure
is nice to know we passed them! :D This graph shows only git,
subversion, cvs, and rcs, since the other ones don't really have any
market share and just clutter the bottom of the graph.

http://tinyurl.com/6hj5ds

It looks like a steady growth but I'm not sure we'll pass CVS before
the "end of the world" in 2012. (Just picked the "end of the world" in
2012 as a random reference point since it's relatively close to now
and most people have heard of that theory.)

-- 
Cheers,

Sverre Rabbelier

Re: looking for "market share" analysis of SCMs.

[Sverre Rabbelier]

On Wed, May 21, 2008 at 6:26 AM, Martin Langhoff
<martin.langhoff@gmail.com> wrote:
> Heh. And trim it to start on 2005-7-12 - the date of the first debian
> package for it. Makes the graph more interesting.

Trimming it does make the graph a lot nicer although doing prevents
you from comparing the grow curves (of CVS and SVN with those of git).
http://tinyurl.com/6hj5ds

> Do I confess I added tla to the list too? Some things in my past I
> rather not talk about...

I'm afraid 'tla' is too insignificant to be of interest :P, it's
steady at, say, 100 users.

-- 
Cheers,

Sverre Rabbelier

Re: looking for "market share" analysis of SCMs.

[Martin Langhoff]

On Wed, May 21, 2008 at 4:44 PM, Sverre Rabbelier <alturin@gmail.com> wrote:
> Trimming it does make the graph a lot nicer although doing prevents
> you from comparing the grow curves (of CVS and SVN with those of git).
> http://tinyurl.com/6hj5ds

Actually, if we define the "market" to be DSCM, and the timeframe to
be existence of the git-core package, this graph of installs is quite
useful:
http://tinyurl.com/4uemg2

>> Do I confess I added tla to the list too? Some things in my past I
>> rather not talk about...
>
> I'm afraid 'tla' is too insignificant to be of interest :P, it's
> steady at, say, 100 users.

Maybe git-archimport has made a dent? ;-)
http://tinyurl.com/5okewp

cheers,


m
-- 
 martin.langhoff@gmail.com
 martin@laptop.org -- School Server Architect
 - ask interesting questions
 - don't get distracted with shiny stuff - working code first
 - http://wiki.laptop.org/go/User:Martinlanghoff

Re: looking for "market share" analysis of SCMs.

[Sverre Rabbelier]

On Wed, May 21, 2008 at 7:10 AM, Martin Langhoff
<martin.langhoff@gmail.com> wrote:
> Actually, if we define the "market" to be DSCM, and the timeframe to
> be existence of the git-core package, this graph of installs is quite
> useful:
> http://tinyurl.com/4uemg2

Hehe, that feels a bit like manually favoring the odds in our benefit,
but it's fun to see how git is the most popular DSCM since 2007 ;).

> Maybe git-archimport has made a dent? ;-)
> http://tinyurl.com/5okewp

It would seem the contrary is true, the usage of tla was steadily
declining, but since git-arch it has been picking up! (Well, that is
what one could naively deduct from this graph, there's probably
another explanation :P).

-- 
Cheers,

Sverre Rabbelier

Re: looking for "market share" analysis of SCMs.

[Peter Karlsson]

Linus Torvalds:

> Add RCS to the list (because real mean use the real thing(tm)), and see 
> how git passed RCS late last year.

I have been using RCS for local version control (like in my /etc
directory), when I did not want to involve a CVS server. I find that
Git does that job much better, so I am switching to Git there as well.
Very nice :-)

The graphs are all missing "cvsnt", which does replace the cvs package
if installed. But even if added, cvs is still losing market share.

-- 
\\// Peter - http://www.softwolves.pp.se/

Re: looking for "market share" analysis of SCMs.

[Robin Rosenberg]

onsdagen den 21 maj 2008 07.21.50 skrev Sverre Rabbelier:
> On Wed, May 21, 2008 at 7:10 AM, Martin Langhoff
> <martin.langhoff@gmail.com> wrote:
> > Actually, if we define the "market" to be DSCM, and the timeframe to
> > be existence of the git-core package, this graph of installs is quite
> > useful:
> > http://tinyurl.com/4uemg2
> 
> Hehe, that feels a bit like manually favoring the odds in our benefit,
> but it's fun to see how git is the most popular DSCM since 2007 ;).

Diff+tar+patch should be counted too. http://tinyurl.com/3frawy Seems
far more popilar than anything else added together, including cvs and
other subversive systems.

-- robin

Re: looking for "market share" analysis of SCMs.

[Daniel Barkalow]

On Wed, 21 May 2008, Robin Rosenberg wrote:

> onsdagen den 21 maj 2008 07.21.50 skrev Sverre Rabbelier:
> > On Wed, May 21, 2008 at 7:10 AM, Martin Langhoff
> > <martin.langhoff@gmail.com> wrote:
> > > Actually, if we define the "market" to be DSCM, and the timeframe to
> > > be existence of the git-core package, this graph of installs is quite
> > > useful:
> > > http://tinyurl.com/4uemg2
> > 
> > Hehe, that feels a bit like manually favoring the odds in our benefit,
> > but it's fun to see how git is the most popular DSCM since 2007 ;).
> 
> Diff+tar+patch should be counted too. http://tinyurl.com/3frawy Seems
> far more popilar than anything else added together, including cvs and
> other subversive systems.

I think tar and diff are necessary to install git-core or any other .deb, 
so it's unsurprising that they've got a 100% market share by that metric. 
Until Debian switches to a git-based package format, git's not going to 
catch up to those...

On the other hand, the "vote" value for patch is a reasonable metric, and 
it's been steadily close to but above the version control systems.

	-Daniel
*This .sig left intentionally blank*

Re: looking for "market share" analysis of SCMs.

[Teemu Likonen]

Daniel Barkalow wrote (2008-05-21 16:07 -0400):

> I think tar and diff are necessary to install git-core or any other
> .deb, so it's unsurprising that they've got a 100% market share by
> that metric. Until Debian switches to a git-based package format,
> git's not going to catch up to those...

Yes, tar and diff are in every Debian system and can't be casually
removed:

  $ aptitude remove tar
  [...]

  WARNING: Performing this action will probably cause your system to break!
           Do NOT continue unless you know EXACTLY what you are doing!
  To continue, type the phrase "I am aware that this is a very bad idea":

Re: looking for "market share" analysis of SCMs.

[Martin Langhoff]

And more popularity contests around Debian

The overall thread should appear in the archive around here - (not
there yet at time of writing):
http://lists.debian.org/debian-devel/2008/05/threads.html#00583

This graph is of VCS usage by package maintainers to keep track of the
local patches applied on top of upstream. It sure makes git look good:
http://upsilon.cc/~zack/stuff/vcs-usage/

cheers,



m
-- 
 martin.langhoff@gmail.com
 martin@laptop.org -- School Server Architect
 - ask interesting questions
 - don't get distracted with shiny stuff - working code first
 - http://wiki.laptop.org/go/User:Martinlanghoff