[Koha] Community Feedback

[paul POULAIN]

Chris Cormack a écrit :
> Theres a lot more than that.
> The whole C4::Context module was done by arensberger.
> arensb	163 commits
>   
Right, I just counted the 10 first.

In fact, I think we should/could separate 2 things :
- historical of Koha
- snapshot of Koha

Let me explain : Oloho displays 8 years of commits.
We can see that we have a large amount of contributors.
http://git.koha.org/gitstat/chart.php can display recent monthly code 
submission.

The differences are interesting I think.
For the last 6 months, we have the following domains : LibLime, 
BibLibre, myacpl, tamil, gmail, washk12, hclibrary, bigballofwax, 
ens-lyon, phonecoop
That's 10 differents organisations (note i've ignored metavore, as it's 
LibLime. gmail I think it's Kyle, although not sure)

on the other side : if you count the number of commits, you'll get :
LibLime : 270
BibLibre : 79
myacpl : 74
washk12 : 24
bigballofwax : 12
All others cumulated : 19

total : 478 commits

percentages : LibLime 56%, BibLibre 17%, myacpl : 16%, washk : 5%, 
bigballofwax : 2.5%, others : 4%

So the question : is Koha "rich of a large number of commiters (10 
differents organisations in the last 6 months)", or "heavily depending 
on 3 major contributors (LibLime, BibLibre, myacpl)" ?

Note : those numbers don't display the RMaint role hdl is endorsing for 
3.0.x He has validated a lot of patches (250) for 3.0.x, but they are 
not counted here (95% of the patches applied on main and 3.0.x, that 
would be a duplicate)

another interesting stats, that I haven't, could be the number of lines 
of a given version written by each contributor.
Something like "there are 100 000 lines of code in koha 3.0 as of dec 
2008. X coming from LibLime, Y coming from BibLibre, Z coming from 
Katipo, ..."
I'm sure the results would be much differents (with my/BibLibre 
contributions being more important, and katipo one much more). In my 
feeling, that would be something like 25% for each me/LibLime/katipo, 
and 25% for all others. but that's just a feeling, and I would be very 
pleased to have numbers !

note that those numbers would be false anyway, as someone commiting just 
a reindented script could be counted as author, which is obviously wrong 
! I think the % is impossible to have. We just can count the number of 
commits (oloho), and the "snapshot" at a given date (git.koha.org)

As a conclusion : none of us should forget the other number : Is LibLime 
proud to be the main commiter this year ? It's fair, but they must not 
forget that they have got a 6-7years old code when they started to 
commit heavily. Katipo or me is proud to be the author of version 1 / 2 
? It's fair, but we must not forget that those days the vitality of Koha 
is not coming mainly from us.

Those are 2 steps of the same piece, and it would not be what it is 
without the 2 !

Developpers & libraries can be thanked as well :
Katipo would not have written koha 1.0 without HLT, I would not have 
been involved in Koha without my catholic community (that asked me to 
find a free software for their library), and LibLime wouldn't have 
existed if NPL had not choosen it in 2004 (joshua used to work at NPL)

/me endorse the role of the wise man ;-)

-- 
Paul POULAIN
http://www.biblibre.com
Expert en Logiciels Libres pour l'info-doc
Tel : (33) 4 91 81 35 08