User: Password:
|
|
Subscribe / Log in / New account

Results of the Apache OpenOffice 4.0 Logo Survey

Results of the Apache OpenOffice 4.0 Logo Survey

Posted May 10, 2013 20:08 UTC (Fri) by rcweir (guest, #48888)
In reply to: Results of the Apache OpenOffice 4.0 Logo Survey by mjw
Parent article: Results of the Apache OpenOffice 4.0 Logo Survey

You can do the math yourself in a spreadsheet, or simulate it in python. Take a model where developers contribute to an open source project and then move on. It happens to every open source project. The contributor base shifts, you have some core group that remains constant, but many others are passing through.

Given that model, what do you expect the relationship to be between the number of committers in a 2Y period of time, versus a 1Y period of time, versus a 3 month period of time?

You can do the measurement yourself on any open source project, and unless it is stagnant you'll see the same thing.


(Log in to post comments)

Results of the Apache OpenOffice 4.0 Logo Survey

Posted May 11, 2013 19:46 UTC (Sat) by mjw (subscriber, #16740) [Link]

I think you are misreading the point that the original poster in that thread was trying to make. That the number of active apache contributors is in a downward spiral and what to do about that. If I do as you suggest some quick measurements on similar projects it does seem the poster was correct and other projects see a different (active) committers pattern. It is hard to do statistics on openoffice since they are still stuck with subversion, but luckily libreoffice has a converted git branch for the project to easily cherrypick.

Take the commits, committers and active committers, who made at least 10 commits in the last 4 months, and compare them with the numbers of the 4 months before that (just the core code, no website edits):

Sep/Dec 2012
aoo          562 commits,  29 committers, 14 active.
calligra     770 commits,  49 committers, 13 active.
libreoffice 6541 commits, 157 committers, 59 active.
Jan/Apr 2013
aoo          576 commits,  17 committers,  9 active.
calligra    1013 commits,  51 committers, 18 active.
libreoffice 7631 commits, 223 committers, 69 active.
Above using git://anongit.freedesktop.org/libreoffice/core master and aoo branches, and git://anongit.kde.org/calligra master branch. LWN has some code to produce much better stats from git repos if you like: git://git.lwn.net/gitdm.git

Results of the Apache OpenOffice 4.0 Logo Survey

Posted May 11, 2013 20:41 UTC (Sat) by rcweir (guest, #48888) [Link]

You can find stats to prove whatever you want. For example, I just looked at the number of committers for January-April 2012 and compared to the number active January-April 2013. And I looked at only the work in the code, not the AOO website. Guess what? The number of committers was greater in the recent time period, 2013.

Results of the Apache OpenOffice 4.0 Logo Survey

Posted May 11, 2013 21:05 UTC (Sat) by rcweir (guest, #48888) [Link]

Another example:

Looking at the "core" repo for LO, let's apply the same stat you quoted in the parent of this thread to claim a "downward spiral" for AOO.

If we look at committers for LO for this year versus last year we see a 19% decline. If we look at active committers, those with more than 10 commits the decline is 27%.

OMG! LO is declining! It is the death spiral!! Angst, angst, angst, Oh my!

Of course, that is entirely bogus. Comparing different length time periods is invalid. However, that is what the numbers say. But a bad metric for AOO is also a bad metric for LO.

So please be a bit more perceptive in what you quote and what you claim. I think Lwn.net would not be harmed by a bit more intellectual honesty here, even among those who might disagree.


Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds