Re: GNOME: numbers of users, patches/revisions

[Sumana Harihareswara <sumanah panix com>]

Allan wrote:

> What are these metrics intended to indicate? The amount of work that has
> gone into GNOME 3? Changes since 2.0 don't necessarily map onto that. I
> also suspect it will be way too difficult to pull out the required
> numbers for this. You could maybe focus on one or two key modules: lines
> of code changed in GNOME Shell since it began, for instance, or the
> amount of changes between Gtk+2 and Gtk+3. It's not quite as good, I
> know...


On 03/25/2011 08:24 AM, Allan Day wrote:
> Dave Neary wrote:
>
> > > Sumana asked: How many patches/code changes have been committed since the 2.0 release
> > > 9 years ago?  (Tried viewing the repos, can't figure out how to get a
> > > total like this.)
> > You can make a list of the modules you're counting, do a git log -M for
> > each of them, concatenate the results, and then run gitdm on the result.
> The tricky part is determining which modules you include and which
> periods of development you define as 'GNOME 3 development'. This will
> vary from module to module: you'd want to include the whole development
> history of new components like GNOME Shell and dconf. Existing modules
> could probably be counted from 2.30.
>
> You might also want to count the number of modules that have been
> depreciated (and count the number of lines of code they contain). That's
> another relevant metric.
>
> The other problem is what you count as 'GNOME'. I tend to think that
> 'GNOME 3' is roughly equivalent to the contents of gnome-suites-core-3.0
> [1]. I suspect that others might disagree.
>
> Figuring this stuff out isn't an impossible task, but it would take a
> bit of time and thought.
>
> Allan
>
> [1]
> http://git.gnome.org/browse/jhbuild/tree/modulesets/gnome-suites-core-3.0.modules


I shall spend a tiny bit of time trying this, then probably give up and 
ask y'all for an order-of-magnitude estimate.

best,
Sumana