Re: What is the future of EOG?

[Felix Riemann <friemann gnome org>]

Hi!

Am Freitag, den 26.03.2010, 17:59 +0100 schrieb Javier Sánchez:
> Well, after my contribution to eog with fit-to-width plugin, it's time
> to continue with the work.
> 
> But i think that it's necessary to set new targets for future releases.
> IMHO, eog needs to resolve some serious problems related with core like:
> preload prev and next files (321603
> <http://bugzilla.gnome.org/show_bug.cgi?id=321603>), which have reported
> for a long time.

Well, generally I see many reports on bugzilla as targets. But ordering
them by priority like in the old 2.20 roadmap on the wiki could help.
Speaking of our wiki. It surely fell a bit under the radar since eog-ng
development and needs some cleanup (e.g. the brainstorming page).
Pierre-Luc Beaudoin has thankfully already overworked the plugins page
for it to be an anchor the eog-plugins package.

Regarding the caching. I wouldn't call its absence a "serious" problem
with eog. When profiling eog while loading an image most of the time is
spent in the decoders (libjpeg, libpng/zlib,...). So to be effective the
cache would have to decode the image in the background and keep the
decoded pixbuf data around until it is needed (or not when randomly
selecting images).
This requires memory and also uses the CPU which might not be desired by
the user.

Generally I'd say eog's core is pretty stable at the moment and has no
really serious problems. The most limiting part right now is I think the
old ScrollView, but at least there is something in the plans (the
Clutter enhancement; see below). Still, I'm planning some changes to the
job system which is has several deficiencies at the moment, especially
when working with plugins.

> Other idea is that it's necessary replace the old EogScrollView with
> other view more powerfull like proposed in (*Bug 546504*
> <https://bugzilla.gnome.org/show_bug.cgi?id=546504> - clutter backend) 
> before reaching Gnome 3.0

Would be cool if we could finally manage to complete this. :)
Though I wouldn't speak of replacing the old view but merely offering an
alternative to it (at least for the first releases). Also we need to
determine what we would like to replace. Just the image view or also the
collection and possibly more.


I've also started to assign the GNOME2.32 target milestone to bugs I
want to have in sight for the next release. It's mixture of already
finished stuff and of bugs I think we should really fix for GNOME
2.32/3.0. Feel free to propose more.

Regards,

Felix