Re: [gnome-flashback] Debug-ability; getting gnome-panel 3.8 out

[Philipp Kaluza <floss ghostroute eu>]

Hi Alberts,

Am 21.10.2013 09:59, schrieb Alberts Muktupāvels:
> Hi!
>
>     Independent of that I would be nice if you or one of the other two
>     maintainers could comment on the bugs that I and Alberts have
>     touched in the last two weeks and review some of the patches, I
>     think except for some of the bigger patches, most of the others
>     can be reviewed and marked as commit-now. I am currently
>     hesitating to review more of the open bugs because there has been
>     no response for the last two weeks, but I am sure I could also
>     write a few small patches for some of the open bugs and I will do
>     that if I feel that its worth the time that I put into.
>
>
> Low activity, no response is reason why I decided to work on forked
> branch in github.
Are you being serious ? As far as I can tell, you got a lot of valuable
reviews in gnome bugzilla in the last two weeks, since you actually
started posting your patches there.

Please define "forked branch". Are you creating a feature branch, or yet
another fork ?

Frankly, I am a bit weary of applying your patches, as some seem like
you replace existing and working code by cargo-culting somebody else's
solution to a specific deprecation. This has lead to a bug in the
shipped 3.8.0 and an undocumented Gtk+ dependency increase. Please put
in the work to understand what the different functions actually do, and
explain it on patch submission. This is a work the patch submitter has
to do, because the code reviewer or maintainer cannot always see the
full context of everything touched.

I am not being paid to work on gnome-panel, so the time I can put in is
quite limited. (Anybody that wants to change that feel free to contact
me.) Therefore I am way more interested in repairing people's experience
with gnome-panel, salvaging what we can from g-s-d, and overcoming the
mate-panel fork, than following every Gtk+ deprecation just for the sake
of it. So that's where I put my time.

I'm not saying there is no value in your deprecation work, in fact I
highly appreciate it, but the benefit / risk ratio just is off for the
current state of the code-base. Please help me ship a good 3.8.1 (will
write a mail about that soon) and then let's discuss this long-term
maintenance work for 3.10 or maybe 3.12.

Cheers
  Philipp

-- 
Philipp Kaluza
Ghostroute IT Consulting