On 02/10/13 19:21, Adam Dingle wrote:
On Tue, Oct 1, 2013 at 4:22 PM, Sebastien Bacher
<[email protected]> wrote:
Le
01/10/2013 21:16, Adam Dingle a écrit :
I've used Ubuntu every day for 7 years and am
active in the GNOME
community. The fact that Ubuntu lags one release behind
GNOME is
already a significant burden for me. I often spend time
building the
newest version of GNOME apps, which can be challenging since
Ubuntu's
libraries lag behind.
If Ubuntu stays with 3.8 for Saucy+1 (i.e. starts to lag two
releases
behind GNOME), I'd quite possibly switch to Fedora or
Debian. Staying
with 3.8 could be fine for most users, especially if
Canonical wants
to focus most of its energy on phones and tablets. But for
anyone who
wants to use the latest GNOME apps and especially anyone who
wants to
contribute to GNOME development, two releases back is just
too much.
adam
Hey Adam,
I'm sorry to read that Ubuntu being behind on GNOME releases
is a burden for you :/
Can I ask if that's the opinion of an user, or from a
developer wanting to contribute to GNOME?
I'm somewhere between those, but actually more of a user. In
other words, I report a lot of bugs and like to comment on the
very latest features, but don't make many code contributions
myself. There's a continuous spectrum from users to power users
to developers, and I think in a healthy software ecosystem they
can all run the same codebase. Suppose that developers are
running release A and users are all running release B. The
greater the distance in time between A and B, the harder it is
to get a useful feedback loop from users to developers (and vice
versa). I think Ubuntu's lag behind the latest GNOME has
contributed to the feeling of separation between the Ubuntu and
GNOME communities, for better or for worse.
As an Ubuntu GNOME developer I think the 1 cycle delay worked out
quite well for Saucy. We had a full cycle to get the 3.8 packages in
the gnome3 PPA's baked, before moving them into the archives, and
largely I think this went pretty well, the only real exception to
this was g-s-d/g-c-c being blocked by ibus for months and ultimately
running out of time to get g-c-c 3.8 in. From the Ubuntu GNOME side,
Saucy is going to be a pretty solid and stable release.
I am however concerned how this will all pan out if it does indeed
end up being a 2 cycle delay. From which I can see it sliding
further downhill as time progresses, until Ubuntu drop the GNOME
stack, leaving us with a big mess to clean up and perhaps mission
impossible to catch up again.
You
probably understand that's it's hard for us to make both
targets happy at the same time, especially in GNOME directions
is less aligned with Ubuntu's which makes harder to include
their newer version.
If you want to write code for GNOME trunk, using the GNOME3
ppa/jhbuild probably makes sense (or Fedora if that seems a
better option for you), you are just not on top of our
priority list for the next LTS (I hope we can get back to a
situation that makes GNOME users happier after the LTS
though).
One of the question we need to answer there, is to know if the
improvements from GNOME 3.10 are going to be enough benefits,
to our users, to justify the bugs/stability issues/lack of
integration that are going to come with the updates? (if we
update, that's going to take our desktop resources, which
means we are not going to be able to do work smoothing rough
edges).
Right. I'm not concerned about specific features from GNOME
3.10 as much as staying closer to the upstream codebase so that
developers and users can work together. I know that the closer
you are to upstream, the more engineering effort it takes to
keep things stable, and that has a cost, so of course this
depends on Ubuntu (and Canonical's) development resources and
priorities.
Anyway, I know I may not be a completely typical Ubuntu user.
The deeper story here is that it feels like Ubuntu is slowly
separating from GNOME, and lagging 2 releases behind GNOME (for
the first time ever in Ubuntu's history, I believe) may just be
the next step in that process.
adam
|
