On Feb 1, 2014 3:57 PM, "Alex GS" <alxgrtnstrngl(a)gmail.com> wrote:
You completely missed the whole point of my post.
> You wrote:
> That's just not feasible. Even though our MATE maintainers invest a
> great amount of effort to keep it going, there just aren't enough of
> them to maintain a collection of software as big as Gnome 2 as Fedora's
> primary desktop.
It will involve equal development resources and effort to adapt Gnome 3
Shell to
the Workstation use-case and the traditional workstation metaphor.
Workstation's have a totally different set of requirements and expectations
than Desktops. You can't play around with the UI/UX at all with a
Workstation, it has a conservative nature.
I think you are very wrong in your estimates of the work required. Adapting
G3 to a more sophisticated workflow would be far easier than bringing G2 up
to the technical expectations that G3 has brought.
Let G2 rest.
>> You wrote:
>> innovate. Not exactly a good match for Fedora, right?
>
>> They have to play catch-up for a while before they can
even start to
> Most workstation users today
either use Mac OS, Windows 7 or if they're
on Linux they're still using Gnome 2 in older versions of RHEL. The point
is that if you intend on having a meaningful user-base moving to anything
but Gnome 2 and the traditional desktop metaphor is not possible with this
product.
> Try telling Windows workstation users to abandon Windows 7
and jump to
Windows 8 and you'll see what I mean.
> One of the major goals of the Workstation PRD was:
I think you've got good point buried in here. :)
The intended audience for G3 is exactly the opposite of the user that
Fedora Workstation is targeting. I want to stress that I certainly don't
mean the workstation user is going to, or needs to, constantly adjust the
theme, wallpaper, fonts, animations, whatever but there are things they are
going to want: like group tab switching/window grouping (the virtual
desktop has been maintained when the metaphor should be jettisoned for what
we are really concerned about: tasks), some way to quickly and consistently
monitor intermittent but running tasks, etc. But G3 has too many good
ideas, and is too close to being excellent for specialized workflows (like,
say, animators at large studios) to drop entirely.
Look at rhel 7: it's using G3 classic mode by default. We don't need to do
as rhel since we aren't rhel but I think it is a good idea, given who this
is targeting, to see what's behind the changes from default G3. Personally
I don't believe that the problem was it being "different". In the end
I'd
say that G3 wasn't radical enough with the ux changes, and by being more
conservative they ended up with a ui that doesn't quite fit in any space.
>> Work towards standardizing and unifying the Linux desktop space
> link:
https://fedoraproject.org/wiki/Workstation/Workstation_PRD
> Gnome 2 has several "modern" different forks;
Unity, Cinnamon and Gnome
Shell. These have all resulted in a Linux desktop space that's fragmented
and in serious decline. If you intend on unifying the desktop space you
have to start at square one, back to Gnome 2. Mate is currently moving to
GTK 3 and when that happens which is soon, you'll be able to start bringing
current Gnome programs and make then integrate better.
I think we'd need to see numbers to be certain but I imagine that the main
reason Gnome lost so many users is because of Ubuntu switching their
default.
Considering the API instability and the fact that Gtk is apparently no
longer intended for writing "big apps", the Gnome stack isn't going to be a
target for widespread third party development. Instead we should focus on
providing THE BEST ootb experience of any desktop and let third party
developers target webapi, or use any of the bundling ideas (I liked
Lennart's a bit more since it was designed for the Linux DE and with
security in mind, but anything that "works" is fine, obviously).
> In the next major release of MATE it is going to support GTK3
while
still being compatible with GTK2 as well. The development is mainly done by
Semmu under Google Summer of Code 2013, with the help of the main MATE
developers. You can follow the development at Semmu's GSoC blog.
> link:
http://wiki.mate-desktop.org/status:gtk
> On Sat, Feb 1, 2014 at 2:51 PM, Lars Seipel
<lars.seipel(a)gmail.com> wrote:
>
>> On Sat, Feb 01, 2014 at 12:42:12PM -0500, Alex GS
wrote:
>> > it should be a TRUE workstation-class product.
>
>> You mean RISC CPUs and a proprietary Unix? ;-)
>
>> > Fedora Workstation = Core + Mate (Gnome 2 +
Compton and/or Mutter)
>
>> That's just not feasible. Even though our MATE
maintainers invest a
>> great amount of effort to keep it going, there just aren't enough of
>> them to maintain a collection of software as big as Gnome 2 as Fedora's
>> primary desktop.
>
>> Now looking at upstream MATE's git there are two
people working on it in
>> a regular manner (hope I didn't miss someone) and MATE relies on a whole
>> bunch of stuff that is dead upstream ...
>
>> They have to play catch-up for a while before they can
even start to
>> innovate. Not exactly a good match for Fedora, right?
>
>> Lars
> --
> desktop mailing list
> desktop(a)lists.fedoraproject.org
>
https://admin.fedoraproject.org/mailman/listinfo/desktop
Adopting anything other than a Gnome 3 based system for Fedora Workstation
just doesn't make a lot of sense to me.
That said I think you have a point that, intentionally or not, Gnome 3 has
an interface that seems to be touch inspired and consumption-focused, which
is at is with the workstation target of software devs, cs students,
sysadmins, and the like.
(As an aside, I have been using G3 as my daily desktop for around 4
years--well before the official release. There's a lot to like in G3,
otherwise I wouldn't use it, but it still feels like the best of a
not-great set of choices.)
Ultimately, it seems like we have a conflict between the goals of Fedora
Workstation and those of Gnome. I know there's been discussion about
altering, downstream, some of the design choices of Gnome to better suit
Fedora's users but nothing came of that. However that was before this big
effort to revitalize Fedora (which I am completely behind, for whatever
that's worth:) ). What I would like to see is Fedora's designers look at
any perceived problems in G3's DESIGN and propose some solutions.
Pardon any typos or unclear writing; I'm typing this on a phone :)
Best/Liam