Re: Module semi-proposal: gnome-shell

[Colin Walters <walters verbum org>]

On Wed, Nov 4, 2009 at 1:24 AM, Christian Neumair <cneumair gnome org> wrote:
>
> Reading some of the responses, major concerns wrt accessibility,
> theming and some other topics regarding "quality control" have been
> raised. I'd like to add another one: All the current desktop
> components (including the panel) have undergone usability studies from
> major companies (Sun etc.) and many individuals. Maybe you could point
> out similar usability studies for the shell?

None yet, but it's increasingly high priority.  We are still missing a
good chunk of the design as it is now to be implemented though.

> A proper investigation of
> usability is a prerequisite for not making GNOME 3.0 worse than GNOME
> 2.30.

I agree.

> What happens if GNOME shell does not match the desired quality by the
> time of GNOME 3.0, i.e. what is the plan B? Do we delay it? Do we ship
> unfinished software that ruins our image of shipping good software?

We delay it.

> I am not trying to be rude. I am rather trying to explicitly state
> some concerns that seem to be under the hood of your very own email. I
> am not against changes. I also had some unpublished sketches and
> concepts regarding "the desktop I want" [which IMO should work like a
> personal life planner, i..e datebook + ..., where ... includes files,
> notes, photos, associations with other datebook entries]. It pretty
> much seems to be what you mention with the "Zeitgeist (data colletion)
> engine to collect a rich view of how the user used their computer over
> time". I don't see this kind of data-centric desktop in the GNOME
> shell, so I don't see any new concepts but "just" lots of
> application-centric UI polish.

Well, we have a little bit of the infrastructure for this; the shell
actively tracks and records application usage data, for example.
We're not (at present) using this data in the UI but I have a plan to.