Re: New module proposal: LightDM

[Ray Strode <halfline gmail com>]

Hi,

On Sun, May 15, 2011 at 1:58 PM, Brian Cameron <brian cameron oracle com> wrote:
> Your report missed the following bug, which is a better example of some
> of the more serious issues Canonical had working with upstream GDM:
>
>  https://bugzilla.gnome.org/show_bug.cgi?id=587750
>
> As you can see in the bug report, Robert was not provided with much
> real support or guidance about how to move forward
As far as I know, every time "no gdmsetup" has been brought up the
response has always been more or less that GDM shouldn't act like a
bolted on app with its own setup UI.  Instead, the various bits of
configuration that are supported should be exposed in an integrated
fashion with the rest of GNOME.  That means putting some stuff in the
(at the time unwritten but planned and frequently discussed) accounts
dialog, relying on system default values for certain things, and
potentially adding "Make default" buttons to places in control-center.

I know you and I personally discussed this at various points (and
seemingly came to consensus on).  I'm pretty sure Robert and I
discussed it (but could be I'm misremembering and it was Martin or
Sebastien I talked too, not 100% sure).  Jon also sort of mentioned it
on the bug report you point at above.

Anyway, we can't all agree on every change.  We have to evaluate each
one on a case-by-case basis and do what's right for the project as a
whole.  I think on the whole, though, GDM does a pretty good job at
what it's supposed to do, has a pretty good track record.

It totally could be that this particular bug is the reason is Robert
wrote LightDM.  I don't know.  I don't know the history there at all.
I guess it doesn't *really* matter though. It's fine he's writing
LightDM!  I wish him the best for that project.  I'm sure that project
has its place and its own intrinsic value.  I just don't think its
place is in GNOME right now.  We have something in that place already,
and it's serving its role well enough.

--Ray