Re: [gdm-list] Branch update

["Matthias Clasen" <matthias clasen gmail com>]

My thoughts on the issues raised here:

 I would be disappointed if after a complete rewrite, all the messy
config options of the old gdm were slavishly copied in the name of
regression-avoidance ("Use circles instead of asteriks in the password
entry", to cite a particularly egregious example).

A major rewrite should be causing a few "regressions" in that things
are no longer the same as before, because they have been rearchitected
to be different and, hopefully, better.

Throwing in any extra features that may be necessary to make the
rewrite a regression-free replacement for kdm, too is a rathole that
we should not go in, imo.

The current gdm already has far too many random features...

Maybe this is an argument for keeping the current gdm as-is for people
who depend on its current feature set, and turn the rewrite into a
separate project,
I don't know.  But I do know that we (Fedora) want to switch to the new code
base without ifs and buts, and don't plan to do the extra work of
maintaining parallel gdm and gdm-ng packages.

Matthias