Login Manager Story

[Mark Gaiser]

On Wed, Jan 15, 2014 at 9:20 PM, Sebastian Kügler <sebas at kde.org> wrote:
> Hi,
>
> In this email, I'm sharing the results of our session about the login manager
> during the Plasma sprint in Barcelona. It's not conclusive, but gives a pretty
> good overview of the situation. These notes concentrate on the two most viable
> options right now, but we realize that the decision for or against a login
> manager is in the hands of our downstreams. What we can do is make the
> integration easy, but also only to a certain degree. Pros, cons and the
> preliminary conclusions are marked with + and - respectively.
>
> SDDM:
> - Wayland support not yet there
> - Core functionality is incomplete
> - we'd likely end up maintaining it
> - Qt5 port unfinished
> + possibly more control about direction
> + GPLv2+
> + Qt/C++, ties in well with our skillset
> + upstream is interested in Wayland support
> + we could collaborate with Red Hat
>
> LightDM:
> - Copyright assignment
> - Wayland story utterly unclear
> - C/QObject-based
> - bad experience with close collaboration with Canonical
> + well-tested, works today
> + Qt5 backend available, front-end to be done
> + LGPLv2
> + short term, it's less work
>
>
> For both cases, the Wayland story is quite unclear still. The decision is
> definitely not a clear-cut one, both have significant benefits and
> disadvantages.
>
> Our idea of direction so far is to provide a set of QML files so that both can
> integrate well visually, and revisit the situation after this summer.
>
> Feedback is of course welcome and appreciated.
>
> Cheers,
> --
> sebas
>
> http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9
> _______________________________________________
> Plasma-devel mailing list
> Plasma-devel at kde.org
> https://mail.kde.org/mailman/listinfo/plasma-devel

SDDM looks like the better candidate.
My reasoning? Well.. Canonical has a tendency to do stuff that other
projects might not like so much hence LightDM might not be the best
solution here.
Also, KDE = C++/Qt all the way, SDDM is that as well. LightDM on the
other hand is C and GObject (not QObject ^_-) based which might not be
ideal either.

So just from those two points the most ideal match is SDDM.

But what about KDM? Will it silently die once Plasma Desktop 2 / KDE
Frameworks 5 is fully released?
^^ Damn, that is a mouth full. I meant to say "once KDE 5 is released"
but we don't call it that way.