Re: MATE 1.8 has now fully arrived in Debian
Thorsten Glaser <firstname.lastname@example.org> writes:
> Yes, I fully agree. But _please_ also realise that there are people,
> a non-neglibile number of them, for whom these frameworks are not an
> improvement, and who wish to be not forced to use them.
That's fine for you to feel that way, but that feeling does not obligate
anyone else to do work, nor does it obligate upstreams who see code and
conceptual simplification benefits for dropping non-logind approaches to
maintain support they don't like. If you want to be able to avoid new
frameworks that the general community of Linux developers feel materially
improve the quality of the Linux desktop experience, you're going to have
to get together with like-minded people and actually do the work, on an
ongoing basis. Some of those upstreams have indicated willingness to keep
other interfaces provided someone is actively engaged with them about
those interfaces and doing development on them. For other upstreams, you
may have to fork, or just say that people who don't want to use systemd
won't be able to use that software.
My family had a Betamax VCR. The format was arguably much better than
VHS. It didn't get support, maintaining both formats wasn't viable, and
it died. We bought a VHS VCR.
That's not necessarily what's going to happen with systemd (in particular,
it's much easier to maintain multiple init infrastructures than multiple
commercial VCR tape formats). But there's still work involved that
someone would have to do. It's easy to express your disappointment on
mailing lists and much harder to say "here's the new infrastructure that
replaces PolicyKit and ConsoleKit, here's what it's going to be able to do
and what it isn't going to be able to do, here are the situations where
you'll just need to use logind instead, and here are patches to GNOME to
make it work properly." Which is why no one's done that yet.
Russ Allbery (email@example.com) <http://www.eyrie.org/~eagle/>