Re: gnome-session
- From: Richard Tibbetts <tibbetts MIT EDU>
- To: Owen Taylor <otaylor redhat com>
- Cc: Richard Tibbetts <tibbetts MIT EDU>, gnome-devel-list gnome org,tibbetts MIT EDU
- Subject: Re: gnome-session
- Date: Sun, 02 Jul 2000 15:59:05 -0400
The problem with a complete rewrite (or even a near complete rewrite)
is that I am working on gsm as part of an somewhat understaffed
project to get GNOME working in the Athena environment. gsm is just
one of the things causing trouble for us, and we have a target
completion date of August 1. At this point I don't have time to take
on a complete rewrite of gsm.
I spent yesterday hacking in the ability to turn session management
off and only use the unmanaged programs list. Apart from a bunch of
newly exposed bygs in other stuff, it is done. This is probably all I
am going to do for now, as it is enough to get us through usability
testing. If you are interested in this patch let me know, otherwise we
will keep it local.
However, I would be more than happy to work on it beginning in late
august or september depending on how well the rest of the Athena User
Interface project goes. I expect that the special needs of Athena
(user sessions will be expected to work on 3 or 4 different OS's all
running GNOME) mean that unless gsm is written here it will not do
what we need.
When is gnome-core expected to branch for 2.0, or do people not plan
that far ahead? Presumably I would need to get a protocol change into
2.0, rather than a patch release.
tibbetts
On 7/2 Owen Taylor <otaylor@redhat.com> wrote:
> Nobody is going to fight you if you want to pick up gnome-session
> and improve it. It's certainly not something I have time to
> work on anymore; I suspect the same applies to Tom. And nobody
> has heard from Felix for a long time.
>
> I would, however, suggest that adding features should not be the first
> step in working on gsm. The way that Felix extended the xsm protocol
> by hacking a protocol on top of property setting has turned out to be
> a big mistake - it makes the code very hard to understand, and
> is also very difficult to extend.
>
> The first thing that needs to be done is to rip out that extension and
> replace it with a CORBA interface. gsm should be using xsm for what
> xsm covers, and a CORBA interface for the things that xsm doesn't
> cover - like getting a list of current clients.
>
> Regards,
> Owen
>
-*- http://www.mit.edu/~tibbetts -*- finger tibbetts@monk.mit.edu -*-
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]