[241] in athena10
Re: Graphical login session configuration
daemon@ATHENA.MIT.EDU (Greg Hudson)
Wed Jun 11 09:23:19 2008
From: Greg Hudson <ghudson@MIT.EDU>
To: Timothy G Abbott <tabbott@mit.edu>
Cc: athena10@mit.edu
In-Reply-To: <alpine.DEB.1.10.0806101652550.26830@vinegar-pot.mit.edu>
Content-Type: text/plain
Date: Wed, 11 Jun 2008 09:22:33 -0400
Message-Id: <1213190553.18347.10.camel@error-messages.mit.edu>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
On Wed, 2008-06-11 at 02:03 -0400, Timothy G Abbott wrote:
> Is there a reason why /etc/X11/Xsession.d isn't ideal? It has the nice
> property that it doesn't require modifying anything existing.
It's what I'm leaning toward. Issues are:
1. What's in there might differ from dist to dist.
2. What's in there on hardy appears kind of confusing and muddled,
though it might make some sense on a more careful reading.
3. I didn't list this as a goal, but processing the user's .environment
or .bash_environment file once in the top level session shell is good
for dotfile compatibility. To do that for tcsh would require the
somewhat underhanded trick of performing the "exec" of the session
command from within the script we drop into /etc/X11/Xsesssion.d.
