[31] in Athena User Interface
Re: Making AUI opt-in for 8.4
daemon@ATHENA.MIT.EDU (Greg Hudson)
Wed May 10 15:26:15 2000
Message-Id: <200005101926.PAA07763@small-gods.mit.edu>
To: Richard Tibbetts <tibbetts@MIT.EDU>
Cc: aui@MIT.EDU
In-Reply-To: Your message of "Wed, 10 May 2000 15:14:29 EDT."
<200005101914.PAA29552@m4-167-9.mit.edu>
Date: Wed, 10 May 2000 15:26:09 -0400
From: Greg Hudson <ghudson@MIT.EDU>
If we're going to do anything of this nature, we need to get aproj.aui
replicated, and maybe even make a new locker for more volatile kinds
of testing. And, of course, we'll need the aui locker to be in decent
shape. (I don't know its current status from a user perspective).
> 1) We frob the 8.4 dotfiles before it goes to release such that
> people who do "set aui" in their .environment cause (for instance)
> /mit/aui/bin/xsession to get exec'd.
This doesn't feel right. It implies too much of a support commitment
in my mind.
> 2) Offer users an xsession to copy into their home directory as a
> private .xsession.
I think I favor a variation on this theme:
* People symlink the /mit/aui xsession file (using the AFS
path), instead of copying someting.
* We have all the warnings as in the #1 case.
* At some future date, when we have software in the release,
the /mit/aui xsession file will remove the user's .xsession
symlink (checking that it is, in fact, a symlink pointing to
the expected place).
That way, we don't leave .xsession files behind in naive users'
homedirs.
