[5108] in SIPB bug reports
tex installation
daemon@ATHENA.MIT.EDU (ghudson@MIT.EDU)
Mon Jan 16 20:34:02 1995
From: ghudson@MIT.EDU
Date: Mon, 16 Jan 1995 20:32:56 -0500
To: bug-sipb@MIT.EDU
Cc: amgreene@MIT.EDU, cfields@MIT.EDU
Okay, I think this reflects a loose consensus between bert, myself,
and Chad on what should happen with the tex stuff currently installed
inthe sipb locker:
* It should be installed in a locker named "tex".
* The existing tex locker should be considered obsolete.
* The new tex locker should be in the sipb cell.
* Symlinks and/or files should be left in the sipb locker for
backward compatibility for a long while until people learn
to reference the tex locker.
The reasoning for this is:
* tex is too big of a package to conveniently remain in the
sipb locker, since the sipb locker is replicated in the
athena cell. This is especially true now that we would like
to support platforms such as NetBSD and Linux which do not
have the tex fonts installed in /usr/athena/lib/tex.
* tex is part of the Athena release, and therefore does not
need as high a level of support (outside of the release) as
is provided by either the sipb locker (with the athena cell
replication) or a locker in the athena cell.
* Having tex be in a locker other than "sipb" allows the
administrative control of the locker to include people other
than sipb members. This might help stop the proliferation
of useful tex utilities into various lockers like consult
and gnu.
* Having the tex volumes be in the sipb cell allows more
flexibility for SIPB types to extend the quota and release
the locker. This is a good thing as long as the interested
parties are SIPB types, which they currently are.
* The name "tex" is a short, descriptive locker name which is
easy for users to remember.
* The existing tex locker hasn't been touched since 1991,
apparently, and I don't think anyone relies on it. If they
do, they shouldn't experience any more trauma than tex users
did with the 7.7 Athena release.
So what needs to happen is:
* We will adapt what's currently in /afs/sipb/project/tex to
be a complete tex installation with appropriate binary
directories, etc..
* After everything is ready, we will ask for the hesiod filsys
pointer for "tex" to point at /afs/sipb/project/tex instead
of /afs/athena/astaff/project/tex. If this is unlikely to
be approved (Craig?), we would like to hear the reasons for
it now.
