[52] in DCNS Development
Re: Filesystem Reorganization for release 7.3
daemon@ATHENA.MIT.EDU (dkk@ATHENA.MIT.EDU)
Thu Jul 4 10:23:13 1991
From: dkk@ATHENA.MIT.EDU
Date: Thu, 4 Jul 91 10:22:05 -0400
To: Ezra Peisach <epeisach@patriot.MIT.EDU>
Cc: Marc Horowitz <marc@MIT.EDU>, Mark Rosenstein <mar@MIT.EDU>,
In-Reply-To: Ezra Peisach's message of Wed, 03 Jul 91 08:52:53 EDT,
> Users who have customized their environment, bypassing the dot file
> scheme which has been in use for over a year, will have to fix their
> environments. Remember that such changes were made by users.
> $athena_path will be modified to DTRT. We cannot help users who
> decide that "they know best".
I think you're overlooking the realities of user-customization. Many
users with customized logins, perhaps even most of them, never decided
that "they know best." They either installed customizations blindly,
following the instructions of someone who knew Athena better than
they; or the modifications were made by another person entirely, and
the owner of the account doesn't know what changes were made, or what
those changes do.
While I think the filesystem re-org is a good idea, I certainly don't
think that "we cannot help users" who customize their account. Most
of the help we can give will be in the form of documentation and
consulting (both staff and volunteer).
Preemptive consulting should help the users who are active over the
summer, but come fall, many old users may come back to broken
accounts. We'll need to have the usual extensive postering, and
educated consulting resources (including SIPB members and other
sophisticated users) to help those users who went away for the summer,
and came back to the feeling that their locks were changed...
We should be able to get an estimate of how many users will suffer by
running a (read-only) script on the big 6 NFS servers. As long as it
doesn't keep track of usernames (just a count of grep-matches), there
shouldn't be any privacy problems.
