[14805] in Athena Bugs
Re: sgi 8.0I: bogus README file when logged in with temporary directory
daemon@ATHENA.MIT.EDU (Theodore Y. Ts'o)
Tue Nov 5 11:18:49 1996
Date: Tue, 5 Nov 1996 11:18:31 -0500
From: "Theodore Y. Ts'o" <tytso@MIT.EDU>
To: Tom Coppeto <tom@MIT.EDU>
Cc: Greg Hudson <ghudson@MIT.EDU>, hoffmann@MIT.EDU, jhawk@MIT.EDU,
bugs@MIT.EDU, postmaster@MIT.EDU
In-Reply-To: Tom Coppeto's message of Tue, 05 Nov 1996 08:51:17 EST,
<9611051351.AA01586@baby-dialup.MIT.EDU>
Date: Tue, 05 Nov 1996 08:51:17 EST
From: Tom Coppeto <tom@MIT.EDU>
I also don't think it's appropriate for something to be added to the
release that affects the mail system without the buy in of the people who
run it and have to discover it after the fact. If there's needs to be a
debate, it should be why that document should be there, not why it should
be taken away.
For the record, this README was present at least as far as the 7.3
release, so it at least has the historical precedence behind it. (It
may be older, but those sys packs aren't on-line on AFS.) Perhaps
whoever added should have talked to the mail system people first, but
keep in mind that it's been this way for at least the last SIX years,
without causing major problems, at least that I'm aware of.
All this being said, it seems to be a bit silly to be having a
meta-debate about should be about whether the debate is to "add
something to the release" or about "whether something should be taken
away".
Can we try to be a bit more customer oriented here, instead of worrying
about each team's own selfish (or should I say stovepipe-ish) concerns?
If a user can't get access to his/her home directory, what do we want to
tell them? (Remember, the message we're talking about only comes up if
the user's home directory can't be attached.)
Here's one (facetious) example:
Bwa-ha-ha-ha!!!! Abandon all hope!!! If your parents were too cheap
to buy you a computer for your dorm room, that's too bad on you;
there's now way to save your work. Besides, you didn't want to
do that term paper anyway.
More constructively, we could tell them about bitbucket, along with all
of its limitations. Athena ops might object to that, though, since it
might imply a slightly higher service commitment about the realibility
of files stored on bitbucket, and of the overall support given to
bitbucket.
Ultimately, I suppose if the various teams can't get together and solve
this problem without invoking management, we can always kick it up to
ACMG.
- Ted
