[212] in SIPB bug reports
[Jerome H. Saltzer: nits: SIPB Inessential guide to Athena]
daemon@ATHENA.MIT.EDU (daemon@ATHENA.MIT.EDU)
Thu Oct 6 08:43:27 1988
Date: Thu, 6 Oct 88 08:43:05 EDT
From: John T Kohl <jtkohl@ATHENA.MIT.EDU>
To: bug-sipb@ATHENA.MIT.EDU
Some comments for the next edition.
------- Forwarded Message
Date: Wed, 5 Oct 88 18:34:37 EDT
To: jik@ATHENA.MIT.EDU, jtkohl@ATHENA.MIT.EDU
Subject: nits: SIPB Inessential guide to Athena
From: Jerome H. Saltzer <Saltzer@ATHENA.MIT.EDU>
The fourth edition is an impressive document. I noticed a few minor
glitches as I read it over:
1. file protection: The MIT privacy committee doesn't even know
there is such a thing as default protection for user files on Athena;
the document shouldn't accuse them of deciding that the default file
mode be private. (The message that used to appear in .login where it
set umask was similarly off-base.) The real story is that the
privacy committee has, with advice and input from all quarters, made
a very general statement of M.I.T.'s policy with respect to privacy.
That statement appears in "Policies and Procedures," and it contains
statements like "The Institute is committed to protecting the
personal privacy of members of the MIT community." (You are welcome
to look at my copy if you are interested.) Every department and
operating organization at M.I.T. takes that policy as a starting
point and translates it into more specific policies and procedures
for itself. Athena's translation, which is its own responsibility,
is that the default file mode for users should be private. This
translation is consistent with the Institute policy, but it isn't
dictated in the way that one would infer in reading the paragraph on
page 7.
2. Footnote 4 on page 18 is a little out of date; whichever upcoming
release of the system contains NFS version 3.2 will allow you to
control-C out of a hard-mounted server that stops responding.
3. The discussion of finger could alert readers to interpret the
phrase "never logged in" as irrelevant, to ignore the "local" line
that usually says "in real life: ???" and to expect that a finger of
your own name will produce n copies where n is the number of active
xterms. A general mention that for consistency with traditional UNIX
fingerit looks on the workstation to see if the user is logged in
there, and if not found asks Hesiod, might make finger less
mysterious.
4. The description of sipb materials still recommends use of
"/usr/sipb/..." rather than "/mit/sipb/". Doesn't the latter now
work? If so, wouldn't it be more consistent to encourage people to
use the same form for this locker as with most other lockers?
5. The list of useful lockers could do well to omit newsletter,
which has never been updated since it was created. Some generally
interesting lockers not mentioned are lab-supplies, X11, and xpix
(programs only at this point, but still useful.) There are also now
a bunch of system lockers that might be mentioned briefly. {gnuplot,
hoops, igks, KCL, prochart, RS/1.new, RS/1.old, saber2, scribe6,
2020}
Jerry
------- End Forwarded Message
