[2536] in Release_7.7_team
Re: IMAP text client
daemon@ATHENA.MIT.EDU (John Hawkinson)
Sun Jan 14 15:21:31 2001
Date: Sun, 14 Jan 2001 15:21:16 -0500 (EST)
Message-Id: <200101142021.PAA16864@multics.mit.edu>
To: Garry Zacheiss <zacheiss@mit.edu>, Greg Hudson <ghudson@mit.edu>
CC: release-team@mit.edu
In-reply-to: "[2535] in Release_7.7_team"
From: John Hawkinson <jhawk@MIT.EDU>
pine versus mutt
I believe the choices before us are:
a) Include pine in the release
b) Include mutt in the release
c) Include pine and mutt in the release
d) Include neither in the release
Greg argues reasonably that (d) shouldn't happen, because we'd like to
encourage users to use an IMAP mail client, or at least test it,
within the confines of the release, and we believe that having a good
supported mail client as part of the release is an appropriate thing.
Presence of a mail client in the release is an indication of
"supportedness," as well as somewhat of a marketting win. In general,
it will always be easier to tell users to use the client in the
release rather than in a locker. Witnessthe popularity of "xmh &" over
"add sipb; exmh &". This is relatively unfortunate, since exmh seems
to be an all-around better mail client than xmh, which has stagnated.
It would be nice to avoid this situation in the future.
But of course, I don't want to delve too far into lockers-vs-packs.
Traditionally, I don't think that we've done the best job of
anticipating support commitment that a given piece of software
implies. But it's an exercise that's important here. Presumably the
criteria for what to import are basically governed by the development
cost (Greg's) and the support cost (owned by nebulous others?). Clearly
the support cost for two clients is higher than the support cost for one.
Is it significantly higher and is it worth the cost? Hard to say.
If we have to choose only one mail client, I think it's quite reasonable
to choose pine. pine is essentially a naive, beginning user's mail client,
with menu-driven operation and some additional features that extend
to intermediate users. mutt is much more of an intermediate and advanced
user's mail client, is more configurable and scriptable than pine, and
has a wider array of features. It is a bit more difficult to use, though,
especially if you don't have a unix background (not "hard", mind you,
but certainly more difficult than pine). pine and mutt are reasonably
comparable for their intersection, the intermediate-level user.
I think it's more important that the client in the release cater to the
beginning user than to the intermediate user, though. Intermediate users
and advanced users can figure out how to use lockers, etc.
I'm reluctant to see mutt lose the marketting advantage of being in the
release, though. I think that many-to-most Athena users would be better
off using mutt than pine, but not necessarily on Day One of their Athena
experience (and mail needs to work from Day One).
So I would advocate installing both in the release, if feasable,
though if we had to choose one, I think pine would be the better choice.
Perhaps incidently (or perhaps not), I think mutt has a higher release
frequency than pine, and so may be somewhat more work from the
development side to keep in the release. I'm not sure whether this
should effect the decision (it seems to me unfortunate if the work of
maintaining mutt was beneath the threshold, but mutt+pine would be
over the threshold).
To summarize, my vote is for both, or barring that, reluctantly for pine.
--jhawk
(a happy mutt user)
