nmh-workers
[Top] [All Lists]

Re: [Nmh-workers] Understanding nmh (aka. What's the goal)

2010-12-01 22:22:04
It seems to me as if you would be doing compatibility for
compatibility's sake. This is sticking to old cruft. Caring to much
for some old userbase likely keep you from getting new users while old
ones slowly vanish.

Why do we need new users?  When did this become a popularity contest?

Sigh.  It's not a popularity contest; it's a RELEVANCY contest.  As
in, nmh is becoming less and less relevant in it's basic function:
communicating with other people via email.

As part of the conversion to git, I had a chance to troll through the
nmh mail archives looking for old email addresses, and something stark
came across to me: we're losing users.

These aren't casual users; for example, Kimmo Suominen, someone who did
a LOT of nmh coding back in the day, announced that he finally gave up
on nmh here (in 2005!):

    http://www.mail-archive.com/nmh-workers(_at_)nongnu(_dot_)org/msg00429.html

Like it or not, MIME messages are a reality in email today.  For
example, when my boss sends me email from her Blackberry, it comes
as a text/plain but encoded in base64, which nmh doesn't handle
very well.  What exactly am I supposed to do here; tell the person
who decides whether or not I get a raise, "Hey, your email client
sucks, get a better one"?  Especially when every OTHER email client
handles this just fine?  I doubt she has any idea what base64 is and
quite honestly there's no reason for her to know about it.

MH was written by and for people who have a deep understanding of how email
works, and who want to exploit the capabilities of email to the n-th degree.
These people also tend to be pretty hard core about the fundamentals of
software engineering, one of which is avoiding change for changes sake.

I agree that was true maybe 20-30 years ago, but I am wondering:
What does Marshall Rose or John Romine use as an email client today?
Maybe they still use mh or nmh, but it sure wouldn't surprise me
if they don't (if we've lost Jerry Peek, then the game really IS
over).  Hell, I met Brent Welch, the author of exmh, a few months
ago at a conference; he told me that while he still wants to support
exmh, most of the time his email client is PC-based.  And if your
email provider goes to an IMAP solution, you'll get to a point where
nmh simply won't work for you anymore unless you're really hardcore
and willing to cook up a bunch of crazy solutions ... and those
people are becoming rarer and rarer.

And MH, by its very intent, is a highly flexible tool.  When you understand
the value of its tool-based approach to handling mail, you'll realize that
most of the functionality you want you can add yourself by writing shell
scripts around the existing commands; there no need to add everyone's
pet functionality into the core.

I don't think anyone is suggesting the basic idea of the tool-based
approach that is the core of the MH concept be abandoned.  But we
do need to think about the email we receive today and how we deal
with it.  My mom sends me email today that nmh doesn't handle well;
I don't know what my daughter's email is going to look like, but I
can pretty much guarantee that she's not going to limit herself to
7-bit ASCII :-/  And while I like the nmh gang here just fine, if
I can't use it to communicate with my mom or my daughter, then what
the hell is nmh really good for?

I don't know yet if I agree with Markus's proposed changes (I will
confess that I haven't read it completely; that's not due to a lack
of interest, but a lack of time), but at least he's starting
a conversation that needs to take place.  For that I am grateful.

--Ken

_______________________________________________
Nmh-workers mailing list
Nmh-workers(_at_)nongnu(_dot_)org
http://lists.nongnu.org/mailman/listinfo/nmh-workers

<Prev in Thread] Current Thread [Next in Thread>