*Mops.reply.m.c_replyall: Msg_Reply -group
That doesn't fix exmh. To fix exmh you need a patch to exmh.
Fixing one user is not a solution.
There isn't much I can do about exmh. You'll have to talk to
Brent about that.
This doesn't strike me as being particularly burdensome for the EXMH user to
do. It also doesn't strike me as being particularly burdensome for EXMH to
have a switch at install time for the user to specify whether NMH or MH is
being used under the hood, thus allowing EXMH to default correctly.
Um, except that fairly recent versions of nmh behave the same as MH in
this regard.
The change for `-group' happened some time ago.
Apparently there is not. The change is not even documented right
in the man page. The old broken -cc flag is still listed.
Incorrect. Read the man page carefully. If you read it, you will
notice that the -nocc/-cc flags are not defined unless the -group
flag is given.
MH was moribund before Richard began revamping it as NMH. Anyone who insists
that Richard should adhere strictly to MH's ancient ways is simply asking for
a return to that period of moribundity, which IMO benefits no-one.
Now nmh is no longer a supported version of MH. It's a supported version
of nmh -- its own thing.
It's always been its own thing.
I'm not saying nmh can't evolve. However the huge issue looming over MH
is that it's an extremely old package (relatively), with lots of history.
If it's going to evolve then there has be ways to maintain backwards
compatibility. Otherwise we're abandoning MH/nmh's huge installed base.
I see the reason now why lots of Linux distributions still ship with
the original MH -- compatibility. nmh is close, but not compatible.
If you want MH, then use MH. I haven't forced anyone to change to nmh.
But even MH evolved and changed when it was under development.
Personally, given the choice of having an nmh which was feature-static and
stable versus incrementally-improving-but-incompatible, I'd pick the former.
I don't look for lots of features in nmh -- I use it primarily within exmh
(though I do have a significant number of scripts and other stuff built
around MH/nmh which I use outside exmh.).
People stick with MH (the system, not the program) because it does stay the
same. If nmh is going to evolve away from MH then a large fraction will
stop using nmh (at least nmh past the incompatibility change) or move on
to something else (like IMAP).
Those who want MH, can use MH.
I am thankful for Richard for taking over nmh. I don't want him to stop
working on nmh. I just think nmh is going in the wrong direction.
Then what direction is the right one? Maintaining 100% compatibility
forever? That's pretty boring.
Hacking new code is fun. Maintaining legacy code is work (which I usually
get paid for). When I stop having fun with nmh, I'll stop working on it.
As a matter of fact, I'm about to start breaking backward compatibility
even more. I've put it off long enough, but some things just can't
stay the same, if MIME support is to improve.
rc