Dan Winship wrote:
>So what's up? Do we have an official new maintainer? Are we going to
>use the offered mailing list and CVS repository? (Is there anyone on
>this list who hasn't subscribed to the new list yet?)
I hope so. Otherwise I'm really wasting a lot of time for nothing. :)
>There have been a bunch of patches posted to fix various 1.0 bugs. It
>would be nice to gather them all together and put out a nmh 1.0.1.
>Perhaps with a new-and-improved -group flag, once we can mostly agree
>on what it should do.
I don't have extreme feelings about -group, but I think it points to a
problem that needs to be solved more generally. In particular: how to
handle interface-breaking updates.
Richard complained that he didn't want to be bound to support the old
interface -- specifically that it would prevent him from spending his
time more productively advancing nmh itself. I'd hope that since more
people will be hacking on it, this will not be as much of an issue
anymore. Though I may be overly optimistic there.
So, the more general quesiton is: Should the old mh-style interface be
maintained forever at all costs, or should nmh advance and be allowed
to diverge from that design? My feeling is that, a -mh-compat flag
would be worth maintaining so that other interfaces like xmh, exmh,
and so on will continue to function with a minimum of pain. But this
won't work for everything. (See the mh-e fixes in the FAQ).
On a related note: What should we look for in nmh-1.0.1, and further
releases. There is a TODO list in the source directory -- I'd say
that provides a good list of things to start with (though I think
getting them *all* done by 1.0.1 would be a tad too ambitious).
On top of that, I'd like to see autoconf configuration finished. It's
a bit of a personal peeve. It makes little sense to have to run
./configure; vi config.h ; make. That's what autoconf is supposed to
save you from doing.
I'm hacking at that, but I'm not an autoconf expert, so it goes
slowly.
Apoligies to those who get this message twice: once from each list.
Since the original was posted to the gatech list I though I should
reply there, but the content of this message seemed like it should be
sent to the new list.
-Doug Morris
