On Thu, 10 Oct 1996, Miguel A.L. Paraz wrote:
> > Here's a copy, since it didn't get into the last beta. Comments?
>
> Excellent... let's start by defining the framework. I've been wanting to
> hack into Majordomo for a long time, but the current code, but, I'm afraid
> I don't understand it too well. I think the switch to Perl5 will help
> readability a lot.
Agreed. First and foremost, since there seems to be many who want to work
or help with mj 2.0, there MUST be organization. Chan is, obviously, in
charge. He has been up to this point, and unless he wants to give it to
someone else, I see no reason that should be changed.
The various packages should be handed over to specific people who would
then be in charge of those. Chan would coordinate between the packages
(resend, majordomo, digest) and the "package-managers" would coordinate
work on their specific packages.
As for somebody working on something that they are not "officially" a part
of, would go by consulting the package-manager -- or CHan if it's a global
majordomo thing.
Might seem like a bit of formality, but if there's going to be a major
overhaul it will definetly help allieviate confusion and multiple
(possibly incompatible) modifications and patches. Another benefit might
be allowing specific packages to release their own updates, with "formal"
global updates released every month with whatever has been released up
until then -- a two-step system.
> I think the core shouldn't even know what the commands are. Instead
> you'll have a command-handling front-end for e-mail, and one for
> WWW. And, another front-end for putting out the messages - whether
> as individual messages, as a digest, as a file. You don't even have to
> limit it to plaintext - you can output it as HTML directly to a WWW
> directory.
Would definetly allow for the easy addition of commands.
> This would lead to a very generalized Majordomo - it's no longer just a
> mailing list manager. You can replace WWW discussion boards,
> and guestbooks. You can use the available NNTP modules and do Usenet
> with it. All using the same engine.
I'm not looking for majordomo to be a "solve-all". There are currently
some very good guestbooks, WWW discussion boards, and Usenet clients out
there. Tying those into majordomo to allow INTER-connectivity is fine, but
replacing them is just too much.
As for mj 1.95, I think it should have a set and defined goal. I don't
want it to take 6 months, because that just delays 2.0. Bugs should be
fixed, uncomfortable situations need to be corrected and anything that we
deem as necessary should be added. Other than bare-minimums and we're
working too hard when energy could be put to 2.0.
-------------------------------------------------------------------------
| Brock Rozen | brozen@webdreams.com | http://www.webdreams.com/~brozen |
-------------------------------------------------------------------------
Follow-Ups:
References:
|
|
