On Thu, 28 Sep 2000, The Hermit Hacker wrote:
> please note that Mj2 is *totally* different then Mj1 ... Mj2, to
> configure/install is a simple 'perl Makefile.PL;make;make install' ... to
> configure a new list is as simple as entering the admin interface and
> typing 'createlist <listname> <listowner>' and it outputs the four lines
> of text you need to add to the aliases file to enable the list ...
But the major difference is that Mj1 is released whereas Mj2 is still not
released.
For those of us in production environments, the release status is vital.
o We happily use a lot of open-source software (sendmail, pine, samba).
o We happily keep up with releases (where necessary).
o We happily investigate bugs, try to fix them, and contribute them back.
o We happily sometimes install beta-releases especially where we have
been active in a bug fix or new feature.
But this is all on released software (or a variant very close to a
release, including an "offical" beta).
I have had a couple of plays with Mj2 in the last two years, but
encountered some problems, which I tried to investigate and at least
report back. But the demands of our production-oriented environment mean
that time on this is severely limited.
If it were "released", then we could begin to make that commitment, and I
(and those in similar positions) could contribute considerably more time
and effort, knowing that there is that "covenant" between the developers
(Jason et al) and use (the enthusiastic, production-oriented, users).
I'm sure there are many of us who would love to start using Mj2, and
contributing bugfixes and new functionality. But its non-release status
acts as a major barrier.
> See, ppl on this list are stuck in old technology, and are basing opinions
> based on it ...
Wrong. We are stuck, by necessity, on released (or close to released)
software.
The Mj2 developers are doing an excellent job, and we'd like to help and
support them.
But it really is time to take the plunge. Let us, the end-users, see the
determination to release Mj2 . Establish a date (say within a month) for
a beta release, using current functionality. During this time, let the
efforts be on stability and portability of what is already there.
Then release it, officially, in beta. Show us how to feed in bugfixes
etc.
> Majordomo1 v1.94.1 was released in April of 1997 ... over three years
> ago. Any other work done on it since then, to the best of my knowledge,
> has been to patch security problems, and that's it ... so, those using Mj1
> right now are using effectively dead code, while the latest, greatest and
> what I'd consider most stable remains unused because the developers have
> too much they still want to add/to do it to go through the trouble of
> giving it a label that means something to the bean counters, and not so
> much to the techies ... beta or release ...
An unfair comparison. Yes, there are bean counters (I'm not one). Yes,
there are "techies" (I'm much closer to being in that group). But many
of us techies (the majority?) work in production environments where
we are accountable for the decisions (Mj2 or not Mj2?) that we make.
These labels are more more than just labels. A conscientious techie
recognises the commitment involved in applying such labels. And this is
probably why Jason has, in conscience and rightly, not wished to apply
such a label yet to Mj2.
The time has surely come to take that decision and make the commitment: to
apply such a label, with all its real meaning and implications.
--
: David Lee I.T. Service :
: Systems Programmer Computer Centre :
: University of Durham :
: http://www.dur.ac.uk/~dcl0tdl South Road :
: Durham :
: Phone: +44 191 374 2882 U.K. :
References:
|
|
