on 9/29/00 SRE <eckert@climber.org> wrote:
>
> At 01:10 AM 9/29/00, Mills CP wrote:
>> Why not release mj2 as it is now, sure, it may have an extensive wishlist,
>> but why not have that ready for v2.5 of something.
>
> I've already asked for a feature set freeze and release, haven't heard
> an answer. I'm behind on writing the docs for the stuff that got done
> last week, but the number of configuration options *IS* getting to the
> point where you can do anything you want if only you can FIND the command.
Keep pushing for this. Every well run project has milestones, scope, feature
set lists and freeze dates. MJ2 needs all of the above to be successful.
>> As I have said above, I have recently been looking for an open source
>> product that I can help develop. I have quite a bit of experience (c. 5
>> years) developing perl and cgi based web applications
>> and some experience in a lot of other languages (c, c++, java, etc). Is Mj2
>> still being written in Perl? If so, is there a list of things which need
>> working on? Would you developers like some help?
>
> Yes, yes, and yes. For example, there's no way to search the archives.
> That's a bounded problem that no one is working on and which is arguably
> something that needs to be in place before a full release. For another
> example, I want a way to dump the ENTIRE database in the form of commands
> to create it from scratch. This already exists for subscriptions and
> personal option settings, but you can't dump the core database that
> contains user passwords and full name with comment fields (the registry),
> nor can you dump the aliases (for people posting from multiple addresses).
Help with MJ2 would come easier of the MJ2 information on the web were more
up to date. Jason needs to give the management of the webpage - maybe the
management of the project itself - to someone else. He admits to not wanting
to do that stuff, so it does not get done. It's hard to get excited about
MJ2 any more. There has not been much information forthcoming in months.
Sifting through the posts in the -dev list is painful unless you are
tweaking code.
>> I sure am willing to offer it. I do not have a lot of time to spare, but I
>> would like to get involved if thats possible?
>
> Need more? Review the archives from mj2-dev and you'll see all sorts
> of stuff I've requested. Most of the big stuff has been done, but
> as you've already read there are really only two code contributors
> and their time is limited.
The archives that stopped being updated in November of 1998? I hope these
are not the latest archives. That was nearly two years ago.
> Hey, if anyone is worried that they can't work on object oriented Perl
> (and that includes ME), I can always use help on the help docs. I've written
> a Perl hack that automates some of the help file maintenance, but we
> really need an index/table of contents to help people find the feature
> they need. Having a newbie review the docs might pick up a bunch of
> wording assumptions and/or gibberish that needs to be rewritten.
>
> SRE
>
> mailto:eckert@climber.org | http://www.climber.org/eckert/
> Info on peak climbing email lists mailto:info@climber.org
>
> Amateurs built the ark. Professionals built the Titanic.
I have to second the concerns raised by others in this thread. Majordomo 1.x
has been a great mailing list manager. It has been extended by many
(including myself locally) to do what we needed it to do, and now it's time
for MJ2 to replace the original version.
MJ2 is an open source project. Mailman is an open source project.
MJ2 is available in pre-release "snapshots" and you can run it at your own
risk. The MJ2 community is currently seen as small and not growing.
Mailman is available in a "public beta" and the community is seen as
growing. Mailman is being developed and supported by volunteers.
The need to release a version on MJ2 that is lacking in some features is
here now. The features that are part of the package need to be solid,
documented, and ready for prime time. It's OK to have a list - even if it's
a large list - of features that are still under development and not yet
ready for prime time.
If MJ2 appears to be vibrant, healthy, open-source software, it will
generate interest, developers will tweak and contribute, someone will write
that database access routine, or the archive search function. Open-source
software works because lots of people contribute to it. MJ2 appears to be
stalling because there are too few authors, too few users, and it's
currently poorly positioned.
MJ2, if it only replaced MJ1.x, would become the most widely deployed
mailing list system overnight. If it takes another year for MJ2 to be
released, it will be an also-ran.
-Tim
--
Tim Hewitt
Information Systems Manager
Fairchild Semiconductor
|
|
