>  I have no trouble following the volume on
> > ruby-talk, and can (still) easily pick what threads to follow or
> > ignore.
>
> Congratulations.

???

> I can't/won't... Many others can't. That's why I
> suggested we start announce@ and that's why sean put it on his server
> and that's why out of ALL of the people subscribed to this list we have
> less that 10 (guestimate) arguing against it.

And how many arguing for it?  Five? 20? Out of ALL of the people subscribed
to this list? I don't see how an informal poll determines the long-term
validity of any idea.  Many people do not publicly comment because they
believe their views have already been expressed by somebody else, and aren't
aware that somebody has decided a vote is in progress.

>
> > I have an interest in, for example, database programming, but I don't
> > want to
> > subscribe to a list solely on that topic, as database programming only
> > occupies a relatively small and sporadic amount of my time.
>
> Who said you were limited to only one list? Subscribe to more than one,
> including -talk, and filter them to the same mailbox or whatever you
> want.

Thank you for that insight.  My point was that, with a list focused on (as
an example) database development, one would need to subscribe to the list to
partake in any brief discussion about databases.  As it is now, ruby-talk
allows for broad-topic discussions on databases, threading, distributed
applications, etc.  With potential fragmentation, discussions covering more
than one area often might require running for one list to another, repeating
the details at each step.

Now, maybe that's just how it has to be, maybe we've reached the point where
multi-topic discussions are impractical on ruby-talk.  Maybe it's worth
discussing before spewing lists.

>
> How detrimental is multiple lists to the ruby community (which we
> already have!)? It hasn't seemed to hurt FreeBSD, or apache, or linux,
> or postgresql, or mysql, or... you get the picture. For the most part,
> those are all thriving online communities.

For the most part?  It is insufficient to point to another community and
count the number of lists. We should also consider the number of lists
relative to the number of list participants and the size of the community,
as well as the topics and organization of the lists, and see what (if any)
bearing these have on the growth and development of the community.

Here are my concerns:

  That an announcement list will distract from the use of the RAA as the
primary means of recording information about new and altered libraries.

  That a precedent will be set for any disgruntled person or group to start
whatever "official" lists they happen to deem useful at the time, risking
Balkanization of the community because of a lack of discussion and
consensus.  I'm excepting lists devoted to independent classes and libraries
(e.g. ruby-tmpl), which should be managed by the owner of the library, and
referring to broad-topic lists (e.g. ruby-databases, ruby-announce).

If there is a real need for additional Ruby lists (and it appears there is,
though the specifics are vague), then the details should be thought-out and
publicly discussed prior to creation, and the lists should be organized
under @ruby-lang.org.


James

P.S.

Where can I go to get a *complete* list of all ruby-related mailing lists?