another question to ponder

Sat Jan 24 08:28:05 EST 2004

Executive summary
--------- -------
A lot of people said a lot of useful, correct, and interesting stuff
about TEI and lists. But IMHO this is a Board, not a Council, issue,
and should be removed from our agenda.

Unless anyone objects, I plan to forward this mail-piece to the Board
next week.

Summary
-------
Two issues: TEI-TECH, aka one vs. two lists; and list consolidation.

On TEI-TECH -- I'm on the fence, and would be happy to implement
   whatever the Board decides. Lou's suggestion to create a new list
   for beginners, perhaps under the auspices of a SIG, is not
   unreasonable.

On consolidation -- yup, we should. I can't even discuss it further
   until Feb.

<p>Details
-------
cw> BTW, what happened to TEI-TECH? Would it be useful to revive that
cw> as a general list for technical discussions as opposed to TEI-L.

pw> I would just point out that we had a TEI-TECH list, but it was
pw> eventually withered away. Not sure why--whether there weren't
pw> enough subscribers to make it viable, or that it was difficult to
pw> define the difference between the two lists well enough.

TEI-TECH still exists, with a mere 37 subscribers. Subscribers
include me, Julia, Sebastian, and Laurent; but do not include
Alejandro, David B., Lou, Matthew, David D., Susan, Natasha, Edward,
Perry, nor Christian.

I just scanned quickly, and it appears that the last TEI-related
posting, as opposed to spam or meta-postings (i.e., discussions of
the list itself) was a query about lang= by Lou Burnard posted in May
of 2001, to which the response was a deafening silence. (For which I
feel a bit guilty, I must admit.) Note that this post was a reply to
a post of Merrilee's on TEI-MMSS, so even it may have been
mis-posted.

cw> I am a bit worried that some readers of TEI-L might be scared
cw> away.

That was the logic, and indeed, I think it is sound, although perhaps
not convincing. My (anecdotal) recollection is that we do see a
couple of subscribers signing off the list most every time there's a
fast-and-furious thread with more acronyms than explanations. On the
other hand, we see far more people abandoning ship when there's spam
spewing about -- 6 people have left in the past 3 days, not including
Micheal Beddow who signed back on with a different e-mail address.

pw> If I'm remembering correctly, posts to TEI-TECH were generally
pw> also cross-posted to TEI-L, so it only doubled the volume without
pw> solving the problem.

Exactly correct, which is why the aforementioned logic although
sound, may be unconvincing. I think two lists makes better sense, but
in practical terms Lou is correct:

lb> Having two lists was just a nuisance -- what's too technical for
lb> one person is too fluffy for another.

But Sebastian may well be correct about *why* it failed:

r> because there was no marketing or explanation of it?

cw> I also see this as a vital point for the way TEI is perceived
cw> from the outside (or even from the edges). Management of the
cw> lists is a mess and done purely on an ad-hoc basis, with at least
cw> three different providers involved.

Hit the nail right on the head. I know of at least four.

sr> Sorry to bang on about this, but if the TEI Consortium ... is to
sr> have a future, it has to be much more proactive about its image,
sr> its communications, its services etc. A sine qua non for that is
sr> a single home for mailing lists, with centralized subscription
sr> management, common searchable archives, and a rational set of
sr> lists for different aspects of the TEI.

I very much think Sebastian is on the right track here, although the
particular features he lists may be a bit pie-in-the-sky (or I may be
misunderstanding the features). SR, do you know of a particular list
server software package that will do what you want?

cw> ... I see the need to streamline this ...

I believe it is technically possible to collect all the lists at
Brown, provide a single address space "lists.tei-c.org", keep the
same folks who are list owners as list owners, and thus provide the
same interface (both web and e-mail) to subscriptions, subscription
archives, archives, management, etc. I do not think it is possible to
provide a single search interface to multiple lists. Moreover, I am
not at all sure this is politically or financially feasible.
Furthermore, the folks at Brown who might deal with this are
currently frantically working on a major project they hope to unveil
on Mon 02 Feb, so there is *no* chance of getting any information out
of them until well after that.

cw> I will put this on the agenda for the next call, but would like
cw> to have a proposal we can discuss by then. I assume also that
cw> this is rather an issue the board should take up, but I am sure
cw> the Board will consider a statement from the Council if we have
cw> one.

Seems to me this issue falls clearly outside the scope of the council
(ala section VI.2 of the By-laws). While I am sure Christian is
correct, that the Board would consider a statement from the Council,
this is a thorny issue that has the potential to both take up time at
Council meetings and be divisive. Given that (as Julia has stated)
the Board is already considering this issue, and that I can guarantee
not to be able to find out Brown's possible commitment before our
call, I'd be inclined to drop the item from our agenda. If others
think the reverse -- we've got momentum on this issue now, may as
well talk about it -- I'm happy to go along.

ss> I agree totally! Sometimes there is just so much going on the TEI
ss> list, that I can't get to it. I'd also love to be able to delete
ss> items, and know I can get back to them. Humanist has a fabulous
ss> archive interface.

?? 
TEI-L uses LSoft's LISTSERV software. Several list-geeks have
referred to its search and search interface as the best in the
business (that was several years ago, though; others may have
surpassed it since). The web interface (see
http://listserv.brown.edu/tei-l.html) permits browsing by author,
date, or topic (although I have to admit, those stupid icons at the
top are nearly undecipherable -- why couldn't the buttons just say
"author", "date", "topic"?), and searching based on any combination
of content, subject, author, and date.

I just tried searching HUMANIST. Although the search capabilities are
not quite as impressive as LISTSERV, it is plenty adequate, and the
Excite engine results page has a potentially useful "confidence"
rating. However, it seems that HUMANIST is digested in a manner that
loses message ID and thus threading information. Thus sorting by
thread is pretty much useless.

So in short, I think the TEI-L archive interface is significantly
better than the HUMANIST archive interface (although the latter does
have some advantages), and thus I'm guessing that you (ss) simply
didn't know about them. If that's the case, besides posting a
reminder to TEI-L itself, is there anything you think I (we) should
do to provide "marketing or explanation of it"?