L. Detweiler <[log in to unmask]>, maintainer of the
of Cryonics FAQ set, complained last month when his FAQ was bounced
back to him from the SCIFAQ-L mailing list.  As I tried to explain to
him then, the problem is not a simple one, nor is the solution easy.
In particular, it is not a problem of installation or management of
the mailing list or the LISTSERV software, but a deeper issue of mail
header format standards.
 
L. Detweiler has posted his (unchanged) FAQ again, and again received
bounced copies from the LISTSERV machine that administers SCIFAQ-L.
His complaints (numbered) are valid and justified (his tone is not):
 
(1) Automatic error messages should include the address of a real person.
 
At this time I have limited ability to modify the set of automatic
messages generated by LISTSERV.  As this is now being changed in
the LISTSERV code, I must wait a few months before I can fix it.
 
 
(2) Bounces from a mailing list should go to the list owner.
 
An important feature of the LISTSERV software (now 1.7f) is an algorithm
that checks for duplicate posts.  Duplicate posts are returned to the
sender.  As owner of a mailing list that uses the LISTSERV software, my
options are to disable the checking feature, or make the list moderated.
I do not want to disable the checking feature, because this is the main
way that "loops" are prevented from propagating through the mailing list.
I am not able to read my e-mail every day, so it is important that I keep
this feature in place.
 
Note, however, that because the loopcheck algorithm sends duplicate
articles back to the sender, it doesn't prevent loops from developing
between LISTSERV and the offending account!
 
Ideally, I think duplicate submissions should be sent to the list owner,
who could then deal directly and in a context-sensitive way with the
original sender.  There has been a fair amount of discussion in LSTSRV-L,
a mailing list about LISTSERV administration, regarding this issue.
Nothing has been decided.  If duplicate submissions were redirected to
the SCIFAQ-L owner (me), I would probably not re-send them to the list,
but I would want the option, as list owner, of doing so without having
to modify the text or disable the loopcheck feature.
 
Also, I wish the loopcheck feature worked within a specified timeframe.
This would simplify the problem of periodic articles that do not change,
and as both FAQs and mailing lists gated to Usenet become more common,
this will be an important feature.  Although SCIFAQ-L articles are not
archived by the LISTSERV, the loopcheck feature still works.  I assume
this is accomplished by saving the checksum of every distributed article.
Would it be difficult to save the date too?
 
 
Here is a possible temporary solution that might satisfy all parties:
 
Because SCIFAQ-L is gated to sci.answers for the purpose of distributing
FAQs, it would not be convenient or appropriate for me to make the list
moderated, unless I could make the gateway itself a "moderator".  It
should be possible to have all errors sent to me, although I could not
guarantee that the Usenet gateway would not also receive some e-mail.
 
Comments?  Would this work, given the existing LISTSERV program?
 
 
L. Detweiler further asks:
 
>p.s. do you know of a general purpose Usenet to mail server that people
>can subscribe to? tx.
 
If the e-mail gating of Usenet newsgroups were "simple", then such a
facility would no doubt already exist.  Many Usenet newsgroups *accept*
postings via e-mail, but do not *distribute* them, for this very reason.
 
I think it is a wonderful thing to be involved in working out how to
make Usenet and mailing lists fully interconnected.  This is a small
part of history in the making, and what we decide now will affect how
the Internet looks to millions of people for some (or many!) years to
come.  We are slowly creating a generalized gateway system.
 
 
 
        Una Smith       Department of Biology           [log in to unmask]
                        Yale University
                        New Haven, CT  06511 USA