Eric - I like the idea of regularly scheduled digests with a cap on the
size of the digest which would force a digest out before the time elapses.
Kicking out a digest at that point would have the benefit of spreading the
network load out and keeping people informed sooner of a high traffic
discussion (as opposed to waiting the entire time period then chopping up
the digest).  Thanks for adding the digest capability...
 
   Would the addition of a "cap" change the scheduling of the regular
digest in any way?  For example, if I have a weekly list that kicks off
every Sunday but hits its limit on Tuesday, would the next digest still
be scheduled for the next Sunday?  This may depend on implementation details
but I think I like that approach.  It keeps a regular schedule "framework"
with additional digests interspersed as needed due to the load.
 
   Finally, if some list editors are interested, I wrote a very
simple-minded REXX EXEC which creates digests from note or Rice Mail
NOTEBOOKS (it would probably work from archives too but I don't think
I've tried that).  For example, I save list searches sent to our
new-list list in a notebook and create a digest every few weeks.
Since the DIGEST EXEC just creates a file you can edit the digest as
you wish before sending it (in fact I've wrapped a local exec around
the process to generate mail to the list with the digest imbedded).
 
   While my DIGEST EXEC isn't totally RFC1153 compliant, it should
allow any reasonable "bursters" to pick it appart ok (it worked with
the RFC1153 author's software anyway).  And it provides a starting
point to you could make it more compliant if you wish.  I'm sure there
are other programs that do this same thing (I know someone was working
on a CMS PIPEs version).
 
   If you'd like a copy of "DIGEST EXEC" you can get it from LISTSERV
on NDSUVM1 (eg. tell listserv at ndsuvm1 get digest exec ) or via
anonymous FTP to vm1.nodak.edu after  CD LOCSOFT   then  GET DIGEST.EXEC .
 
     Marty