Roger (and others),
 
  You mention  receiving messages saying  that "0  files were distributed  to 0
recipients". I think this  'bug' has been fixed a long time ago  -- if it still
happens, please tell me which server  sent this message. Since version 1.4 this
message is suppressed if 0 file is sent out.
 
  I agree that  receiving acks can be a pain.  However, FRECP11-LISTSERV allows
you to define the amount of ack info you  want to get on a list per list basis.
Either FILE ack with  short stats summary (when you live in  FRECP11 this is an
absolute necessity), MESSAGE  ack with short stats summary, or  "no ack". Until
now I have equated "no ack" to a single message, "your mail has been successful
ly processed  for list  xxxxx". Which  is to  say, half  the messages  that the
BITNIC version would send you. This message was here only for test purpose, and
will be removed in  the next version. Then "noack" will  really mean nothing at
all.
 
  Additionally, the list  owner has the ability to define  the DEFAULT value of
this option,  to be used  for people who  did not specifically  request another
option. This is the "Ack=" keyword in the list header.
 
  This option  is automatically forwarded  to the  peer servers by  the 'entry'
server, the one  that receives the piece  of mail from you in  the first place.
There was  a typo in  version 1.4 (c/X-LSVOPTS'/X-LSVOPTS:'/ in  LSVXMAIL EXEC)
which caused this forwarding to be ignored, though.
 
  However,  mail from  a  non-FRECP11 server  does not  have  the "ack"  option
indication in the  header, and therefore causes the default  value to be taken.
The default for the default value is "file ack", for the reasons I have explai-
ned above.  Also, as I  said previously  on the UG-L  list, I wanted  people to
realize the  load they inflict  on the  network when they  mail to a  list like
PROFS-L. I myself  sent mail once to  PROFS-L and was utterly  surprised when I
saw how much load I had imposed in the network. I consequently decided to redu-
ce mailing to this list to a minimum.
 
  Finally I would  like to repeat that this acknowledgement  is sent INSTEAD of
sending a  copy of  the original mailfile  back to the  sender like  the BITNIC
version did. I would  not like to start a year-long  discussion on that matter:
FRECP11-LISTSERVs  interface  fully to  each  other.  They support  the  BITNIC
type servers for  obvious compatibility reasons, but to a  limited extent. I do
not want  to waste  any time  working on schemes  that would  allow BITNIC-type
users to access the FRECP11-type  functions that would be otherwise unavailable
to them. Suggestions such as "just allow  the user to specify which kind of ack
he wants by  a 'Comment: NOACK' tag"  will not be heeded,  feasible though they
might be.
 
  Eric
 
PS: on the 'operator receiving a reply' issue:
 
    1) There IS  already an IGNORE =  'u@n1 u@n2...' keyword in  LSV$PROF EXEC.
    2) You  should NOT  run with  WNG  IUCV, but  with  WNG ON  -- the  default
       profile shipped with the code runs with WNG ON.
    3) My personal opinion is that NO  network server should run with WNG IUCV,
       and that NO user  should run WNG OFF. In the place where  I was before I
       came to FRECP11, there was a rule that  any user who was MSG OFF and WNG
       OFF could be  CP FORCEd without warning,  even though he had  a phone in
       his office and  could be reached there. The operator  has more important
       things to do than to send dozens of phone calls to people who don't care
       about messages. But this is another story...