>: for the following  reason: "Sender:", "From:" or "Reply-To:"  field pointing
>to
>: the list has been found in mail body.
 
>and the only field that could be construed as 'pointing to the list'
>is "To:" which was quoted.  Is it the error text which is wrong, or
>is the logic which caused it to bounce in error?
>(I believe this is version 1.8b.)
 
     Neither: it is the user that is "wrong" based upon the
     collective net ;-) experience of 10 years (Happy Birthday
     L-Soft) of running LISTSERV on Bitnet and Internet where
     e-mail loops used to be the norm for mailing list managers
     that ignored LISTSERV's logic.
 
>A second question would be "why is this going to the list owner
>instead of the sender?"  It is the sender who has control of the
 
     Because it is a loop error that LISTSERV is detecting, not
     some some user error.
 
>content, and this bounce requires the list owner to write back to
>the sender and explain things, or else, puzzled by the non-appearance
>of his message, the sender will just keep doing it.  "My" list
>is only 70-75 people, but this could be a major headache for large
>lists, I would think.
 
     That's why I send them a canned (boilerplate) response.
     Most people will (eventually) learn how to interact
     proplerly with mailing lists.
 
 
>Third:  I don't see what the problem is with included headers pointing
>back to the list might be; it's *just slightly* problematic if at the
>beginning of a line.  If the concern is something autoreplying and
>quoting everything *including* headers, then why is the ">" quote
>OK but not the "> " quote?
 
     Not being a LISTSERV developer, I can't say for sure but my
     guess is that the problematic behaviour (loops) was more
     often than not detected with the one style rather than the
     other style of quoting/forwarding.
 
>Yes, I've read 9605-UD-02; it says "what" (i.e., this will happen,
>unless the quoted headers are quoted by ">" instead of "> ") but
>it doesn't say *why* and the logic completely escapes me.
 
>The text in 9605-UD-02 (11.2 at the end) which says  "The ultimate
>solution to the problem is to warn subscribers..." seems to me to
>be a copout for not making LISTSERV a bit more intelligent.
 
>OK, your turn... convince me I'm wrong.
 
I'm sure I haven't convinced you, but *I'm* sure glad that
LISTSERV is doing the "right thing" -- I seldom have to deal
with loops any more, though I do have to deal with the
occasional "error" generated by the loop-detection.  Proper
netiquette is its own reward: less CPU time, less archive
space, less bandwidth, less time downloading/reading.
 
Here is the boilerplate that I use to inform the user how to
interact with the list:
 
Dear
 
If you want this distributed to all of xxxxxx-L, you will need to
remove the original "headers" that you included in your reply: the
software does not allow you to include them, thinking that it is
a "loop."
 
Additionally, please remove all other original signatures and text
NOT essential for context -- sometimes all you will need is a
relevant SUBJECT: RE: line in your reply.
 
Otherwise, simply forward to your intended correspondent.
 
Thanks.
 
--  co-owner:   INFOSYS, TQM-L, CPARK-L, ERAPPA-L, JANITORS, LDBASE-L, et -L
URL:mailto:[log in to unmask]  "Don't drink n' drive on the info. super hi-way"