Re: question about the new port

[Mark Crispin <MRC@xxxxxxxxxxxxxxxxxx>]

Bob -

     Guess what?  "Try the port" is the only thing that's any good anyway.
WKS records are utterly useless, at least in this country.

     I think I said it before, but I'll say it again:
. 8 bit SMTP is not sufficient to solve the 8-bit e-mail interconnectivity
  problem.
. 8 bit SMTP is not necessary to solve the 8-bit e-mail interconnectivity
  problem.
. Whatever you do, you have to deal with feeding 8-bit e-mail through a 7-bit
  pipe and you have to make it work.

     Some people, in spite of the above, want 8-bit SMTP.  OK.  We have at
long last determined a way by which we will allow it; to wit, a different
service on a different port which exchanges 8-bit mail.  SMTP is, and will
always be, 7-bit.  Sad but true.  Now, it may be that this new service looks a
lot like SMTP but is 8 bits instead of 7 bits.  That's fine.  It's whatever
the people who use it want it to be.

     I will support any reasonable proposal for an 8-bit MTA protocol under
TCP.  Changing what listens on port 25 is NOT reasonable and as such is not
subject to any further discussion.

@begin(flame)
     I will not offer serious consideration to any 8-bit SMTP on port 25
proposals except from an organization that is willing and has the resources to
convert every single piece of 7-bit mail software on ever single machine on
the network (including all those MX sites) at no operational or resource
expense to *any* site -- and has the resources to undertake such an effort.
This conversion includes converting all non-mail tools such as editors,
spoolers, etc. which may also operate on mailbox files.  This also includes
supplying more disk drives (and paying extra maintenance and electric costs)
for any systems such as PDP-10's that pack 7-bit data denser than 8-bit data
and thus would require more disk capacity to convert to 8-bits.  No sources?
Well then, disassemble and re-engineer them.
@end(flame)

-- Mark --