MW> Why would it be unreliable? It works perfect, and every messages that 
MW> does contain a correct MSGID/REPLY kludge is linked as they should be.

HS> No, you are depending on the crc which is fine.  That means that it

what CRC, hector??? nowhere in FTS-9 is CRC mentioned...

  * the second and third parts of the line are used in the linking *

ie: MSGID: 1:3634/12 12345678

the address and the serial number ARE SIGNIFICANT... that's why you can
have identicle serial numbers from different systems and not cross link
the messages... because the address is significant... no matter what
kind of address it is...

ie: MSGID: mlewis{at}wplace.cssnet.sanford.nc.us 42ab42ab

BOTH PARTS are significant... the address and the serial number...

HS> requires that every message whether is it a reply or a follow up has
HS> the original crc. Thats your lack of foresight to see there would be
HS> a problem in follow up situations.   But to use JAM and say
HS> "MSGID/REPLY IS NOT FOR ADDRESSING" is horse crap and no excuse for
HS> inproper reply/chain in jam. 

HS> Here I do have an solution:  For follow up, use the address of the
HS> original reply, and the crc of the original msgid.  That would
HS> atleast satisfy your chaining requirements.

see above... BOTH PARTS are SIGNIFICANT! the address and the serial
number...

HS> What you are TELLING me if that I shouldn't TAKE the address from
HS> the REPLY line and put that into the ADDRESS field because it is
HS> UNRELIABLE. 

HS> Well I'll be monkey's uncle... Mats perfers to live with
HS> imperfection. 

but we still have you, hec... >

)\/(ark