Text 10535, 45 rader
Skriven 2018-12-09 19:57:32 av Kees van Eeten (2:280/5003.4)
  Kommentar till text 10530 av Markus Reschke (2:240/1661)
Ärende: What would YOU do?
==========================
Hello Markus!

09 Dec 18 19:04, you wrote to Carol Shenkenberger:

 MR> The FTSC documented one:
 MR> - if there's an address in the IBN flag use that one and only that one
 MR> - if the IBN flag doesn't include an address take all the addresses from
 MR> all INA flags

 MR> The undocumented standard:
 MR> - take all addresses in IBN and INA flags

 Putting the hostname in the system field was a kludge needed at the time.
 With the introduction of hostnames in the IBN: an INA: flags, putting a
 hostname in the system field should be discouraged.

 Undoubtedly there is still software around that will for a hostname in
 the system field. The system field should be reverted to its original
 purpose, even if only for folklore.

 Whatever the docoments say. My understanding of where hostnames are listed,
 is, when only one protocol is supported, list it with the protocol flag.
 When more protocols are listed and supported on the same host, list the
 hostname with the INA flag. If hostnames are different for the protocols
 list the hostnames with the protocols. Listing a protocol port with with
 INA makes no sense, as that would mean that multiple protocols use
 the same port. Off standard port numbers should be listed with their
 protocol flags.

 It makes sense that software that reads the nodelist, cares for the misuse
 that some software and peolple have introduced. This misuse may be docomented,
 but should not be included in a standard.

 Just my two cents.

 Defining some program that has been built long ago, as how a standard
 should be set, should be out of the question.

 And yes, I do not use any of the programs laying around to create my own
 binkd nodelist.

Kees

--- GoldED+/LNX 1.1.5
 * Origin: As for me, all I know is that, I know nothing. (2:280/5003.4)