Text 10248, 82 rader
Skriven 2018-11-11 02:02:02 av Alexey Vissarionov (2:5020/545)
  Kommentar till text 10223 av andrew clarke (3:633/267)
Ärende: FTS-1 and FTS-4
=======================
Good ${greeting_time}, andrew!

10 Nov 2018 00:21:32, you wrote to All:

 ac> Policy4 refers only to FTS-1 with regards to ZMH.

Yes. That's why new revision should be named "Technical requirements for a
Fidonet node" or somewhat like that.

 ac> Expanding on Andrew's suggestion, I propose that:
 ac> 1. FTS-0001.017 should be issued, superceding FTS-0001.016.

Yes - it would be just a next edition. Fully rewritten, but next.

 ac> 2. The newly issued FTS-0001.017 is used solely as a reference
 ac> document that only points to other documents, for ease of
 ac> maintenance.

Not only references, but the requirements like "dialup nodes must support EMSI
protocol as defined in FSC-0056", "IP nodes must support binkp protocol as
defined in FTS-1026", "email-based nodes must reply to the PING requests as
defined in section 5.10 of FTS-5001 in at most 3 days (72 hours)", "the message
editor should support multi-level quoting as defined in FSC-0032" etc.

 ac> 3. FTS-0001.016 should be renamed FSC-0001.016

FRL-0001.016 (it once was FSC-0001.016 in the past, according to itself).

 ac> to be used as an historical reference, and is otherwise left
 ac> completely untouched. This completely bypasses potential issues with
 ac> the document's copyright.

Yes - just add a standard header before the original text ("This document is a
part of Fidonet Reference Library (FRL) and preserves the text of FTS-0001.016
obsoleted by newer publications").

 ac> 4. We prioritise documenting the required (non-optional) parts of a
 ac> compatible node, especially Internet nodes, as they are the obvious
 ac> majority of FidoNet in 2018.

That's quite simple: just require accepting uncompressed (this word is much
better than "unpacked") PKT2+ bundles over binkp.

 ac> 5. Ensure these parts are documented separately, again for ease of
 ac> maintenance, and refer to them in the new FTS-1.

Yes: this would require issuing several FSPs and FTSes.

 ac> 6. (optional, but probably sensible) Have FTS-0001.017 refer to (and
 ac> apply to) FSC-0001.016 in the case of remaining dialup nodes.

Only if we'd find enough FTS-0001.016 dialup nodes not supporting EMSI. And I
seriously doubt there is even a single such node left alive.

Also, as there always will be some idiots who would try to obstruct this by
setting up such nodes, these nodes _must_ be searched among listed at 01 Nov
2018 (well enough to avoid playing with timezones or whatever else).

 ac> Effectively this would mean the new FTS-1 would apply only to
 ac> Internet nodes.

No: it _must_ apply to any and all nodes regardless of connection type.

 ac> This could obviously change in future as the need arises, with
 ac> further issues of FTS-1.

Of course.

 ac> 7. Re-document the optional parts (eg. stored message format) as time
 ac> and interest permits. In the interim simply refer to the appropriate
 ac> section in FSC-0001.016.

No: the FTS-0001.016 _must_ (in the sence of FTA-1006) be buried in oblivion.


--
Alexey V. Vissarionov aka Gremlin from Kremlin
gremlin.ru!gremlin; +vii-cmiii-ccxxix-lxxix-xlii

... god@universe:~ # cvs up && make world
--- /bin/vi
 * Origin: http://openwall.com/Owl (2:5020/545)