[ietf-nntp] draft21.pre-2

[Ken Murchison]

Russ Allbery wrote:

> Clive D W Feather <clive at demon.net> writes:
> 
>>Ken Murchison said:
> 
> 
>>>Section 8.5.1.2 says:
>>>
>>>"Support for the first (message-id) form is optional. If it is not 
>>>supported, the generic response code 503 MUST be returned."
>>>
>>>I have no problem with this, but I wonder why we don't have a capability 
>>>to announce this as well.
>>
>>[...]
> 
> 
>>I'm happy to add this, but I would like to hear at least one or two more
>>voices on the topic.
> 
> 
> My feeling is that it's not worth it, but I definitely have no strong
> opinion and if other people do, I won't argue against it.

The argument for this is that we have gone to great lengths to allow the 
client to determine which optional parts of the protocol are supported 
by simply calling LIST EXTENSIONS.  Skipping this for OVER by msgid and 
forcing the the client to use trial and error seems like a step 
backwards (or at least not a step forwards).

Unless there is a good technical reason why we shouldn't/can't do this, 
I think we should add it just for completeness sake.

-- 
Kenneth Murchison     Oceana Matrix Ltd.
Software Engineer     21 Princeton Place
716-662-8973 x26      Orchard Park, NY 14127
--PGP Public Key--    http://www.oceana.com/~ken/ksm.pgp