ietf-nntp HDR

Clive D.W. Feather clive at demon.net
Mon Jan 7 01:48:50 PST 2002


greg andruk said:
>>> Yes. I support the suggestion of ":Lines" and ":Bytes".
>> I don't, at least yet.  I think that syntax is bizarre and completely
>> unintuitive.  It feels very ill-thought-through to me.
> I would rather see these things that aren't really headers put into a
> separate verb (ARTINFO or something like that).

That's an option, but it's a major departure from the existing situation
(which mixes metadata and header data in the overview database).

> I wouldn't want to see us waste time trying to define what these other
> values would be right now, just leave this space for future extensions
> to play in.

I agree that we want a separate metadata space.

My thought is that, since existing practice uses the same commands to
access header data and metadata, let's see if it's possible to continue in
that way but still have something cleaner. In turn, that requires a way to
distinguish the two.

Now, since 2822 allows a header name to include *any* ASCII character
except space and colon, and since we don't want to rely on UTF-8 for
primary syntax, the only way to separate out the two namespaces is to use
colons in some way. The three obvious cases are:
(1) no colon
(2) leading colon
(3) trailing colon
It would be nice to use (1) for metadata and (3) for headers, but
unfortunately existing practice is to use (1) for headers. Using (3) for
metadata would be perverse, which leaves only (2).

So, for those objecting to my views:

(A) Should OVER/HDR be allowed to return metadata ?

(B) If no, how do we address existing practice for OVER and the Lines
situation for HDR ?

(c) If yes, how do we distinguish headers and metadata in HDR ?

-- 
Clive D.W. Feather  | Work:  <clive at demon.net>   | Tel:  +44 20 8371 1138
Internet Expert     | Home:  <clive at davros.org>  | Fax:  +44 20 8371 4037
Demon Internet      | WWW: http://www.davros.org | Mobile: +44 7973 377646
Thus plc            |                            |



More information about the ietf-nntp mailing list