ietf-nntp Re: draft-ietf-nntpext-streaming-00.txt

[Clive D.W. Feather]

> A URL for this Internet-Draft is:
> http://www.ietf.org/internet-drafts/draft-ietf-nntpext-streaming-00.txt

Okay, a first pass through this.

We took the leading spaces out of LIST EXTENSIONS.

I would actually remove the whole of section 2 and replace it with the
formal stuff required by section 8 of the base document.

Do existing implementations actually return STREAMING to LIST EXTENSIONS?
If not, and this is an invention on your part, can we change the name to
something more specific, like "pipelineable posting commands" and
"PIPEPOST".

I'm not clear what's going on with MODE STREAM. I take it it's a command
like MODE READER that possibly changes internal state. In which case, I
suggest you copy the entirety of the MODE READER section and then edit it
to describe the actual situation. Questions that immediately come to mind
for me are:
- Does using MODE STREAM make certain commands unavailable?
- Does LIST EXTENSIONS report STREAMING when MODE STREAM before this
  command is used?
- Can I use CHECK without using MODE STREAM?

We've already discussed freeform text in responses.

Don't include generic responses, including 480, in the tables of responses.
[Note that MODE READER lists 400 and 502 because they can cause a
connection close in an unusual situation, and arguably 400 shouldn't be
listed because it always results in a connection close.]

You should make it clear that, while a 438 from CHECK means TAKETHIS is
likely to be a waste of effort, it's not illegal.

Make it clear that the message-id in the responses MUST be that in the
initial command.

Don't put that "[the entire article ...]" bit in the syntax, put it in
the description; look how I address the same issue for POST.

Can I suggest that we say a server MAY respond with 431 or 438 to a
TAKETHIS command, with 439 meaning "either of these but I'm not saying
which". That allows future servers to be more helpful, while existing
clients will see 43x and just treat it as an error. If this change isn't
made, make it clear that 439 means that CHECK might return 431 or 438, so
retrying might or might not be sensible.

Private question: are you hand-editing to match my style or using xml2rfc?
If the latter, do you want to use my tools as well?

-- 
Clive D.W. Feather  | Work:  <clive at demon.net>   | Tel:    +44 20 8495 6138
Internet Expert     | Home:  <clive at davros.org>  | *** NOTE CHANGE ***
Demon Internet      | WWW: http://www.davros.org | Fax:    +44 870 051 9937
Thus plc            |                            | Mobile: +44 7973 377646