ietf-nntp new draft of base document available

[Clive D.W. Feather]

Stan O. Barber said:
>> If the editor isn't taking part in them, they aren't discussions in a
>> useful sense of the term.
> 
> I using information from the list as a basis for changes in the document. The
> fact that you have already noted that changes have been made confirm that.
> 
> You have also noted that some changes you have suggested didn't make it in. In
> every case where I chose not to include something, I was making the best
> judgement I could based on all the information I had available. If you disagree
> with a specific decision, then please bring it up on the list.

The problem is that you haven't *said* that you've made a change or given a
reason for rejecting it. On other lists I get responses like "done" or "I
don't want to make that change because ...". This list feels like a black
hole for comments.

>> I'm still waiting for answers to basic questions before I can write
>> proposed text. Like "what the hell is the point of the MODE READER command
>> *in terms of the protocol* ?".
> 
> Here is what is in the text:
> 
>             MODE READER MAY be used by the client to indicate to the
>             server that it is a news reading client.

But since nowhere else (and, in particular, section 3) talks about any
other kind of client, aren't *all* clients "news reading clients" ?
[Remember, I'm looking at this as a black-box specification being read by
someone who's never written a client or server before.]

>             This command may be
>             entered at any time. The server MUST present a greeting code
>             (as described in section 7.1.2.1) appropriate to the server's
>             ability to provide service to this client in this mode.
> 
> So, the command is basically a no-op unless the server chooses to make use of
> the
> information to better optimize its behavior to provide service to a news reader
> client.
> 
> Are you suggesting an explanation of that nature be included in the text? Is
> that an 
> implementation detail?

Well, I can see three possibilities:

(1) MODE READER is effectively a no-op (particularly at the start of a
newsreading session; I note your other email about resetting things). In
that case, the first sentence needs to go.

(2) MODE READER or some implementation-specific alternative (such as
MODE NEWSPEER) is required before many common servers will provide service.
In this case we should be documenting this fact.

(3) MODE READER is recommended if many ARTICLE/HEAD/BODY commands are going
to be issued, because many servers will optimise themselves accordingly.
In which case we should say so.

The hypothetical new client author currently has *no* idea why this command
exists and when - if ever - it should or shouldn't be issued. If someone
will tell me what the damn thing actually does, I'll rewrite the text for
you.

-- 
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 1037
Demon Internet      | WWW: http://www.davros.org | DFax: +44 20 8371 4037
Thus plc            |                            | Mobile: +44 7973 377646