From: Charles Lindsey (chl@clerew.man.ac.uk)
Date: Mon Dec 15 2003 - 07:56:36 CST
In <200312140129.hBE1TqXR027056@jefferson.patriot.net> "Shmuel (Seymour J.) Metz" <Shmuel+gen@patriot.net> writes:
>In <200312122254.hBCMsGT05223@clerew.man.ac.uk>, on 12/12/2003
> at 10:54 PM, Charles Lindsey <chl@clerew.man.ac.uk> said:
>>These will rarely occur, but users need to be aware of them when they
>>do.
>Is there any reason to not recommend that a client make all header
>fields available to the user on request, even if they are not
>displayed by default?
Yes, I have made a note to say that in 3.3.1. Oddly, GNKSA seems silent on
that.
>>> c) Reply by e-mail, with "Subject: " and "To: " headers derived
>>> appropriately from the original article. (see #5 and #8 below)
>Does it say anything about recognizing a .invalid TLD?
Yes, USEAGE cover that fully. GNKSA says nothing (but GNKSA was written in
days before people started munging to any extent).
>>USEAGE 3.1 currently has SHOULD for creating _any_ header.
>I would be opposed to weakening that.
I agree.
>>That MUST NOT seems a bit strong ...
>If it were in USEFOR I would agree. In USAGE I believe that it is
>appropriate. Certainly mailing unwanted copies of Usenet posts is
>disruptive.
Yes, but it was the bit that said the user MUST NOT even be allowed to
re-configure it to post and mail that I thought was OTT.
What USEAGE currently says (3.2.1.3) is that the user SHOULD be warned and
asked to confirm if he tries to do it (which is what USEFOR used to say
before we split it). You want that upgraded to MUST?
>I'd like to see some encouragement for offering the user the ability
>to select disjoint blocks of text to quote; one block initially and
>additional blocks as he composes his reply.
OK, if we agree to add the section 3.5 on user interface issues, that
could go in it.
>>> If feasible, the software SHOULD try to guarantee that this address
>>> actually belongs to the person using the software, and actually accepts
>>> e-mail.
>>But this is much more dubious in these days of munged addresses.
>With a proper reference to the .invalid TLD it is no longer dubious.
Agreed. I think what USEAGE says is about right (and we did argue long
enough over it). As I said above, GNKSA was written before it was
fashionable.
But what I would really like to hear from people is to what extent we
should try to be consistent with GNKSA. This comes in two parts:
1. Where we are waeker than GNKSA, do we upgrade (usually means changing
SHOULDs to MUSTs).
2. On issues where we are silent, should we add text to match?
I think the answer to 1 is pretty clear. In most cases we should upgrade
to match (though I am doubtful about a few of them).
2 is not so clear. Some of them surely Yes. Others may need more
discussion.
But I need to hear views on both groups.
I sent a copy of this to the author of GNKSA and invited him to join us
here to discuss it. But I have had no response as yet.
-- Charles H. Lindsey ---------At Home, doing my own thing------------------------ Tel: +44 161 436 6131 Fax: +44 161 436 6133 Web: http://www.cs.man.ac.uk/~chl Email: chl@clerew.man.ac.uk Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K. PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5