From: Charles Lindsey (chl@clerew.man.ac.uk)
Date: Wed Dec 03 2003 - 04:54:28 CST
In <3FCD1289.9050403@durham.ac.uk> Nick Boalch <n.g.boalch@durham.ac.uk> writes:
>It looks good. The tone seems appropriate and the draft consistency reads with
>clarity.
At Last! Some Discussion!
>Would it be useful for readers to have, in addition to the clearly explained
>and annotated items currently presented in the draft, a summary in simple
>bullet-point form of the requirements, perhaps presented as a conclusion to
>the draft?
Yes, an appendix might be possible. The GNKSA has a short summary in it as
well as the main texts. However, first priority is to decide what issues
we want to include/not include. Summaries can come later.
>On individual points:
>Section 3.1.1.2 on address munging suggests user-agents make it difficult to
>munge addresses....
> ... the practice. However, it shows how to do it "correctly"
> and it is NOT recommended that implementors facilitate other
> means of disguising addresses.
I think this stems from discussions some years back. We were taking the
line that munging was a Bad Thing, though possibly necessary in the
current climate, and we looked forward to the day when it would no longer
be necessary (well, perhaps now that the US Congress has decided exactly
what Spam is, that day may have come :-( ). Anyway, that was the origin of
the wording which suggested that if anyone was going to munge, then he
would have to do it on his own without the benefit of any "munge button"
in his browser. OK, let us hear views on what we should say here. It is
only a NOTE, but it does stick its neck out a bit at the moment.
>An internal note in section 3.1.1.3 refers to a draft on Message-ID
>construction prepared early on in the project. Are you referring to
>'Recommendations for generating Message-IDs' (Matt Curtin and Jwz) or
>'Guidelines for the Generation of Message IDs and Similar Unique Identifiers'
>(Claus Andre Faerber) -- I'm assuming the latter.
No, I was thinking of the former. Can you give me a pointer to the latter?
And do people think this WG should be promoting an Informational document
on those lines (no, NOT NOW! Later when these present drafts are out of
the way)?
>Section 3.1.1.4 - I don't see any real reason why the MUST NOT needs to be
>downgraded. Are there hordes of people queuing up to start their
>Subject-headers with "cmsg "?
I dunno. Are there still extant servers that interpret a Subject: cmsg...
as a control message and create new groups from it? If so, they are
terribly, irretrievably broken and I didn't see why people should still be
expected to pander to them. Usefor makes it clear that they are broken,
but is now otherwise silent on the matter.
>Do we have any comment to make in 3.1.1.9 about posting agents generating
>Followup-To headers pointing to groups which do not exist on their immediate
>host? (I believe most current servers disallow this behaviour.)
There is some wording in 5.5 of Usefor that says you SHOULD NOT post
unless at least one of the groups in the Newsgroups-header is known to
exist (that is legitimate to say in Usefor, because such an article would
fail to propagate, or even to be stored locally).
But the point you raise is where a malicious poster posts to, say, a
pro-abortion group with followups set only to an anti-abortion group. In
which case I think you have a point, and I have made a note of it. Anyone
else support the idea?
>On a separate issue, I definitely think that co-operation between (y)our
>efforts on this BCP and the GNKSA people needs to be encouraged:...
Yes, I agree. My next move is to prepare a summary of the differences, and
after that we should probably bring the GNKSA author in.
-- 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