Re: ISSUE 1583 status
Harald Alvestrand <[email protected]> Tue, 03 Feb 2009 13:21:18 +0100
| Newsgroups | gmane.ietf.usenet.format |
|---|---|
| Message-ID | <[email protected]> |
Russ Allbery wrote:
> Here's proposed wording, which uses SHOULD rather than MUST for the
> honoring side to leave wiggle room for things like a lost serial number.
>
> --- usepro.xml (revision 5620)
> +++ usepro.xml (working copy)
> @@ -1953,12 +1953,14 @@
>
> <t>The <chksernr> argument may be any positive integer. If
> present, it MUST increase with every change to the newsgroup
> - list and MUST NOT ever decrease. If provided, news servers
> - SHOULD remember the <chksernr> value of the previous
> - checkgroups control message honored for a particular hierarchy
> - or sub-hierarchy and decline to honor any subsequent checkgroups
> - control message for the same hierarchy or sub-hierarchy with a
> - smaller <chksernr> value.</t>
> + list, MUST NOT ever decrease, and MUST be included in all
> + subsequent checkgroups control messages with the same scope.
> + If provided, news servers SHOULD remember the <chksernr>
> + value of the previous checkgroups control message honored for a
> + particular hierarchy or sub-hierarchy and decline to honor any
> + subsequent checkgroups control message for the same hierarchy
> + or sub-hierarchy with a smaller <chksernr> value or with no
> + <chksernr> value.</t>
>
> <figure>
> <preamble>For example, the following Control header
>
>
I'm not very worried about this issue, but this seems fine with me.
(couldn't find the change to the text surrounding the SHOULD, but I
think the SHOULD leaves wiggle room enough for the case where weird
stuff has happened.)
Harald