--On Friday, 17 December, 2004 12:39 +0100 Harald Tveit
Alvestrand <harald@xxxxxxxxxxxxx> wrote:
> --On torsdag, desember 16, 2004 16:37:09 +0100 Eliot Lear
> <lear@xxxxxxxxx> wrote:
>
>>>> RFC1269 Definitions of Managed Objects for the Border
>>>> Gateway Protocol: Version 3
>>>
>>>
>>> Why would this be cruft? The BGP4 MIB was just recently
>>> approved...
>>
>> Good thing too. Take a good look at 1269. I don't think it
>> would pass a MIB compiler test today. If you approved the
>> BGP4-MIB, ought not that have obsoleted this guy?
>
> draft-ietf-idr-bgp4-mib-15.txt says:
>
> This document obsoletes RFC 1269 and RFC 1657.
>
> and the I-D tracker says:
>
> In State: Approved-announcement to be sent :: Point Raised -
> writeup needed
>
> which usually means that the shepherding AD needs to fiddle
> with the RFC Editor note in the announcement before sending it.
>
> It's one of the oddities of the way we process data that it's
> quite hard to know that something's already obsoleted between
> the time the obsoleting document is approved and the
> publication of the RFC.
>
> But I think the old-standards team can take RFC 1269 off the
> list with a note saying "obsoleted by draft-ietf-idr-bgp4-mib,
> no action necessary".
Harald,
Sorry, but I've got a procedural problem with this. I-Ds can't
obsolete anything, even I-Ds approved by the IESG. While "fiddle
with the RFC Editor note in the announcement..." may be the
usual reason for delay, we all know that documents sometimes
change significantly between the last-published I-D and actual
RFC publication. In theory, the announcement could be posted,
the IDR WG membership could take a look at it and conclude the
AD's RFC Editor note does not reflect WG consensus, and an
appeal of the announcement could be filed. As far as I know,
that has never happened, but the procedures clearly permit it
and I can think of a case or two when maybe it should have.
While we have safeguards to prevent it, it is even possible that
a document inadvertently would change enough during the RFC
editing process that the WG would no longer believe it was an
appropriate replacement for the earlier document. Moreover, it
isn't the I-D that obsoletes the older document, it is the I-D,
plus any RFC Editor notes, plus the editing process, plus any
corrections made in 49 hour last call... a set of considerable
uncertainties.
"Work already in progress to supercede this document, hence off
the list" would be a reasonable statement (and that would be a
reasonable statement if such work were at a much earlier stage
of some WG process). But "obsoleted by draft-ietf-..." is not,
IMO.
john
_______________________________________________
Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf