[rfc-i] Comments on RSE models

Ted Hardie ted.ietf at gmail.com
Thu Jan 13 09:53:26 PST 2011

Hi Nevil,

Thanks very much for sharing your experience.  I'd like to draw you out
on one aspect, if I may.

On Wed, Jan 12, 2011 at 8:12 PM, Nevil Brownlee
<n.brownlee at auckland.ac.nz> wrote:
>  However, now and then
> something unusual comes up that needs to be resolved, e.g. a question
> like "can references to I-Ds include their version number?"  Such
> questions need an answer within a reasonable time; that's hard to get
> via a Committee Discussion.  It needs 'rough consensus,' with a
> definite path chosen - exactly as happens within a Working Group.

I personally believe that answering this sort of question can be done
by the RSE in either the model Glenn put forward or the model Paul
put forward.  Where I think it differs is when the RSE spots a larger issue.

Let's say for this one, the RSE decides that a normative reference to
a document which must expire or where the text referenced has changed
should not be allowed.  First, I'd expect the RSE to document that in
the style manual along with what to do.  There are a variety of ways
of handling
that (including the relevant text in an appendix and referencing that
being one I've personally seen), but having a documented default would be

But if the RSE decides that this raises the larger question:  is it time to
treat IDs as archival, since they functionally are archival within most meanings
of the term?  In Paul's model, that question gets chewed over by the
oversight committee and they decide whether to raise the question to the
community.   In Glenn's model, the RSE brings it to the community directly
and, for some value of the manages, manages the community toward consensus.

What would be the difficulties for you as ISE in those two models?  Or do
you believe one model forces the oversight committee to decide issues at
the "can X be a reference" level?

> Again, the production team focus on getting RFCs published.  That
> means they don't have time to spare to do things like updating the
> Style Manual.  I don't believe that a Committee would get that done in
> a reasonable time, there needs to be someone who does that as part of
> their job.

I think most folks here agree with you, and that having someone with the Style
Manual development in their job description seems to be a point of
common agreement.
But this is a pretty different task than the executive management that Glenn's
document also implies, and it's not clear to me why it couldn't or shouldn't
be done by a Senior Editor in Paul's.

> Overall, I agree with Brian Carpenter, who said "we don't want either
> a robot or management by committee.  An autonomous Senior Person and
> an oversight committee seems a much more effective way forward to me."
> Furthermore, I believe that's what Glenn's proposal is trying to achieve.
> Cheers, Nevil
> --

thanks again for sharing your thoughts,

Ted Hardie
> ---------------------------------------------------------------------
>  Nevil Brownlee                    Computer Science Department | ITS
>  Phone: +64 9 373 7599 x88941             The University of Auckland
>  FAX: +64 9 373 7453   Private Bag 92019, Auckland 1142, New Zealand
> _______________________________________________
> rfc-interest mailing list
> rfc-interest at rfc-editor.org
> https://www.rfc-editor.org/mailman/listinfo/rfc-interest

More information about the rfc-interest mailing list