WGLC: draft-faltstrom-5892bis-04.txt

[John C Klensin]

--On Wednesday, May 04, 2011 11:33 +0200 Simon Josefsson
<simon at josefsson.org> wrote:

> "Jiankang YAO" <yaojk at cnnic.cn> writes:
> 
>> Reminder. The WG last call will end soon. If you still have
>> any comments, pls kindly give it before the deadline.
> 
> Could you please clarify the last sentence in the WGLC?  As I
> asked before, is the intention that this document will be
> marked as updating any earlier RFC or not?  Right now the
> document does not include any "Update:" tags, so I'm confused
> by your statement that the document updates an earlier RFC.

Simon,

Personal opinion, speaking for myself only.

This is a judgment call.  If the document were _changing_ the
5892 specification, e.g., by adding an exception rule, it would
clearly be updating 5892.  But it doesn't.  It has no effect
whatsoever other than to document our decision to not do
anything.   So, if we had the requirement (one that I've often
advocated) that any document that says "updates XXX" must
describe what was changed and why, it would have to say
something like "despite the claim that this updates 5892, it
changes absolutely nothing about the spec even though it lets
the expected changes in derived tables with Unicode changes
happen".

Similarly, if 5892 contained a comprehensive explanation of when
we do or do not add exceptions, or this 5892bis document did, it
would be updating 5892.  But 5892 essentially says "judgment
call on a case-by-case basis" and 5892bis doesn't change that
either == we are just making that call.   Our intent, from the
original discussions, was that we accumulate that experience and
then see if we set out clear rules.  I don't think we are much
more ready to do that now than we were 18 months ago.

At the same time, the document doesn't exist without 5892 and we
don't have a category other than "updates" for "hangs off 5892
somehow".

My recommendation to the WG is that we not waste time on this.
The WG Chairs (or other shepherd) should attach a note to the
IESG pointing out the difficulty (borrowing text from your notes
and the above if it would be useful) and leaving it to the IESG
and RFC Editor to decide whether an "updates" header is
appropriate.  They will need to decide anyway and the issue is
really a procedural and editorial one, not one on which the
AppsAWG needs to decide.

Sorry Peter and Pete :-)

best,
     john