Protocol-16, Rationale-13, Defs-11

[John C Klensin]

These three documents are now in the posting queue.  

Changes from the last-posted versions:

Rationale-13: Section numbers added to references to other
IDNA2008 documents, plus some very small tuning changes.  I
don't think any of the changes are significant.

Defs-11: See the change log, but a number of changes have been
made here pursuant to comments during Last Call that both seemed
safe* and didn't get any opposition.  Those changes include
clarifying the upper-case issue with A-labels (and hence ASCII
characters appearing in U-labels), removing the 63 character
limit on U-labels, replacing it with warnings about programmer
cautions.  Section numbers have been added here too.

Protocol-16: A fairly large number of changes have been made in
this document.  I believe that all of the controversial ones
have been discussed on list (or at least raised on list and not
disputed).   They include tightening up the discussion of
changes to Punycode, clarifying version matching between IDNA
code and Tables and Unicode versions, and making all of the
editorial changes suggested by Martin that I believe are safe*
(including another rewrite of Section 4.1).  Section numbers
have been added here too.   See the change log and/or do a diff
for details.

* As discussed on the list and confirmed by Vint, I
	think we should be reluctant to make major editorial
	changes or rearrangements of text at this point because
	of the risk of causing unexpected problems or of the
	changes having side-effects.   The term "safe" is used
	above to describe changes that seemed reasonable and
	didn't seem to violate those principles and are hence
	best left for a Draft Standard version of these specs.
	Even eliminating apparent redundancies may fall into the
	"unsafe" category because they may not be as redundant
	as they appeared.   I'll try to prepare and post a
	detailed summary of the change proposals that I have
	deferred within the next week or, but I believe that all
	remaining suggestions are editorial rather than
	substantive.
	
     john