Re: RE: I-D ACTION:draft-ietf-ipcdn-pktc-mtamib-08.txt

Hi -

Completing my thought...

In  the places where the document currently says things like:
             "The calculation of 
               384 octets comes from the RFC2279 UTF-8 encoding 
               definition whereby the UTF-8 encoded characters 
               are encoded as sequences of 1 to 6 octets." 

We might do something like
             "The calculation of 
               384 octets came from the RFC2279 UTF-8 encoding 
               definition whereby the UTF-8 encoded characters 
               were encoded as sequences of 1 to 6 octets, based on the
               the assumption that code points as high as 0x7ffffffff might
               be used.  Subsequent versions of Unicode and ISO 10646
               have limited the upper bound to 0x10ffff.  Consequently,
               the current version of UTF-8, defined in RFC 3629,
               will never require more than four octets to encode a valid
               code point." 

This would effectively make the RFC 2279 reference informative,
and the RFC 3629 reference normative.

If we wanted to be *really* rigorous, we'd identify a normalization form
as well, but I think that is overkill here, since these aren't used for
indexing.

Randy