[idn] Editorial comments on nameprep

[Erik Nordmark <Erik.Nordmark@sun.com>]

draft-ietf-idn-nameprep-09.txt
------------------------------

I've reviewed the nameprep document to make sure it is
ready for IETF last call.

Here are some editorial comments that it would be useful to
fix before the last call.

Section 1 says that the applicability is for a domain name label
as opposed to a domain name. It would be helpful to briefly state
in the text why this choice was made (which is basically to leave U+002E 
related processing above nameprep) was made and also point out that 
the user of nameprep needs to separate the domain name into labels 
before applying nameprep.

Section 4. It would be useful to have an explicit reference to the
Unicode table for NFKC (e.g. the file name). The text reads as if
stringprep defines NFKC when it is in fact defined by Unicode.

Section 5.1: The text says that the input mechanism can choose to map
U+3002 to U+002E. But that would potentially cause problems for non-domain
name use of that code point. I think the correct statement in nameprep
is that the application, when it is handling a domain name and splits
in into labels before applying ToASCII, the *application* (not the *input
method*) can do this mapping.
Thus changing "input method" to "application" seems the right thing here.

The RFC 2119 citing text does not include all the 2119 words.
The document uses "SHOULD NOT" which is not included.
The safest thing is to use the boiler plate that lists all the 2119 words.

Section 7 reads like the designers were just too lazy 
to handle similar looking characters, which clearly isn't the case. 
Thus I think it makes sense to add some language saying that
similar-looking is a very hard problem and is (believed to be) impossible
to solve without some additional context, and that context is
not available in DNS lookups.

Split normative vs. non-normative references.

---