I'll leave the authors to respond on the nits; at first glance they all
seem reasonable to address. On the two more substantive items:
On 10/17/14 9:15 AM, Edward Lewis wrote:
Comment 4
---------
Section 6.
Slight discomfort: I suppose this is not internationalizable given the
nature
of the programming tools being ASCII-based. So no formal objection, but a
desire to allow identifiers to be able to break out of a ASCII-only set
''is
a pony I'd like to have.''
No suggested change, but curious ... if ...
Do note that the restriction described says SHOULD. To quote 2119,
"there may exist valid reasons in particular circumstances to ignore a
particular item, but the full implications must be understood and
carefully weighed before choosing a different course." You *could* go
look at http://www.unicode.org/reports/tr31/ about using Unicode in
identifiers, or you could go look at the Identifier Class in the
soon-to-be-released
https://datatracker.ietf.org/doc/draft-ietf-precis-framework/, and then
try to use one of these for your prefixes, but you had better fully
understand what you are embarking on. :-)
Comment 6
---------
Section 8.1
Addition to third paragraph:
Suggestion: ''For documentation in other than the RFC series, IANA is
requested
to hold in escrow copies of the documentation.'' (This was needed for a DNS
RR type defined in an ATM Forum document.)
This sounds like something that is not unique to this particular
document (that is, any registry that uses "Specification Required" would
have this issue), and might have weird copyright issues. I'm inclined
*not* to put this in the IANA Considerations of this document, but
rather mention this issue to IANA and/or IAB folks and let them work out
a policy to deal with this on all registries that use "Specification
Required", if that is needed.
pr
--
Pete Resnick<http://www.qualcomm.com/~presnick/>
Qualcomm Technologies, Inc. - +1 (858)651-4478