Hello Rich, others,
On 2024-07-10 02:07, Salz, Rich wrote:
UTF8 is upward compatible with ASCII. That is, all seven-bit ASCII characters are valid UTF8 characters.
And while we are at pedanticism, it's UTF-8, not UTF8.
Regards, Martin.
[with my charset reviewer hat on]
On 7/9/24, 12:59 PM, "Dino Farinacci" <farinacci@xxxxxxxxx <mailto:farinacci@xxxxxxxxx>> wrote:
Reviewer: Rich Salz
Review result: Serious Issues
Thanks for your comments Rich.
This is a very short draft that adds "names" to LISP identifiers.
Major nit: Why is ASCII used for names rather than UTF-8? Related, no mention
of punycode as a UTF8 alternative. If UTF8 was considered and then rejected as
not needed, there should probably be a justfication for that decision in the
document.
This came up in the working group and we just decided ASCII was sufficient. And since implementations lead the draft we didn't want to obsolete them or create a compatibility issue.
A document which is part of a system "which are intended to replace most use of
IP addresses" that limits names to the ASCII character set should not be
approved.
Minor nit: "Distinguished Name" has a long history with X.509 certificates and
I could not get past my confusion. Is another name possible? Okay if the
answer is no.
We had to inherit that name since we decided to use AFI=17.
Dino
--
last-call mailing list -- last-call@xxxxxxxx
To unsubscribe send an email to last-call-leave@xxxxxxxx
