Re: [apps-discuss] apps-team review of draft-ietf-hokey-ldn-discovery-06

[Mark Andrews <[e-mail]>]

In message <[e-mail]>, Eliot Lear writes:
> Correcting something I wrote (thanks to Rob Elz):
> 
> On 2/4/11 12:44 PM, Eliot Lear wrote:
> > According to RFC 4282 it should be possible for an NAI realm to begin
> > with a digit.  Strictly speaking this is not allowed by RFC 1035, which
> > RFC 3315 refers to, which this document references.  As a matter of
> > correctness, I might simply reference RFC 4282 instead of RFC 3315.
> 
> There is confusion here.  1035 has display guidelines relating to the
> use of numbers.  RFC 4282 indicates makes the claim that the limitation
> is stronger.  RFC 2181 clarifies this in Section 11.  Either reference,
> therefore, would work, and the authors should ignore the above quoted
> text.  In time it may be worth RFC3315bis referencing 1035 and 2181, and
> RFC 4282 doing the same.

RFC 1035 puts NO, NONE, NADA restrictions on labels other than length.
Hell even 0x00 is allowed as a octet in a label "\000" is how you
enter it.

The restrictions on hostnames comes from RFC 952 as is relaxed by RFC 1123.
The restrictions on mail domains comes from RFC 821 or 822 (I don't care to
look up which).  This may now have been relaxed when these were updated but
RFC 1123 failed to formally do so as it only updates RFC 952.

Mark
> Eliot
> _______________________________________________
> Ietf mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/ietf
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: [phone]                 INTERNET: [e-mail]