[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: draft-eastlake-2606bis-00.txt: Suggestions for modifications



At 13:37 +0200 10/20/05, Harald Tveit Alvestrand wrote:
--On torsdag, oktober 20, 2005 21:17:34 +1000 Mark Andrews <Mark_Andrews@isc.org> wrote:

 2) A different conversation led to the (to me) surprising conclusion
 that  there is no IETF document that conclusively states that top level
 domains  shouldn't be all numeric. I think this is an appropriate thing
 for the IETF  to state in a BCP, since 4-component all-numeric domain
 names are hard to  tell from IP addresses - a technical consideration in
 many protocols.

 	RFC 1123 comes close.

            If a dotted-decimal number can be entered without such
            identifying delimiters, then a full syntactic check must be
            made, because a segment of a host domain name is now allowed
            to begin with a digit and could legally be entirely numeric
            (see Section 6.1.2.4).  However, a valid host name can never
            have the dotted-decimal form #.#.#.#, since at least the
            highest-level component label will be alphabetic.

Yep - but I'm betting that this will be ruled non-normative eventually,
because some people want IDNs in TLDs, and Punycode uses numbers in its
encoding.

It's a long leap from "must be alphabetic" to "can be all-numeric" - but I'd
prefer to have something explicit somewhere, so that we don't end up there by
accident.

I *think* it's uncontroversial. But better safe than sorry.

As many of us have said before, 1123 refers to host names, not domain names. (Host names are a subset of domain names.) In an NS and MX record, you want a host name, not necessarily so in a RRSIG record.

I disagree with the suggestion to bar all-numeric names. Mostly because I don't like "rules" that are unnecessary. I don't see the necessity of such a restriction.

I am not arguing against what 1123 says. From what I understand, if a domain name represents a host, then a numeric label can cause confusion for an application like telnet (to pick on an old, tired horse). There is a necessity for that, even if there is a "newer version [of the application] out there."

But in general, I don't see the harm of an all numeric label, even a TLD.

--
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Edward Lewis                                                +1-571-434-5468
NeuStar

True story:
Only a routing "expert" would fly London->Minneapolis->Dallas->Minneapolis
to get home from a conference.  (Cities changed to protect his identity.)

--
to unsubscribe send a message to namedroppers-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/namedroppers/>