Re: [registrars] FW: [council] ALAC statement on resolution of non-existing domain names
- To: "Tim Ruiz" <tim@xxxxxxxxxxx>
- Subject: Re: [registrars] FW: [council] ALAC statement on resolution of non-existing domain names
- From: Eric Brunner-Williams in Portland Maine <brunner@xxxxxxxxxxx>
- Date: Wed, 17 Sep 2003 13:48:48 -0400
- Cc: registrars@xxxxxxxx, brunner@xxxxxxxxxxx
- In-reply-to: Your message of "Wed, 17 Sep 2003 08:11:21 CDT." <029f01c37d1d$313b4f80$fa05a8c0@TIMRUIZ>
- Sender: owner-registrars@xxxxxxxxxxxxxx
In reply to the comments of my former co-worker, and please do forward this
There are not two sides to every debate. Consider the IAB's technical
comment of 27.09.99 on the unique DNS root .
Authoritative name servers for .COM and .NET zones now, without notice,
perform name-to-address mapping differently now, and no longer return
NXDOMAIN for exact non-matches.
The scheme originally implemented by RealNames in the CDN market (redirect
on 8th-bit to a non-DNS directory service presented as an extension to DNS)
is neither novel, nor open, though RealNames made an effort in that direction
prior to termination of its relationship with Verisign and its subsequent
liquidation. Both a flaw in the UTF-8 code in the IE browser, and the RN
agreement with MS and Verisign to provide non-DNS DS, resulted in significant
trans-pacific ("overseas bandwidth") costs for Asian network operators, for
pseudo-resolution (directory search) of broken, or "enhanced" names in the
RN and VGRS pseudo-zones.
It is inexact to characterize the new Verisign functionality as simply the
deployment of a mapping layer on top of the DNS, regardless of the motive.
The DNS relies upon exact match semantics, a 1-to-1 relationship, or ERROR.
The new Verisign functionality does not provide exact match semantics.
As you (Jeff Neuman) are personally familiar with the implementation used
by your business partner (CNNIC) for resolution of names which contain one
or more characters outside of the -0-9a-zA-Z range, I'm surprised to see
you (Jeff) ask for examples of users making choices about how to query the
authoritative server for some zone. These users do in fact (via the CNNIC
provided plug-in) modify their MS browser to effect a change of function.
For an IP lawyer to offer a comment on the gravity of the technical concerns
is as useful and interesting as a DNS expert to offer a comment on the utility
of IP case law.
I don't mind zone-specific semantics, but they should not surprise.