Re: [alac] Redirection of non-existing domain names, again
At 03:40 PM 09/15/2003 +0200, Thomas Roessler wrote:
On 2003-09-15 14:53:33 +0200, Roberto Gaetano wrote:
> I agree with Vittorio for a fast ALAC position, rather than
> coordinating with others. In any case, this matter will be
> discussed by the Board and the synthesis will be done there.
If time is as pressing as it appears to be now, I'm not objecting
against issuing a statement quickly.
> Who can point me to the previously agreed document? I don't
> remember having seen it, and therefore suspect it was drafted
> before I joined ALAC.
Here's what we said last time (in June):
ALAC observed with some concern the recent test in which Neulevel resolved
unregistered domain names in the .BIZ TLD to a search engine rather than
returning the expected "no such domain." The unexpected resolution of
unregistered names may confuse Internet users and the programs they
use. Inconsistent, undocumented responses do not promote the stability
of the Internet. We recognize that this particular instance was a test,
but recommend that ICANN consider the technical and policy implications of
changing DNS responses before permitting further such tests. We would
certainly recommend consultation with relevant technical bodies and
standards groups before adopting any policy endorsing resolution of
unregistered names.
Perhaps we can add the "stability" concern to Thomas's draft.
--Wendy
I don't think we have a formal statement so far.
Here's a first draft:
The At-Large Advisory Committee is deeply concerned by
recent reports that registries might plan to return resource
records in response to DNS queries for names that have not
been registered by any registrant. We understand that the
records to be returned would then point to a special search
engine, that would return a list of possible corrected
domain names to users' web browsers.
This possible practice raises grave technical concerns, as
it would de facto remove error diagnostics from the DNS
protocol, and would replace them by an error handling method
that is tailored for HTTP, which is just one of the many
internet protocols that make use of the DNS. We will leave
it for others to explain the details of these concerns.
For registrants of new domain names, implementation of the
proposed scheme would most likely imply that domain name
servers' caches all over the net will be poisoned with
records pointing to a registry-sponsored search engine for a
certain time after a domain name has been registered. This
would have the effect of diverting net users from
newly-established offers.
From the individual Internet user's point of view, an
implementation of these rumored plans would centralize error
handling at the registry that is rightly done in application
software run on users' computers. Users are deprived of the
opportunity to chose those error handling strategies best
suited for their needs, by chosing appropriate products
available on a competitive marketplace. Software makers are
deprived of the opportunity to compete by developing
innovative tools that best match the user's needs.
We would recommend that the board take whatever steps are
necessary to stop registry operators from implementing such
a service.
At the same time, we observe that possible market demand for
a typo-resolving service at the registry level could easily
be implemented by creating a service separate from the
registry's domain name resolution functions. This service
could then compete with other offers in the marketplace. Of
course, this would also lead to questions concerning the
equal access to zone files of players possibly competing
with each other in this market.
Regards,
--
Thomas Roessler <roessler@xxxxxxxxxxxxxxxxxx>
At-Large Advisory Committee: http://alac.info/
--
Wendy Seltzer -- wendy@xxxxxxxxxxx
Staff Attorney, Electronic Frontier Foundation
Fellow, Berkman Center for Internet & Society at Harvard Law School
http://cyber.law.harvard.edu/seltzer.html
Chilling Effects: http://www.chillingeffects.org/