Re: [nc-impwhois] Melbourne IT WHOIS implementation comments

[Thomas Roessler <roessler-mobile@does-not-exist.net>]

On 2003-01-17 11:03:24 -0800, Rick Wesson wrote:

> you bring up an excellent point. our demo is just that, a demo --
> it is not connected to our production system. as we have had
> significant abuse of the demo through the web-services API. We
> stopped updating the data in November and also reduced the data
> sets coverage.

Fine.  So I'll just assume that your production system does the
right thing about telephone numbers in certain countries (and, more
generally, about telephone numbers it can't make a judgement about).
Congratulations for fixing this.

Anyway, the key point of my earlier message (which should be
self-evident) remains: Automated, database-backed systems can check
the PLAUSIBILITY of WHOIS data elements, and only that. Plausibility
is no proof for accuracy.  On the other hand, great care must be
taken before "non-plausible" data can be considered proof of
inaccuracy, and, hence, give reason to the cancellation of domain
names.

Plausibility checkers will always have a certain error margin due to
outdated information, data entry errors, programming errors, and the
like.  They can't be used as the sole source for judgements on the
accuracy of address information.

Any procedures designed by this group, and any suggestions that
something can be done by an automated system, MUST keep this caveat
in mind.

Hope that helps,
-- 
Thomas Roessler				<roessler@does-not-exist.org>