[IP] PTO blows it big time with DNS searching patent

[Dave Farber <dave@farber.net>]

------ Forwarded Message
From: Dewayne Hendricks <dewayne@warpspeed.com>



!20030516  PTO blows it big time with DNS searching patent

     In the last ten years I have been doing PATNEWS, software patent
quality
has not improved one iota.  And in the next ten years, it won't improve one
iota.  The PTO, the patent bar, and Congress still hold patent quality in
contempt.  There is no independent outside quality reviewers of the PTO
(internal quality measures are as dubious as for all other government
agencies), the PTO publishes no non-trivial statistics about patent quality,
the Patent Advisory Board is a joke, and the patent bar is still trying
to figure out how to spell "tough question".

     For example, consider the following totally unpatentable software
method
as of August 1998, and borrowing from out chemical colleagues, call it a
genus-type patent claim (though utterly not genius):

       1. A method in a data processing system with distributed databases,
       each responsible for maintaining application-specific records for an
       associated application, comprising:

       receiving user input containing a query string, a plurality of
       applications and at least one search criterion corresponding to
       the plurality of applications;

       transmitting a request for a search of the query string to each of
       the distributed databases associated with the plurality of
       applications;

       receiving search results from each of the distributed databases
       associated with the plurality of applications indicating, for each
       of the specified applications, whether an application-specific
       record exists for the application in the specified distributed
       database; and

       displaying the search results.

In short, a method of doing distributed database searches, a technique
well established in the 1980s.  Now consider the following species
restriction:

       distributed databases        -->  DNS servers
       application                  -->  domain [registration]
       application-specific records -->  registration records of domain
names
       query string                 -->  domain name

and substitute the restricted species terms into the above genus claim,
resulting in:

       1. A method in a data processing system with DNS servers, each
       responsible for maintaining registration records of domain names
       for an associated domain, comprising:

       receiving user input containing a domain name, a plurality of
       domains and at least one search criterion corresponding to the
       plurality of domains;

       transmitting a request for a search of the domain name to each
       of the DNS servers associated with the plurality of domains;

       receiving search results from each of the DNS servers associated
       with the plurality of domains indicating, for each of the specified
       domains, whether a domain name record exists for the domain name
       in the specified domain; and

       displaying the search results.

As a programmer, and someone with a degree in computer science, let me say
that this type of software technique "genus restriction" is completely
obvious to one skilled in the art, especially in light of mountains of
non-patent prior art for distributed database queries from the late 1980s
and early 1990s, especially distributed database queries of Internet
information.

SO HOW THE HECK DID THE PTO JUST ISSUE THIS PATENT CLAIM TO VERISIGN
(OTHER THAN THE FACT THAT THE PRIOR ART SUBMITTED WAS BASED ON A
REALLY CRAP SEARCH)?

       Patent 6,560,634                                         6 May 2003

       Method of determining unavailability of an internet domain name

       VeriSign, Inc.
           Filed: August 13, 1998

       Abstract
           Methods, systems, and articles of manufacture consistent with the
       present invention provide an improved query server that overcomes
       the shortcomings of existing domain name searching techniques by
       performing a multitude of searches simultaneously, transparent to
       the user. Specifically, the improved query server searches for
       existing domain name records in various domains and then displays
       the results in a formatted manner, thus eliminating the need for
       a user to perform individual searches.

I MEAN, DID ANYONE EVEN BOTHER TO PRETEND TO SEARCH FOR THIS PATENT?
For example, how were the following papers missed, all published ten
years on more before the patent was filed:

     The Clearinghouse: a decentralized agent for locating named objects
         in a distributed environment
     ACM Trans. Office Information Systems, July 1983, 230

     Designing a global name service
     1986 ACM Symposium on Principles of Distributed Computing, 1

     A name service for evolving, heterogeneous systems
     ACM Symposium on Operating Systems Principles 1987, 52

     A model of name resolution in distributed systems
     6th Int. Conference on Distributed Computing Systems 1986, 523

This is all a contempt of science and engineering.  Applicant and PTO
searching is still a joke.  And I have still not heard one word that
convinces me that outsourcing PTO searching will improve anything.
Especially in light of patents like this.

Greg Aharonian
Internet Patent News Service
patnews@patenting-art.com

Archives at: <http://Wireless.Com/Dewayne-Net>
Weblog at: <http://weblog.warpspeed.com>


------ End of Forwarded Message

-------------------------------------
To manage your subscription, go to
  http://v2.listbox.com/member/?listname=ip

Archives at: http://www.interesting-people.org/archives/interesting-people/