You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Brian Behlendorf <br...@organic.com> on 1998/02/06 04:01:52 UTC
is something like this really patentable?
I could have sworn something like this was in my algorithms textbook in
berkeley CS classes:
>FASTER ROUTING ON THE WAY
>With Internet destination addresses set to increase in size from 32 bits to
>128 bits, computer scientists at Washington University have patented a
>mathematical procedure for speeding up router performance -- a process they
>liken to the game "20 Questions." The router first divides the address in
>half and compares one half of it to a database. The router then either
>keeps it or discards it in favor of the other half, and then repeats the
>process. Using the procedure, a router should be able to find the needed
>information in no more than seven steps. A number of large router makers
>are negotiating licensing deals with Washington U. (Business Week 9 Feb 98)
Oh yeah, and you Canadians now have one less reason to feel so superior to
us Americans:
>IS BIG BROTHER REALLY WATCHING?
>A secret hearing of Canda's Immigration and Refugee Board was told the
>Canadian government paid $31-million during the early 1980s for
>state-of-the-art software to track Canadian citizens by interfacing with
>credit card transactions, banking data, driver's license information,
>pension records, taxation information, criminal records and immigration
>records, according to transcripts. The U.S.-made Promis system could provide
>details of a person's health care and even library transactions. Updated
>versions are reportedly still being used by the RCMP and CSIS, but neither
>agency could be reached for comment. (Ottawa Sun 2 Feb 98)
Brian
--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
specialization is for insects brian@organic.com
Re: is something like this really patentable?
Posted by Ben Laurie <be...@algroup.co.uk>.
Gregory A Lundberg wrote:
> Anyway, I think the method you're looking for is called a 'binary search'
> and should be in just about any text book on data structures or
> algorithms.
Errr. I doubt that. A binary search for a 128 bit number takes 128
steps, doesn't it? Or am I being completely dozy?
Cheers,
Ben.
--
Ben Laurie |Phone: +44 (181) 735 0686|Apache Group member
Freelance Consultant |Fax: +44 (181) 735 0689|http://www.apache.org
and Technical Director|Email: ben@algroup.co.uk |Apache-SSL author
A.L. Digital Ltd, |http://www.algroup.co.uk/Apache-SSL
London, England. |"Apache: TDG" http://www.ora.com/catalog/apache
Re: is something like this really patentable?
Posted by Gregory A Lundberg <lu...@vr.net>.
On Fri, 6 Feb 1998, Ben Laurie wrote:
> I've been thinking about this, and I don't get it. It may be "like" 20
> questions, but there's no doubt that it is equivalent to 128 question.
> Clearly, the 7 comes from 2^7=128, but how? I can sort of see how
> dividing the thing in half may help, but... anyone got a reference for
> the actual description of the method?
I've given up trying to understand US patent law. They're issuing patents
on basic stuff just because they don't have any _qualified_ experts
reviewing things. The requirement that it not be obvious to an expert in
the field seems to have fallen aside when it comes to computer software.
Anyway, I think the method you're looking for is called a 'binary search'
and should be in just about any text book on data structures or
algorithms. I guess the idea of using something other than a linear
search _might_ be totally new when applied to routers, so we should all
applaud WU for such forward-looking, original research.
----
Gregory A Lundberg Senior Partner, VRnet Company
1441 Elmdale Drive email: lundberg@vr.net [205.133.13.8]
Kettering, OH 45409-1615 USA voice: +1 (937) 299-7653
Re: is something like this really patentable?
Posted by Ben Laurie <be...@algroup.co.uk>.
Brian Behlendorf wrote:
>
> I could have sworn something like this was in my algorithms textbook in
> berkeley CS classes:
>
> >FASTER ROUTING ON THE WAY
> >With Internet destination addresses set to increase in size from 32 bits to
> >128 bits, computer scientists at Washington University have patented a
> >mathematical procedure for speeding up router performance -- a process they
> >liken to the game "20 Questions." The router first divides the address in
> >half and compares one half of it to a database. The router then either
> >keeps it or discards it in favor of the other half, and then repeats the
> >process. Using the procedure, a router should be able to find the needed
> >information in no more than seven steps. A number of large router makers
> >are negotiating licensing deals with Washington U. (Business Week 9 Feb 98)
I've been thinking about this, and I don't get it. It may be "like" 20
questions, but there's no doubt that it is equivalent to 128 question.
Clearly, the 7 comes from 2^7=128, but how? I can sort of see how
dividing the thing in half may help, but... anyone got a reference for
the actual description of the method?
Cheers,
Ben.
--
Ben Laurie |Phone: +44 (181) 735 0686|Apache Group member
Freelance Consultant |Fax: +44 (181) 735 0689|http://www.apache.org
and Technical Director|Email: ben@algroup.co.uk |Apache-SSL author
A.L. Digital Ltd, |http://www.algroup.co.uk/Apache-SSL
London, England. |"Apache: TDG" http://www.ora.com/catalog/apache
Re: is something like this really patentable?
Posted by Dean Gaudet <dg...@arctic.org>.
On Thu, 5 Feb 1998, Brian Behlendorf wrote:
> Oh yeah, and you Canadians now have one less reason to feel so superior to
> us Americans:
>
> >IS BIG BROTHER REALLY WATCHING?
> >A secret hearing of Canda's Immigration and Refugee Board was told the
> >Canadian government paid $31-million during the early 1980s for
> >state-of-the-art software to track Canadian citizens by interfacing with
> >credit card transactions, banking data, driver's license information,
> >pension records, taxation information, criminal records and immigration
> >records, according to transcripts. The U.S.-made Promis system could provide
> >details of a person's health care and even library transactions. Updated
> >versions are reportedly still being used by the RCMP and CSIS, but neither
> >agency could be reached for comment. (Ottawa Sun 2 Feb 98)
Our (canadian) government just decided not to cover it up. I'm sure
you're all being investigated right now.
Dean