OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

 


 

   Re: RDF, the "semantic web",and the nadir of AI (was RE: Realisticpropos

[ Lists Home | Date Index | Thread Index ]
  • From: Ronald Bourret <rpbourret@rpbourret.com>
  • To: xml-dev@lists.xml.org
  • Date: Wed, 18 Oct 2000 10:18:24 -0700

One of the interesting things about this thread is that there seems to
be an underlying assumption that the semantic Web is a well-defined idea
and that it's implementation would work reliably. I've never thought of
the term "semantic Web" as anything but marketing talk ("vision" if you
want to use the latest in upper-echelon-corporate-speak). That is, it
pointed people in the direction of a Web that had some chance of
answering questions like, "Find me all reasonably priced pizza joints in
a radius of 50 miles." (Sort of the same way the idea of a Dick Tracy
communicator fuzzily guides my idea of where palmtops and cell phones
are headed).

It never occurred to me that such a Web would be significantly more
relaible than the Web that exists today -- it would simply support a
more targeted set of questions and links.

Shortly before I started working with XML, I read a number of academic
papers explaining how to extract data from the Web. These were
invariably ridiculous, explaining systems where you wrote a special
parser for a Web site not under your control but on which you knew there
was something like a temperature or price you could extract. My
immediate reaction was that the system didn't scale -- not even to one
Web page.

Then I looked at XML and thought, "This is the answer and it's pretty
obvious, just not all that technically exciting." You throw out the AI
bit and go for the human solution: ask everybody to cooperate and tell
the world what kind of data is on their Web page. It might sound like a
hard sell, but it really isn't, no more than getting people to post Web
pages in the first place.

I figure the semantic Web is just the next logical step: all it does is
provide technology that allows people to build links and metadata to
support the kind of queries that aren't possible today. It's no more
reliable than the people that enter the data/metadata and, in the case
of robot-generated information, not that reliable at all. But it does
allow me to ask my pizza question, sift through the results, and get the
answer I want, just like Alta Vista and Yahoo do today.

As to a semantic Web reliable enough to automagically run
business-to-business transactions -- such as sending an XML-based
invoice to a supplier that you've never dealt with before without having
to set anything up beforehand -- well, it's a nice idea but, sweet
dreams.

Aiming low,

-- 
Ronald Bourret
Programming, Writing, and Training
XML, Databases, and Schemas
http://www.rpbourret.com




 

News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS