Why am I so excited??

Artificial Intelligence researchers have studied such systems since long before the web was developed. Knowledge representation, as this technology is often called, is currently in a state comparable to that of hypertext before the advent of the web: it is clearly a good idea, and some very nice demonstrations exist, but it has not yet changed the world. It contains the seeds of important applications, but to unleash its full power it must be linked into a single global system.

It's not really about Semantics

Many of the people in this audience are focused on putting semantics on the web. But this is the small issue -- the exciting one is What does it mean to have a web of semantics

The semantic web is to ontologies (etc.) as the web is to documents!!!

-> ignore the links, and you ignore the web!

A Language for ontologies

<daml:Class rdf:ID="WeatherEvent">
  <rdfs:comment>Superclass for all weather events
  <rdfs:label>Weather event
</daml:Class>

<daml:Class rdf:ID="WeatherReport">
 <rdfs:comment>Superclass for all handled weather reports</rdfs:comment>
 <rdfs:label>weather report</rdfs:label>
   <rdfs:subClassOf rdf:resource="#DatedWeatherEvent"/>
   <rdfs:subClassOf>
     <daml:Restriction daml:minCardinality="1">
       <daml:onProperty rdf:resource="#hasWeatherEvent"/>
     </daml:Restriction>
  </rdfs:subClassOf>
  <rdfs:subClassOf>
     <daml:Restriction daml:cardinality="1">
       <daml:onProperty rdf:resource="#hasStationID"/>
      </daml:Restriction>
  </rdfs:subClassOf>
  <rdfs:subClassOf>
     <daml:Restriction daml:cardinality="1">
        <daml:onProperty rdf:resource="#hasReportTime"/>
     </daml:Restriction>
  </rdfs:subClassOf>
  <rdfs:subClassOf>
    <daml:Restriction daml:cardinality="1">
    <daml:onProperty rdf:resource="#hasReportDate"/>
  </daml:Restriction>
  </rdfs:subClassOf>
  </daml:Class>

Using the web of Ontologies

• e.g. relate terms in different XML schemas, to...
• RDF schemas describing relational data, to...
• Web service descriptions.

This is a REVOLUTIONARY idea

• This is not the same old AI! Examples:
  • What is the definition of "an ontology" when they are all linked together into a single whole? (transitively)?
  • When third parties can use your knowledge in ways you didn't intend, what can follow?
  • There is NO lexical ambiguity:
    • "Person" doesn't exist -- it is http://www.../ont1#person or http://www.../ont2#person and these are NOT the same thing
    • URIs change the whole game -- constants aren't, variables don't!!
  • When N is huge tractability doesn't matter - decidable becomes heuristic!

The Web IS Huge

• Consider DAML+OIL:
  • Last night the crawler found:
    • results
  • Making it the most used AI language in history!!!
  • which means we're only ont .0006% of current web pages
• If AI people must be in the loop (incl. training), it will never work!
  • And note - number 1 complaint received by Google is that it is too hard to use...

What can (should) I do?

• This works best if we play together
  • Put your ontologies in the ontology library (179 so far) -- it is open, free, and easy.
  • Try your tools on the library, datasets, etc -- it's a "naturally occuring" testbed - or at least the closest we've got.
  • Keep this talk in mind - if your work won't scale to the web, it's not Semantic WEB research.
  • and one more thing...

GIVE IT AWAY

• There is, and will be, no market for any of this unless we create it!
• No one will make money selling their tools until we have MANY more users
• Make small, cheap, easy to download version of your tools available
• Open source made the web (no one paid a penny for Mosaic, most still don't have to pay for IE or Netscape)

closing quote

Start small but viral and you can change many things.