[
Lists Home |
Date Index |
Thread Index
]
- From: Bill dehOra <wdehora@cromwellmedia.co.uk>
- To: xml-dev@lists.xml.org
- Date: Fri, 20 Oct 2000 10:02:51 +0100
>Now whether or not this is the Semantic Web is up to TBL, it's
>his vision,
>and he has my respect for being the inventor of something as
>Mind Bombish as
>the WWW. I would love to visit him sometime, in a lecture-hall setting
>perhaps, and listen to his mind talk about the Semantic Web.
>Or I'd like to
>interview him, Charlie Rose-style to pull the vision out of
>his mind, in
>words, so the rest of us can get busy making software for it.
Minimum requirements for a semantic web:
-2 pieces of markup.
-an explicit link between them.
-a program that processes the above and outputs a new piece of markup.
In one sense, this is just processing over hyperlinks. These are the same
minimum requirements for a non-semantic web. We've being doing this for 30
years, thanks to folks like Doug Englebart, Ted Nelson and Tim Berners Lee.
An aside to anyone involved in organising www10: you should really try and
get these guys into one room.
The semantic web is really the notated web. Think Hypernotes instead of
Hyperlinks.
Logical inference (automagicking the semantic web) is just finding the
implicit links betwen bits of markup and telling someone about it.
The big idea behind the semantic web is that we can do more significant
stuff than turn some words blue and draw lines underneath them. But the
significance is in the eye of the beholder. Qualia nuts notwithstanding, the
semantic stuff is not different from the data stuff or the metadata stuff.
Significance makes the difference.
Most of the tools exist already to cobble this together. What is needed are
some good data structures to create the rich links/notes between bits of
markup.
-Bill de hÓra
|
