Wednesday 10 November 2004

Semantic Web, Take Two

Recent discussions on the Semantic Web seem to emphasise the downside, which is quite valid. Some people seem to think that "Semantic Web" means "more links". Some others are of the opinion that this is either too much work or not human-friendly. Some have emphasised the fact that the web is easy enough to use right now: we have plenty of external resources to consult for further information if we want. This last one is true enough, but I'd like to rebut it anyway.

The point of the semantic web is machine-friendliness. With the current web data space, if you want more information, you follow roughly this procedure:
  1. Locate an external resource to consult
  2. Determine its reliability
  3. Learn to use it
  4. Search for your extra information
  5. Consume information, repeating this procedure as required
These are all labour-intensive, time-consuming and tedious. With a fully-functional semantic web, your procedure for finding extra information is as follows:
  1. Ask


Mokalus of Borg

PS - That's the difference, friends.
PPS - It's still a very long way off, though.

No comments: