The Semantic web ‘lightbulb’ moment

Talk to most people in the GI industry about the semantic web, eyes glaze over or people try to move the conversation on to something else.
Why is this ?
The productive use of Geographic information online is an area where taking a semantic approach is useful, and many would argue vital, if we are ever to develop a true GeoWeb!

It does not help that trying to understand the semantic web involves understanding an impenetrable soup of acronyms and terms, RDF, OWL, SPARQL and ontology etc, however it really helps to have an expert to explain and there is no one better than Tim Berners-Lee who presented at yesterdays Terra Future event as reported by ZDNet.

Semantic tube map

For many people attending i think the real “lightbulb” moment in understanding what the semantic web was all about, was Tim’s slide of the semantic web as represented by a Tube map; after-all GI people love maps..

This really does encapsulate the need to link different types of data (the lines) semantically to extend the usefulness of particular types of application or user need (the stations).

Using the example above, a photo taken by Tim at the Ordnance Survey can be linked via its embedded time stamp to his calendar, which in turn can then be linked to the Terra future event he was attending which in turn could be linked to the address of the Ordnance Survey.

While each of these applications and their associated databases have been developed for specific purposes, by describing and making their contents available using an Resource Description Framework (RDF) description of their meaning (ontology) we can join up these databases.

You cannot underestimate the challenge or importance of a semantic approach to solving the problem of linking geographical databases, an approach of linking databases based on a single view of the world as characterised by the DNF approach can only take you so far.

My thoughts from the other excellent speakers who presented at Terra Future tomorrow..

Written and submitted from home, using my home 802.11 network.

Written by Posted in GIS

3 comments

  1. brindy

    Thanks for that. That diagram and your explaination certainly help a lot.

    Personally I “got” RDF/semantic web a while ago while doing my masters degree. While it’s reassuring that it is something being educated about I think it needs to be brought down a few levels if it is to become a well used standard. i.e. it should be taught at degree level or lower where appropriate.

    However, it hasn’t been until recently, reading your blog in fact, that I’ve finally been able to see where it could be useful. In fact – “GeoWeb” itself is something that isn’t really considered by the community. I’m sure if I mentioned it to my buddies, they wouldn’t have a clue and wouldn’t see it’s potential value.

    So somehow we need to raise the profile of “GeoWeb” and related concepts to those people who use the internet, but don’t necessarily understand it (i.e. the majority of people).

    btw, your link to the DNF is broken.

  2. Ed

    Martin,

    I understand the contra view – but then again Tim was clear to point out that in his views ontology’s are something best developed from a bottom up approach ?

Post a comment

You may use the following HTML:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>