Where Are All the RDF-based Semantic Web Applications?

Kingsley Uyi Idehen

kidehen@openlinksw.com — Thu, 02 Oct 2008 13:29:35 GMT

Richard MacManus wrote:
<<
Kingsley, thanks for the post. I must admit on first reading, it was all greek to me. Remember my job is to try and interpret what you guys do to 'the mainstream' (heck even early adopters like me find it difficult to follow!) So you have to help guys like me out, by spelling it out in plain english. Anyway, I got an interpreter in to help me with your post, and basically we came away with this question:

1) if this technology automatically identifies entities in text (ala Calais) and injects RDF annotations, then it is top-down

2) If this technology allows editors to high light entities in text and then generates RDF, then it is bottom up

So, which is it? :-)
>>

Our products are "Top-down" heavy. By this I mean, we produce RDF based Linked Data "on the fly" without user intervention.

They are also "Top-down" heavy because our company mantra and product architecture philosophy is simply about:
Maximum incorporation of new and emerging technologies with minimum (if any) disruption to existing infrastructure.

The Document Web (or Web of Linked Documents) is "existing infrastructure" in our world-view, so our products are geared toward distillation of the entities associated with Web Documents, and the projection of linkage (intra and inter) between these entities in the manner espoused by the "Linked Data" meme.

I am happy to spend sometime with you on a "Linked Data Web" in plain english type discourse, it's clear to me that we have much work to do in this critical area.

Richard MacManus

editor@readwriteweb.com — Thu, 02 Oct 2008 02:57:08 GMT

Kingsley, thanks for the post. I must admit on first reading, it was all greek to me. Remember my job is to try and interpret what you guys do to 'the mainstream' (heck even early adopters like me find it difficult to follow!) So you have to help guys like me out, by spelling it out in plain english. Anyway, I got an interpreter in to help me with your post, and basically we came away with this question:

1) if this technology automatically identifies entities in text (ala Calais) and injects RDF annotations, then it is top-down
3:48 PM

2) If this technology allows editors to high light entities in text and then generates RDF, then it is bottom up

So, which is it? :-)

Danny

danny.ayers@gmail.com — Wed, 01 Oct 2008 23:35:52 GMT

Really glad you responded to this one. Still think there's a long way to go to mainstream, even given cool tech. But we're getting there :-)