RE: N3 and N-Triples (was: RDF in HTML: Approaches)

Interesting ideas, and I must read up on the MIDs. I'm not yet sure about
the gain of shared/serialised bNodes though, if you're going down that path
then all you're really saying is that application A (local) and application
B (remote) are essentially the same application.

I certainly think there may be some work needed to tie down the scope of
graphs though - I would suspect that to be able to correctly reason with a
graph then all bNodes would have to have (terminal) URIs on all their edges.
But there may be a good argument against this - taking the extreme case,
from a certain perspective the semantic web could be seen as one big graph,
in the same way the www is one big hyperlinked structure. The ends of
bNodes' dangling edges may the 404s of this big graph.

 Local isomorphism at bNodes of graphs
>isn't enough, especially when it comes to update.

Why not? (or, how local is local?)

Cheers,
Danny.

Received on Thursday, 6 June 2002 14:00:40 UTC