What David is describing is that any URL you supply in a <link
rel="exhibit-data"> tag is going to be retrieved with XMLHttpRequest and
parsed according to whatever importer you've assigned (Exhibit's JSON,
by default). If you place a mechanism on the triple store end that can
query your triples and return JSON, you wouldn't have to do anything
with Exhibit.
You could also write up a converter on the Exhibit side that took
whatever your triple store normally outputs and turned it into something
Exhibit can internally consume.
These would all be batched; the server would deliver the entirety of the
graph you want to peruse, regardless of which side makes it Exhibit
friendly.
>>> To post to this group, send email to
simile-...@googlegroups.com.
>>> To unsubscribe from this group, send email to
>>>
simile-widget...@googlegroups.com.