- From: Keith Alexander <k.j.w.alexander@gmail.com>
- Date: Thu, 06 Sep 2007 12:56:59 +0100
- To: cr <_@whats-your.name>, semantic-web@w3.org
> all im saying is feel free to invent what you want, but youre not > inventing the simplest thing that could work. Maybe we need to define our terms - simple for what? We think this RDF/JSON is simple for (eg): * expressing the whole RDF model in a consistent, resource-centric structure * iterating over resources, properties of resources, and (typed) values of those properties * accessing a specific property of a specific resource. * knowing whether an object is a literal or a uri, or a typed literal, or has a language token. and also relatively simple for being able to merge RDF from different sources (though not so simple as an array of triples). If you can suggest a simpler way of achieving those aims, we'd be very happy to entertain it - or if you would persuade us of more important aims, that might be interesting too. > youre also throwing away the ability to return graph results in a way > that the programmer might want to naturally loop over them. but then > SPARQL doestn really suport recursive query results so i could see why > that wasnt on the radar Do you mean that resources should be nested? The trouble with nesting resources, is that it would be the publisher who decides, arbitrarily, how resources should be nested, The consumer then doesn't know where to find a given resource in the structure. Yours, Keith -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
Received on Thursday, 6 September 2007 12:04:59 UTC