Re: Blank nodes as predicates Re: Blank Nodes Re: Toward easier RDF: a proposal

On 11/22/2018 5:50 PM, Holger Knublauch wrote:
> Also I personally doubt that it would make RDF more widely used. Most 
> other languages used by average developers (e.g. JSON) have a basically 
> object-oriented data model, where objects have named properties, and the 
> values of these properties are either literals or other objects. 
> Following the same assumptions in RDF limits that mismatch.

I don't actually actually see this as different in RDF, if you equate 
"objects" with "nodes".  An programming language object is connected to 
(usually named or typed) property values that may be literals or other 
objects.  An RDF node (the subject) is connected to Objects or literals 
via named or identified by Predicates.  It's no different.  Just read 
"object" for "Node", "property" for "Object", and "property name" for 
"Predicate" (I'm capitalizing the RDF terminology here because of the 
potential confusion between the two uses of "object".

What's different is the type systems and how various predicate types are 
allowed, as well as how graphs can be combined and logical conclusions 
can be drawn.

Received on Friday, 23 November 2018 03:08:31 UTC