Re: Bob is not a shape [Was: Re: Shapes are Classes, even if you don't use rdf:type]

On 1/28/15, 3:41 AM, Karen Coyle wrote:
> Without doing some major surveying of uses, I don't think it wise to 
> make statements that begin with "most."

Every statement from anybody in this list could be prefixed with "in my 
experience", or "I think". Well, this would make it quite unreadable 
quickly. Nobody knows the absolute truth, not even the statistics that I 
faked myself.

> Our context should be the capabilities of RDF and OWL, not what 
> happens today within some applications. If RDF/OWL support a function, 
> then our solution should also support that function.

LDOM is compatible with RDF and OWL. It uses RDF syntax, some URI terms 
(rdfs:Class etc) and operates on SPARQL. SPARQL can be executed on top 
of a triple store that has OWL inferencing activated.

> Otherwise, we are designing for something other than RDF. Which, 
> Holger, it sounds like you prefer.

I wonder how you come to such a statement? What do I need to change to 
make myself clearer? Of course RDF is fully covered.

Having said this, I see a great opportunity to build something that 
could also be used by tools that do not use RDF internally. In 
particular, many LDOM structures serialized with JSON would be useful by 
normal JS-based software - declarations of classes and properties 
basically align with mainstream "UML style" models. IBM reported similar 
stories from their tool integration platform.

HTH
Holger

Received on Tuesday, 27 January 2015 21:40:22 UTC