RE: lightweight reification (was Representing trust (and other co ntex t) in RDF)

>
>But that compact representation does not extend to the model.  As I 
>understand it, there is no 'aboutEach' node or property in the graph.
>

There is another line of argument about that, but for now lets stick then
with the syntax as in the example I provided, i.e. using production 6.12.

>>
>>Yup.  Do you feel that the digest approach brings significant 
>advantage?
>
>Sergey pointed out that his similar approach reduces the RDF statement 
>triple overhead from 400% to 100%.  Your mileage may vary:  
>this may not be 
>an exact measure of the overhead but it seems a reasonable estimate.
>
>I'd quite like to lose the 100%, but that may be hoping for too much.
>

I'd argue that we have established:

  o  the syntax I outlined in my previous message gives a representation
     as compact as one would get with the digest approach

  o  the processing required is not significantly different to the
     digest approach

Brian




>#g
>
>------------
>Graham Klyne
>(GK@ACM.ORG)
>