On 4/15/2015 7:44, Tom Johnson wrote:
> Thanks Holger,
>
> Dataset seems right, as long as it's clear that the Dataset MAY be
> internal to a validator, or even an individual process. As noted
> above, I think Shapes would benefit from some mechanism for specifying
> the Graphs to include.
On the latter question, I had suggested two new properties that would be
comparable to owl:imports:
- sh:include - state a dependency so that when Graph A is validated, it
should also validate Graph B
- sh:library - state that Graph A uses constraints/shapes defined in
Graph B (forming the "Shapes graph)
If there is any input on this, I'd appreciate to hear about it.
Holger
|