Are RDF and RDFS two separate documents?

Question

When writing RDF and RDFS are you supposed to create a single document covering both? or create two separate documents? Similar to xml and xmls.

Answer 1

Either approach is fine, and this is simply down to your own preference. In fact you can split your RDF(S) data between one, two, or more files however you want.

Some things to keep in mind: for small/simple datasets, having everything in a single file may be more convenient, but once you start working with a larger schema (or a lot more instance data), having some separation might make handling the data more flexible.

Answer 2

I think this really comes down to a question of reusability. For instance, at present you have:

• A. a schema
• B. a collection of some instance data

Given that you have those two things, you have two options:

• X. a single document with schema (+) and instance data
• Two documents:
  • Y.1 one for schema (**)
  • Y.2 one the collection of instance data

Now, if expect in the future to have another collection of instance data, that's distinct from the first, then have a two options:

• two documents:
  • X.1 one with the schema (+) and first collection of instance data
  • X.2 one with the schema (+) and second collection of instance data
• three documents:
  • Y.1 one with schema (**)
  • Y.2 one with first collection of instance data
  • Y.3 one with second collection of instance data

If you take the second approach, you'll end up with three documents, but the one marked (**) carries over from the first, and you don't have to worry about keeping it in sync. In the first approach, you end up needing to make sure that the copy of the schema marked (+) is the same in both cases, and you're duplicating that information in multiple places.

In practice, which of these is more important probably depends on what you expect to have to handle in the future.