RE: JSON-LD Basic

On 06/28/2011 10:27 AM Manu Sporny wrote
> Thinking out loud now:
> 
> I think we could not mention RDF at all in the JSON-LD Basic
> document...

Strong +1


> perhaps there are two document conformance levels (Basic and Advanced)
> and one processor conformance level (Advanced)?

Do we really need to have that? A document should always conform to both Basic and Advanced. Why not just create a Basic Primer/Tutorial and an advanced one. The specification itself and the JSON-LD processors should be independent of those two "use cases". Interested developers will also work their way through a slightly more complex specification, others will never bother to even try to read a spec; we should thus avoid the confusion created by having several different specifications.


> Basically, with JSON-LD Basic, there is one primary kernel of truth:
> 
> If you have a valid @context, you have Linked Data.
> 
> What your Linked Data means is fully up to the @context document... and
> most people don't have to care about that at all. They just take their
> JSON markup, add a context and they're done. They know that anybody
> else
> that uses that same context with the same attributes expresses the same
> information. Full stop.

What if developers then start to rely purely on the default context? How do you distinguish plain JSON documents from JSON-LD documents? Are we going to create a new MIME type as well?

I would strongly suggest to remove the default context as it is defined now altogether. Why not simple host that default context document somewhere and require developers to reference it? If we put it in the spec as it is we will never be able to remove an item in there again. If we host it somewhere we could simple upload a new version under a different URI and have a clear upgrade path. Of course developers would be allowed to just copy the "default context" document to their own domain to have full control of it. This will also allow developers to easily check what's included in the default context without having to look at the whole specification.


--
Markus Lanthaler

Received on Tuesday, 28 June 2011 07:37:41 UTC