Re: Resolving vocabulary URIs?

On 02/12/2015 08:48 PM, Nathan Ridley wrote:
> I am new to the list, so my apologies if this has been covered
> earlier, or elsewhere. I couldn't find an answer, in any case.

Unfortunately, the response is tribal knowledge at this point and really
should be documented somewhere.

> The context of a document references the URI of one or more 
> vocabularies. My understanding is that this whole idea of linked
> data allows the web to be machine readable by smart clients, however
> I'm generally seeing that the referenced vocabulary URIs (schema.org 
> <http://schema.org> in particular) just go to the site's home page, 
> which is not machine readable in any standardized way.

Sure it is :)

curl -H "Accept: application/ld+json" http://schema.org/

That'll give you a machine-readable document that is then used by the
JSON-LD processor.

The machine-readable vocabulary for schema.org can be found here:

http://schema.org/docs/schema_org_rdfa.html

>From what I understand, there's nothing to link

> So, given that most types ultimately drill down to basic data types
> (string, date, integer, etc.) I have assumed that I would be able to
> look at a JSON-LD document and "follow the trail" back to
> machine-readable sources that would give me enough standardized
> information that I can then generate an appropriate representation,
> without technically having had to know anything about that vocabulary
> in advance.

In general, yes, that's the idea. Some contexts and vocabularies are
better than others. Here's an example of another one:

The JSON-LD Context for a Web Payments CG spec:
curl -L -H "Accept: application/ld+json" https://w3id.org/payswarm/v1

And one of the machine-readable vocabularies there:
curl -L -H "Accept: application/ld+json" https://w3id.org/commerce

> Is this the idea, but which has yet to be realised?

It's realized in the sense that it's technically possible to do what you
want to do (if I understand what you want to do correctly).

However, some communities don't need to or choose to go "full throttle
Semantic Web". schema.org is one of those communities. It's sort of a
Semantic Web / Linked Data Lite, and has been improving with time.

> I'm trying to build a small sample reference client using JSON-LD and
> Hydra, and the lack of machine-readability in the response from a
> vocabulary URI is making me think I have to decide what vocabularies
> I want to support and maintain a copy of each on my own server, such
> that the client can look there instead for the exact definitions of
> each type.

Depends on what you mean by "definitions of each type". There's what a
JSON-LD processor expects (which is a little less strict). Then there's
what a semantic reasoning system expects (which can be very expressive
and strictly typed).

I guess the first question I have for you is: Exactly what are you
trying to do w/ the machine-readable vocabulary URIs?

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: The Marathonic Dawn of Web Payments
http://manu.sporny.org/2014/dawn-of-web-payments/

Received on Friday, 13 February 2015 04:43:52 UTC