- From: Markus Lanthaler <markus.lanthaler@gmx.net>
- Date: Sun, 7 Jun 2015 10:58:19 +0200
- To: <public-linked-json@w3.org>
Hi Maik,
On 3 Jun 2015 at 23:32, Maik Riechert wrote:
> Hi everyone,
>
> Assume I have:
>
> "spatial": { "@type": "dct:Location", "geometry": "POLYGON((...))" }
>
> Which is dictated by the ontology I need to use. Ideally though, I'd
> like to have:
>
> "spatial": "POLYGON((...))"
>
> and hide the rest somewhere but in such a way that it can be reconstructed.
>
> I know that this is not possible with vanilla JSON-LD, but it may be
> related to the functionality of
> https://github.com/antoniogarrote/json-ld-macros which a client would
> have to use to reintroduce the inner nodes as given above. As a first
> question, is that right?
Yes, JSON-LD macros are intended to transform JSON to JSON-LD.
> I have a feeling that this is not an uncommon use case. I had the idea
> of actually putting the transformation of JSON-LD Macros somehow into an
> extension section of the @context which certain clients that know the
> extension can use, and all others would fail. Bad idea I guess, for
> interoperability.
Yep :-)
> Except it gets a popular extension and is implemented
> everywhere. How are your thoughts in general on this issue?
Why would you want all JSON-LD clients to do additional work instead of doing the pre-processing just in your client? Or, asked differently, why do you want to use "spatial": "POLYGON((...))" in the first place
--
Markus Lanthaler
@markuslanthaler
Received on Sunday, 7 June 2015 08:58:50 UTC