- From: Alexey Zakhlestin <indeyets@gmail.com>
- Date: Thu, 26 Sep 2013 12:53:24 +0400
- To: Michael Pizzo <mikep@microsoft.com>
- Cc: "public-linked-json@w3.org" <public-linked-json@w3.org>
- Message-Id: <B36FF0A6-0A4E-467A-B81F-8381D9BE7C54@gmail.com>
Another option would be to build OData on top of JSON-LD and use RDF-style namespacing. JSON-LD is very generic, while OData is just a "schema" which can be mapped into RDF and then encoded using _any_ of RDF serialisation formats. Am I wrong? On 26.09.2013, at 3:22, Michael Pizzo <mikep@microsoft.com> wrote: > Yes, I'll grant that my hopes of making this extensible for custom keywords requires either a registry mechanism or a way to associate with a namespace. > > Or both -- the two are actually not mutually exclusive. We can reserve a few "common" prefixes (like "jsonld" and "odata") and allow others to be associated with namespaces through a general mechanism. > > OData actually already has the ability to associate a namespace with a prefix, but it doesn't (yet) prepend the "@" symbol to differentiate the "keywords" from other properties that might contain a period. It would be great if JSON-LD had a way to do the same thing, but for now I'm happy with just prefixing the JSON-LD keywords with a well-known prefix. J > > As far as convincing the rest of the JSON developers, if OData and JSON-LD can agree on a convention, then I think that will help drive others to adopt the same convention. And even if it doesn't become a ubiquitous convention, at least we have better compatibility/interoperability between OData and JSON-LD. -- Alexey Zakhlestin CTO at Grids.by/you https://github.com/indeyets PGP key: http://indeyets.ru/alexey.zakhlestin.pgp.asc
Received on Thursday, 26 September 2013 08:54:02 UTC