W3C home > Mailing lists > Public > public-rdf-comments@w3.org > May 2012

Re: JSON-LD Syntax request for FPWD via RDF WG

From: Manu Sporny <msporny@digitalbazaar.com>
Date: Tue, 22 May 2012 21:43:22 -0400
Message-ID: <4FBC40BA.60006@digitalbazaar.com>
To: RDF Comments <public-rdf-comments@w3.org>
CC: Linked JSON <public-linked-json@w3.org>
On 05/22/2012 06:17 PM, David Wood wrote:
>>> A couple of questions: * Is JSON-LD an RDF serialization?
>> I think the best way of saying it is: it can be used as an RDF
>> serialization, although applications may use it directly, too,
>> without referring to RDF.
> It seems to me that any such format (e.g. Turtle) could be used as
> "just a format"

I disagree. JSON (and thus JSON-LD) can be parsed into native data
structures in a variety of languages. Many languages have built-in JSON
processors. There is no such standard API or data structure for any RDF
serialization that I know of (and no, triples don't count). :)

> but we know that the RDF serializations represent data in accordance
> with the RDF data model.

Yes, but JSON-LD can support things that are not supported by the RDF
data model - like predicates that are string values, or 2-tuples, for
instance (things that you can do in JSON, but not RDF). Sure, you can
model this stuff in RDF... but you have to do strange things like assign
a key-value pair a bnode subject... or generate an IRI from the string
value to ensure that the predicate-expressed-as-a-string "fits" into the
RDF model.

> I would therefore prefer to see JSON-LD described as an RDF
> serialization.

It's the other way around. RDF is a subset of JSON-LD.

> Is there any reason not to do so?

All of the reasons above. :)

-- manu

Manu Sporny (skype: msporny, twitter: manusporny)
Founder/CEO - Digital Bazaar, Inc.
blog: PaySwarm Website for Developers Launched
Received on Wednesday, 23 May 2012 01:44:38 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:29:53 UTC