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

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

From: Richard Cyganiak <richard@cyganiak.de>
Date: Wed, 23 May 2012 12:39:51 +0100
Cc: RDF Comments <public-rdf-comments@w3.org>, 'Linked JSON' <public-linked-json@w3.org>
Message-Id: <9F6C09E2-A639-4B0D-B82F-2EE244EBCF99@cyganiak.de>
To: Manu Sporny <msporny@digitalbazaar.com>
Manu,

As I've said before: The JSON-LD syntax spec doesn't explain how to serialize an RDF graph to JSON-LD or how to parse a JSON-LD document to an RDF graph. This may or may not be appropriate given the intended audience. But I don't really see how RDF-WG could publish an FPWD that isn't formally connected to RDF in any way. Elsewhere in this thread, Gregg has proposed what I thought was a reasonable and workable approach to that problem. What you say below seems to disagree with Gregg.

Best,
Richard



On 23 May 2012, at 02:21, Manu Sporny wrote:

> On 05/22/2012 10:42 AM, Richard Cyganiak wrote:
>> Is there a particular reason why the RDF mapping is in the API spec
>> rather than in the language spec?
> 
> The API spec is where the toRDF() and fromRDF() methods are specified.
> There are examples at the bottom of the JSON-LD Syntax spec that explain
> how to express things written in TURTLE, RDFa, Microformats and
> Microdata in JSON-LD. However, explaining every feature of mapping the
> RDF model to JSON-LD is not a priority for that spec.
> 
> That said, you will note that the JSON-LD specification does actually
> cover how one expresses IRIs, literals, literals with language, typed
> literals, lists, and all other RDF model features in the spec. We just
> don't explicitly state that they came from the RDF model (because, in
> reality - they didn't - they came from the Linked Data model... which
> borrows heavily from the RDF model). They're effectively the same thing,
> but the distinction we're making is this: JSON-LD stands on its own...
> you don't need to understand RDF to use it.
> 
>> Is the proposal that RDF-WG should take both the API spec and the
>> language spec to REC?
> 
> The RDF WG should take the JSON-LD Syntax spec to REC. The JSON-LD API
> spec does not have a home right now... it could be this group, or it
> could be the job of a re-chartered RDF Web Apps WG. That decision is up
> in the air right now... and the JSON-LD API spec needs more
> implementation experience before we take it to LC.
> 
>> At first glance, these sections look great. I notice three things
>> though:
>> 
>> 1. I'd prefer if the algorithms were defined in terms of standard RDF
>> terminology (RDF graph, triple, IRI, etc.) rather than API interfaces
>> that use quite different terminology (array of Statements, Statement,
>> NamedNode, etc.)
> 
> RDF graph - maybe. triple - absolutely not, since a Statement is
> effectively a quad. IRI - yes, if we don't do that - it's a bug. The API
> interfaces use things like "Statement" and "NamedNode" because that's
> what we call the concepts via the WebIDL in the spec. We could change
> some of the WebIDL to match the language in the RDF Concepts document,
> if it makes sense to do so. Also note that many of these names came from
> the RDF API spec.
> 
> Whatever we do, I think we should be consistent (not that I think anyone
> is arguing against consistency). The API spec has not had the same
> amount of review that the JSON-LD Syntax spec has, so I wouldn't say
> that it's ready for a thorough vetting at this point. It should give you
> a good idea of the direction in which we're headed, though.
> 
>> 2. Examples would be great.
> 
> Agreed, added a bug:
> 
> https://github.com/json-ld/json-ld.org/issues/127
> 
>> 3. Is it possible to serialize an RDF graph into a “pretty” JSON-LD
>> document using a context?
> 
> Yes, absolutely.
> 
>> I presume the answer is “yes” and involves Compaction of the basic
>> serialized output.
> 
> Yep. You could also frame the RDF graph if you only wanted a subset of
> it as well.
> 
>>> Btw. I can't join the RDF WG mailing list. Is there anything I must
>>> know?
>> 
>> I believe that the list is members-only (with public archive). I
>> don't know why. There is an open comments list here, which is
>> generally better if non-WG-members are involved:
>> http://lists.w3.org/Archives/Public/public-rdf-comments/
> 
> I think it's vital that the following three editor/contributors to the
> JSON-LD spec are added to this group as Invited Experts: Gregg Kellogg,
> Niklas Lindström, and Markus Lanthaler. These three are intimately
> familiar with JSON-LD and have been on just about every JSON-LD telecon
> since the JSON for Linking Data Community Group was started.
> 
> Talking about JSON-LD on public-rdf-comments cripples the discussion
> because I don't think that many of the RDF WG members check this mailing
> list (as evidenced by the relative lack of activity on this list).
> 
> -- manu
> 
> -- 
> Manu Sporny (skype: msporny, twitter: manusporny)
> Founder/CEO - Digital Bazaar, Inc.
> blog: PaySwarm Website for Developers Launched
> http://digitalbazaar.com/2012/02/22/new-payswarm-alpha/
> 
Received on Wednesday, 23 May 2012 11:40:26 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 23 May 2012 11:40:27 GMT