- From: Ivan Herman <ivan@w3.org>
- Date: Thu, 24 May 2012 09:05:49 +0200
- To: Manu Sporny <msporny@digitalbazaar.com>
- Cc: RDF Comments <public-rdf-comments@w3.org>, Linked JSON <public-linked-json@w3.org>
On May 24, 2012, at 04:40 , Manu Sporny wrote: > On 05/23/2012 09:12 PM, Richard Cyganiak wrote: >> On 24 May 2012, at 01:58, Manu Sporny wrote: >>> The JSON model is capable of round-tripping the RDF model. The RDF >>> model is not capable of round-tripping the JSON model. >> >> That should not be an issue — trying to round-trip HTML+RDFa loses >> the pure-HTML parts too, for example. > > Good point. > > Here's what's concerning me - we don't say that HTML+RDFa is a > serialization of RDF... we say that RDFa is a serialization of RDF that > can be expressed in HTML... HTML can hold many other types of data. > > Similarly, JSON-LD is not /just/ a serialization of RDF... you /can/ > serialize RDF using JSON-LD, but you can also mix-in regular JSON. So > JSON-LD can hold other types of data than just RDF. > > This conversation is getting so pedantic that I want to slap myself... > so I'll stop here. :) Nevertheless... putting it this way the comparison of JSON-LD and RDFa, ie, JSON and HTML, is compelling and good. Which actually also reinforces the issue that came up yesterday on the call: after all, the RDFa spec goes into quite some details, as you well know:-), on how the RDFa content can be turned into RDF. That is an integral part of the RDFa spec and that we were looking for in the case of JSON. Which raises another point: would it be possible (certainly would be great!) to have a JSON-LD Primer as a WG Note, eventually? Ivan > >>> In order for the RDF model to support round-tripping the JSON >>> model - it would have to support null-able subjects >> >> What's that? Just property-value pairs without a subject? > > Yep. > >>> and it would have to support plain literal predicates. >> >> Right. Why not just stick them onto the end of some standard >> namespace? > > (super-pendantic mode unlocked) - because that an IRI, not a plain literal. > >>> I would be fine with this language: >>> >>> JSON-LD is a language that is capable of expressing the RDF model >>> and performing full JSON-LD to RDF to JSON-LD round-tripping. >> >> Shouldn't that be RDF to JSON-LD to RDF? > > Yes, you're right. > >> So how about: JSON-LD is a language capable of serializing any RDF >> graph, and performing full RDF to JSON-LD to RDF round-tripping. > > Works for me, committed: > > https://github.com/json-ld/json-ld.org/commit/584dec5b3d328fdd8f8d85bc9403b256fbc83ed3#L0R2368 > > and live (see Appendix B.1): > > http://json-ld.org/spec/latest/json-ld-syntax/#rdf > > -- 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/ > ---- Ivan Herman, W3C Semantic Web Activity Lead Home: http://www.w3.org/People/Ivan/ mobile: +31-641044153 FOAF: http://www.ivan-herman.net/foaf.rdf
Received on Thursday, 24 May 2012 07:02:30 UTC