- From: Dave Longley <dlongley@digitalbazaar.com>
- Date: Tue, 16 Feb 2016 20:12:10 -0500
- To: Robert Sanderson <azaroth42@gmail.com>, Linked JSON <public-linked-json@w3.org>, Web Annotation <public-annotation@w3.org>, "public-socialweb@w3.org" <public-socialweb@w3.org>
On 02/16/2016 07:51 PM, Robert Sanderson wrote: > > Dear JSON-LD folks, > > (CC affected Annotation and SocialWeb WGs) > > At TPAC last year, there was discussion that led to the recommendation > of aliasing @id and @type to "id" and "type", if that is possible given > existing json keys and structure. For the Social Web WG and Annotation > WG, we have followed this recommendation for our current JSON-LD based > specs. > > An issue generated for the Annotation group [1] was to provide JSON-LD > frames to make it easier for developers to produce the required layout. > In working on that, it seems that the framing algorithm _requires_ > @type, and does not process the context looking for aliases. Is that > true and intentional? > > An example in the playground: http://tinyurl.com/hqsphco > And the same result is obtained using the Python implementation. > > To make both implementations work, I need to use @type in both the input > graph and the frame document, despite the alias being present in the > referenced (and processed) context doc. > > Bug? Seems like a show-stopper for providing frames? I don't see an alias "type" for the keyword "@type" in the context given in the example. Instead "type" is mapped to "rdf:type". Here's a more minimal example showing framing working with "type" as an alias for "@type": http://json-ld.org/playground/#/gist/a4080a8fa2ce7ddf624f -- Dave Longley CTO Digital Bazaar, Inc.
Received on Wednesday, 17 February 2016 01:12:45 UTC