- From: Erik Wilde <dret@berkeley.edu>
- Date: Mon, 03 Jun 2013 16:49:05 -0700
- To: Wes Turner <wes.turner@gmail.com>
- CC: public-ldp@w3.org
hello wes. On 2013-06-03 15:10 , Wes Turner wrote: > application/api-problem+turtle may be a good solution. > text/turtle is the mimetype for Turtle RDF Syntax. you can only have one of these two ;-) my proposal was based on the convention of most web standards nowadays to mint media types. the second one is based on the point kingsley made that in RDF, this often is pushed into the generic RDF media types. if you want to go this route (regardless of the media type), then you'll have to come up with a mapping of the currently JSON-based model into RDF-land. currently, JSON is the canonical model, and the XML syntax is derived from it. notice that this took a bit of negotiating, because JSON is more permissive than XML in its name syntax, so we restricted the names in potential extensions so that they don't cause trouble in the XML syntax. it's currently just a "should" in http://tools.ietf.org/html/draft-nottingham-http-problem-04#section-4, but at least we have made an effort and documented the reason. my guess is that mapping the problem model into RDF also requires a little bit of tweaking to accommodate for its roots in JSON, and to make the extension model reasonable. i am not 100% sure how to best do this, but i am pretty sure to get to good results it takes some handwork instead of just mechanically mapping JSON structures. cheers, dret.
Received on Monday, 3 June 2013 23:49:44 UTC