vs
this is not a performance test. this an ease and capabilities of mapping comparison.
the input json is characters.json
see
- rml/out.ttl
- json2rdf_and_sparql/out.ttl
- 
RML required a preprocessing stage and quite a big mapping file ( rml/mapping.rml.ttl) (though the mapping is expressed in triples (which is nice)).
- 
JSON2RDF + SPARQL was much easier than i expected it to be and it did not make superflous triples like RML did. - It turns out that JSON2RDF isn't needed either. You can turn json into jsonld (that is almost identical to the output of JSON2RDF) by injectecting some @contextinto the json. I'll update the repo with that alternative soon.
 
- It turns out that JSON2RDF isn't needed either. You can turn json into jsonld (that is almost identical to the output of JSON2RDF) by injectecting some