- From: Alex Milowski <alex@milowski.com>
- Date: Fri, 27 Apr 2012 20:57:39 -0700
- To: W3C RDFWA WG <public-rdfa-wg@w3.org>
On Fri, Apr 27, 2012 at 6:12 PM, Gregg Kellogg <gregg@greggkellogg.net> wrote: > It's okay for a processor to generate more triples. That's one of the reason we use SPARQL to test the results rather than graph comparison. > This doesn't make a lot of sense from a QA perspective. The RDFa 1.1 sequence algorithm specifies an set of triples that is equivalent within adjusting for blank nodes and document base uris to any set generated by a processor. As such, the test results should specify the exact graph expected to ensure the processor is not generating extra triples it shouldn't. I've implemented my test harness to check equivalence and, so far, most tests have worked without much adjustment. I have had to add a few "uses vocabulary" triples but there have been ones like this one where there are a lot of triples missing. This all makes an implementor wonder whether they've got it right or not. -- --Alex Milowski "The excellence of grammar as a guide is proportional to the paucity of the inflexions, i.e. to the degree of analysis effected by the language considered." Bertrand Russell in a footnote of Principles of Mathematics
Received on Saturday, 28 April 2012 03:58:09 UTC