Welcome to the
Semantic Web Interest Group
scratchpad generated automatically from discussions on
IRC at
Freenode channel #swig 2001-2018 approx
by the chump bot.
DanC_pm: bug in hello world example: <People/ID=7> <rdf:type> <People> . DanC_pm: that should be rdf:type . no <>s DanC_pm: W3C Working Draft 24 March 2011 DanC_pm: "This edition of this document presumes hash identifiers" er... except it doesn't. <People/ID=7> <People#ID> 7 . that would be: <People/ID=7#_> <People#ID> 7 . DanC_pm: I still think treatment of keys is counterproductive at this level. It doesn't cover 80% of the cases. Just push it off to the rules-integration layer. DanC_pm: ah... found some test cases 2 hops away. why no mention of them in the status section? DanC_pm: the status section is pure boilerplate. sigh. DanC_pm: not much motivation and tons of "denotational sematics" and other spec geekery. DanC_pm: in a test case: ex:id "10"^^xsd:integer; DanC_pm: that's the sort of thing that gives RDF a bad name. just write: ex:id 10 DanC_pm: wild... the mapping layer includes embedded SQL. I wonder how that's specified. DanC_pm: re motivation, use cases. again, why no mention in the status section? DanC_pm: UC1 - Patient Recruitment <- that's how to get me engaged. That sort of stuff is my day job. Please use that as an example in the spec. DanC_pm: wordpress, tax control use cases are engaging too. I hope to find time to study the requirements. DanC_pm: "Optional Requirements" bzzt. contradiction in terms. try "objective" or "goal" DanC_pm: no motivation is provided for NAMEDGRAPH. smells fishy DanC_pm: the links in the references section of UCR don't work. why the heck not?!?! DanC_pm: "The R2RML language must allow mapping specification to have sufficient information to enable transformation of SPARQL queries over the RDF view into efficient SQL queries over the relational database." in all cases? that's a tall order. DanC_pm: no requirements re RIF integration. interesting.
DanC_pm: that should be rdf:type . no <>s
DanC_pm: W3C Working Draft 24 March 2011
DanC_pm: "This edition of this document presumes hash identifiers" er... except it doesn't. <People/ID=7> <People#ID> 7 . that would be: <People/ID=7#_> <People#ID> 7 .
DanC_pm: I still think treatment of keys is counterproductive at this level. It doesn't cover 80% of the cases. Just push it off to the rules-integration layer.
DanC_pm: ah... found some test cases 2 hops away. why no mention of them in the status section?
DanC_pm: the status section is pure boilerplate. sigh.
DanC_pm: not much motivation and tons of "denotational sematics" and other spec geekery.
DanC_pm: in a test case: ex:id "10"^^xsd:integer;
DanC_pm: that's the sort of thing that gives RDF a bad name. just write: ex:id 10
DanC_pm: wild... the mapping layer includes embedded SQL. I wonder how that's specified.
DanC_pm: re motivation, use cases. again, why no mention in the status section?
DanC_pm: UC1 - Patient Recruitment <- that's how to get me engaged. That sort of stuff is my day job. Please use that as an example in the spec.
DanC_pm: wordpress, tax control use cases are engaging too. I hope to find time to study the requirements.
DanC_pm: "Optional Requirements" bzzt. contradiction in terms. try "objective" or "goal"
DanC_pm: no motivation is provided for NAMEDGRAPH. smells fishy
DanC_pm: the links in the references section of UCR don't work. why the heck not?!?!
DanC_pm: "The R2RML language must allow mapping specification to have sufficient information to enable transformation of SPARQL queries over the RDF view into efficient SQL queries over the relational database." in all cases? that's a tall order.
DanC_pm: no requirements re RIF integration. interesting.