DanC: I think so, but I can't say why, for sure. Am I just lacking sleep?
DanC: see my tentative acceptance of the P3P design
DanC: see my tentative acceptance of the P3P design
dajobe: more work on RDF/XML Syntax
dajobe: now got RelaxNG schema working in non-XML and XML
dajobe: the latter pass most of the good 2001-09-12 test cases and fail most of the error ones
dajobe: also pass/fail appropriately with 2001-11-15 test cases
dajobe: now got RelaxNG schema working in non-XML and XML
dajobe: the latter pass most of the good 2001-09-12 test cases and fail most of the error ones
dajobe: also pass/fail appropriately with 2001-11-15 test cases
bijan: DMiles butts heads with the Chumpbot
AaronSw: Comments are appreciated before I unleash it on RDF Core.
DanC: "The API features the usual notions of Graph, Statement and Node" <-- good, consistent use of syntactic deelies. No "resources". I'd rather s/Node/term/, but I can deal...
DanC: I've been on the edge of learning ruby for a while; might be the best of both the python and perl worlds...
DanC: NTriples.nt2graph() <- you'll want a base-URI argument there, before long, I bet. It's not needed for N-triples, but it is for RDF/xml.
DanC: db.ask <-- that name (ask) is traditionally bound to a much more powerful query function.
DanC: "Provenance: I want to keep track of where statements came from" <- yup!
danbri: ntriples: yeah, needs base URI. I'm concentrating on roundtripping ntriples first, not handling anons nor literals properly yet. The whole parser thing (renamed s/NTriples/Loader/) isn't very objecty.
danbri: Re db.ask(), more powerful query function is an ambition here. Having ask_dump() ask_clever() etc would be one style; I was thinking instead about just having ask() and throw exceptions for query types it doesn't support.