I'm currently searching for reasons why set difference may not have been included in SPARQL. The places I have seen it mentioned all seem to suggest that SPARQL would be incomplete without it [1][2][3]. And another example works around it using FILTER [4]. I'm looking into this as one of the ways OPTIONAL could have been implemented would've been a combination of project, join, difference and union (3 of which are supported in SPARQL directly). [1] http://www.hpl.hp.com/techreports/2005/HPL-2005-170.pdf [2] http://www.polleres.net/publications/poll-etal-2006b.pdf [3] http://lists.w3.org/Archives/Public/public-rdf-dawg-comments/2005Mar/0079.html [4] http://dblab.sogang.ac.kr/icde06/workshops/data/055_SWDB02.pdfReceived on Sunday, 9 July 2006 13:12:11 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:17:05 GMT