>this gets back to relational database design. sometimes a tad of >denormalization is recommended for performance. There is a pretty good argument that denormalization for performance reasons is impossible [1] : normalization is a logical concept whereas performance is determined by the implementation. It has also been noted that denormalization is often undesirable as the maintenance of integrity is likely to require additional application-level support (i.e. more work). Whether or not this argument carries over to n-tuples of RDF is another matter... [1] http://searchdatabase.techtarget.com/originalContent/0,,sid13_gci528153,00.h tmlReceived on Saturday, 26 May 2001 12:55:45 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 08:34:46 GMT