The claim is clearly nonsense. However, there is some plausibility in the suggestion that the binary relational model is the theoretical ultimate in normalization of data, and thus a reasonable candidate for an Esperanto if you want to write 2*n converters rather than n^2 converters to get from anything to anything else. (I guess I'm stubborn in calling it the binary relational model rather than the RDF model. It's just that I first came across it in 1974, in Jean-Raymond's paper on Data Semantics, about 23 years before RDF was invented, and I think we should give credit where it's due.) Michael Kay Saxonica |