Well I’m glad to hear that my solution was sane, but I just don’t see what technological innovation was contributed by the RDF. The file was canonical N-triples

Well I’m glad to hear that my solution was sane, but I just don’t see what technological innovation was contributed by the RDF. The file was canonical N-triples, AKA a CSV with space instead of comma. The predicates establish relationships between subjects and objects, but those relationships could be one to one, one to many, or many to many. Should a given predicate be modeled relationally as a column, or with some kind of join table? I have no idea from the RDF. Say, those objects, are the types I see attached to the values the only types that could possibly appear for that predicate? Who knows! Sure, the data has been interchanged, but the “format” is so generic that it’s useless. Why not just give me a 3-column CSV and tell me to figure it out on my own, rather than pretend that RDF provided some improvement?