Re: WWW/Internet 2009: 2nd CFP until 21 September
Date: Mon, 10 Aug 2009 20:57:07 GMT
Walter Mitty wrote:
> Thanks, but you're getting ahead of me. I wasn't really asking if it was a
> viable alternative to RM. Not yet. First I wanted to get clear on whether
> it's even a proposed alternative. ...
Well, I take it that you are really serious about this question. Berners-Lee goes way over the top of a mere proposed alternative. He is suggesting that to "allow the Web to reason as widely as desired" (talk about submarines swimming!), XML "says nothing about what the structures mean". Then that "Meaning is expressed by RDF, which encodes it in sets of triples, each triple being rather like the subject, verb and object of an elementary sentence. These triples can be written using XML tags." A fifth-grader who has been told about intransitive verbs might question these New Emporor's Clothes. Codd went further but in what I'd say is a "rather more" coherent direction, eg., relations of only one attribute and his followers even further with relations that have no attributes. So if Berner-Lee sees RDF as a kind of inside-application of XML maybe he means that RDF is in fact the alternative. I still want to know what are the operators to make an inference from RDF data.
Codd's papers are very different from the other fuzzy stuff, especially because even when he suggested add-ons that many question, such as recording unknowns, he never promised the moon. He offered a classical interpretation for a system that automates the inferences, letting us take advantage of the concise definitions logicians laboured for centuries to minimalize, whereas it looks to me that most RDF systems will end up with a lot of redundancy, recording variations on the same data 'statements' over and over.
I admit I can't take the question very seriously. If Berner-Lee actually believes this stuff, I guess he has as much right to be wrong as the Academy Award voters who perenially deny Bill Murray his just due, a vote in most of the world usually being confused with thought. I don't understand why it is so endemic in IT to want a bandwagon to jump on. I say the people who think Bill Murray is a great actor or that Berners-Lee is wrong about this are wasting their time shouting, better just to think it out for themselves and plough ahead with what they think is right (even if they tried, probably most would barely accomplish only one of those two).
Personally, though my little editorials might indicate otherwise, I think exploring the vagaries of SQL can be more productive. There is too much that has been left unsaid by the xml/rdf/semantic web people whereas at least the sql people gave tried to take a full shot. As somebody said, we learn more from mistakes than from success. Mistakes outnumbering successes in the world by a wide margin, they are more available too. Seeing the real omissions in anything is a pretty rare talent, we can't all be Codds.. Received on Mon Aug 10 2009 - 22:57:07 CEST