Pattern Recognition

I am slowly working my way through Quicksilver. This is most definitely not science fiction, which is interesting the same could be said about William Gibson's last novel. Why is it that two of the big Cyberpunk authors have moved so far from the style that first brought them onto the scene? Interesting question, but it's not like I am going to stop reading them, they'r moving into different realms and taking me along for the ride, which is what it's all about.

Quicksilver (The Baroque Cycle, Vol. 1)

Anyway, I need to go back and look at some of the source materials, but the roots of the Royal Society and the things they were working on are fascinating. Here is an interchange between John Wilkins, founder of the Royal Society, and Daniel Waterhouse, starting with Wilkins:

"If we stayed away from politics we could be flying winged chariots to the Moon within a few generations. All that's needed it to pull down certain barriers to progress---"

"Such as?"

"Latin."

"Latin!?. But Latin is---"

"I know, the universal language of scholars and divines, et cetera, et cetera. And it sounds so lovely, doesn't it. You can say any sort of nonsense in Latin and our feeble University men will be stunned or at least profoundly confused. That's how the Popes have gotten away with peddling bad religion for so long --- they simlpy say it it Latin. But if we were to unfold their convoluted phrases and translate them into a philisophical language, all of their contradictions and vagueness would become manifest."

"Mmmm... I'd go so far as to say that a proper philisophical language existed, it would be impossible to express any false concept in it without violating its rules of grammar," Daniel hazarded.

"You have just uttered the most succinct possible definition of it --- I say, you're not competing with me, are you?" Wilking said jovially.

"No," Daniel said, too intimidated to catch the humor. "I was merely reasoning by analogy to Cartesian analysis, where false statements cannot legally be written down, as long as the terms are understood"

"The terms! That's the difficult part," Wilkins said. "As a way to write down the terms, I am developing the Philosophical Language and the Universal Character --- which learned men of all races and nations will use to signify ideas."

While this is a fictionalized discussion, it is rooted in fact and outlines what Wilkins was really working on. Neal Stephenson has a further writeup on the Philosophical Language which was eventually published as "An essay towards a real character, and a philosophical language" in 1668.

So does that Philosophical Language remind you of anything? The "terms", those pesky terms which identify objects and concepts, and do so unambiguously and universally. You should be reminded of the semantic web and RDF which uses URIs in place of those "terms".

So at the very least people were working on these ideas back in the days of the Royal Society. When I talk about the problems that people working on RDF are tackling are large and that they have been worked on for a long time, this is what I meant: 350 years.

I'm currently reading the novel, too, and this part struck me as extremely thought-provoking. What's strange is that I was reminded of RDF, too - but a few pages later, when Daniel actually works on categorizing knowledge and turning it into concepts in the Philosophical Language, and theory and practice seem to clash ;-)

Posted by Stefan Tilkov on 2004-01-24

Sounds a good read, added to my list. Pretty easy to express falsehoods in RDF, but at least the machine is unlikely to complain... btw, there's a spoken logical language (no connection with RDF) - Lojban http://www.lojban.org/

Posted by Danny on 2004-02-07

(Damned the presentation is screwed up) Hmm I think I should translante in english my entry on the different kind of people and their relationships to the Semantics Web. http://www.la-grange.net/2004/01/21.html http://translate.google.com/translate?langpair=fr%7Cen&hl=en&ie=ISO-8859-1&safe=off&prev=%2Flanguage_tools&u=http://www.la-grange.net/2004/01/21.html The main problem of the Semantics Web right now are people over-dreaming about it. They have so many fantasies that they finally drop the water of the bath and the baby with it and they say the Semantics Web is crap. There is not one Semantics Web, there are many semantics web: small applications that works in a given community because they share an approximate and similar language. People developing ontologies know exactly the problems. For example, you can't often translate an ontology from one language to another one. When I say "translate", I do not mean "translation of the words in the ontology" but "translation of the concepts in the ontology". These people are participarting to the effort of the Web Semantics, they know already the problems. Some of them are researchers, some of them are industrials, but they are all working together. So what are the benefits of RDF. The benefits of RDF are very simple (naive?) having a single language to express the knowledge of communities accross the Web. So when you need to share this information or to link both information together, people have less engineering to do. The Semantics Web has not been created to remove the ambiguities of the notion of semantics between culture. It will be here for ever and that's an interesting problem, a very interesting one, to study and it will be easier to do if the language which gives you the possibility to model your concepts is common with others. The choices of vocabulary and ontologies are at an upper level. Now, read this message, I have used a broken english certainly. First part of the misunderstanding. I'm French from France and so we do not have the same process in analizing things, in putting them in Perspectives for philosophy, for culture, for politics, etc. But we have been able to communicate, we will have to make adjustments to have that really productive but it will take times and that's normal. Does that mean we should stop? Another point, if I say: 1+1 = 2 We assume the same concepts, and the same results. We are at a level of abstraction which carries concepts. But even this simple things could be confusing because I didn't give the context of the mathematics I was using. What about :) 1+1 = 1-(-1) = 1-i^2 = (1-i)(1+i) Semantics Web is not about fantasies, it's all about making an effort of collaboration to share knowledge. It doesn't remove the difficulty of describing knowledge.

Posted by Karl on 2004-03-07

It's not RDF, it's not Semantic Web, it's not even RDFS, OWL or DL: rather call it Ontology

Posted by Rinke on 2004-04-23