Re: semantic data structure/web architecture and innate logic

HI Tony:

Yeah, I am not a db expert in any sense. I like to fly fish for tarpon.

I said that I was the investor in my first post on Oracle and IBM forums, may have not said that in this one.I have spent a lot of my personal fortune on this human memory simulation software because I am trained in biology (euphemism for BS in Bio). And it deeply interests me. And, I do see machines getting smaller, faster and more capabale in every way...except in the aspect of intelligence. I ask myself "what is intelligence?" I happen to believe the foundation of intelligence is
"innate logic". I go to trade shows and see "knowledge management" and
intuitive software and intelligent agents and neural nets, neural agents, Bayesian theories and blinking "smart" signs everywhere and think to myself- how is it that we humans are able to process intelligently and generate knowledge? How am I able to hear a question, like, "where were you on September 11 when the second plane hit the tower?" And I ask myself, how, with all the data stored in my memory, am I able to pinpoint the specific answer to that question? N squared proceessing nematode-like? Or relational tables, fields and subfields and subrows... or tree/directory structures (trees are only so big on this planet) or what? My partner is a neurophysicist who is able to answer this question. Further, he has spent his life's work developing what he is claiming is the unified theory of cognition. F=ma, PV=nrT and now this new unified theory of cognition. I can pass you along to the patent application if you're interested, but this is no place to talk about the implications of f=ma or e=mc squared. Or our very simple law which describes human logic processing. And the truly amazing thing is that within the human memory structure, simulated with a very simple law, is innate human logic- logic, is indeed, structure determined. You were warned I am talking a different language! But the results are simply startling. a=b, b=c, then a=c.. AND the fact remains that db experts are, if they care to analyze what they are doing at all, simply information organizers. Hey, so is our brain! And it handles multiformat  data, automatically does many to many relations, after all, it contains innate logic. So, I beg the question again, what is logic? And why is it our "blow your horn, create a business plan and make a billion dollars" theme for this forum? Because logic is automatically for our system what you guys do expertly and rules-based. In the classic sense it is automatic IF-THEN relationships. If I have conveyed that, then ask yourself, "how do we specify addresses without sorting or schema?" The brain utilizes methodologies called cognitive functions, which neural net people try to simulate all the time. What is deduction? One way down a linear tree like structure. What is reduction? The opposite of deduction. What is recall? Goes down the structure multiple levels and branches out in varying pathways. What is reasoning? Same, just the opposite direction, BUT goes through a specified node. And so on. Each function finds the most logical node to begin processing the query. Unifying all the retrieval assets combinatorially at the most semantically relevant node. How do we claim logN processing speeds? In my mind, I see what the traditional rules-based people doing is "finding a needle in a haystack". What we do, in our system and in the brain, is analogous to a moth and a light bulb. You ask a specifying question and get a lot of data. You ask a more specific question and get ONE accurate answer. I ask you, how many people in France bought blue cars yesterday? Then I refine my query (OK?) and ask, How many people bought blue Citroens in Nice yesterday? How many people bought new blue SX model Citroens in Nice between the hours of 1PM and 2:30PM?? We can specifiy and deliver these answers without sorting through "who bought golf carts in Spain?" Or why was the sky blue in Madird?" just because the words BLUE or BOUGHT were in the query and the database. Not keyword, but "conceptbased".  What is this query about? We humans don't bother sorting through Golf Carts, unless the query was phrased as, "who bought a blue vehicle in Europe yesterday?" Might get motorbikes, boats and helicopters, right? Wow,I hope you've read this. Believe me, I AM just the investor, but I consider myself somewhat as smart as the average garbage man or deep sea scallop diver. Should have gone into abalone diving...

Now ask yourself what REALLY REALLY is the web? The www? To me, it is a repository of a whole lot of knowledge. Multi-formatted (jpeg, txt, exe. email and so on) with links out. However, access is limited by the
"structure". Rules, language, specificity problems, keywords (whose
keywords are those anyway?), but let me reiterate in my own words what W3C and Berners-Lee are attempting with XML/RDF/OWL and so on. They recognize the www and HTML is non communicative. It is expert language, it is pre-determined by the rules/language/ontology committee and it is NOT SEMANTIC. They are shut down with simply providing 3 associations to each sirte. Some ontology... Sorry...Semantic here means "content oriented". What is your home site "about"? How would I, as a searcher, find your homesite? I am looking for "a nose flute player from Algeria for hire for my daughter's 15th birthday party on September 29. The party will be held in Portland, Maine at 6PM". How do I find this player??? Directory structures have no deeper row than "wind instrument" players. HTML has no way to assemble and parse the solution to my very specific query (check Google- they get bog down at fourth description level, demonstrating rules-based algorithms which crash big time after they loop on third or fourth tier levels... I must say, better than last year when they only handled a couple). Hey. What's the optimum query language? Well, I should ask YOU, what's the universal query langauge? Natural Language, right? We all speak, write and think in it. So, our system is developed with that in mind. Natural language front end- we call it Communication Agent, with a simulated human memory structure as the architecture of the next gen NATURAL LANGUAGE DESCRIPTOR semantic web. Natural Language descriptors are the meatadata. Meaning is resolved when you consider "what is the metadata ABOUT?" Descriptors ARE Content is what it's about. We provide the application to provide
"communication" between web site builders and web site searchers. 100%
accuracy, provides for infinitely associative specificity and targetability, and is processing at logN speeds due to innate logic. Many to many and so on.... Does this sound like I've jumped yet? i MAY HAVE...;-) I would be honored to pass along a demo application... if you are willing to think about this outside the box of rigid db expert rulesbased  programming. TYhere are NO RULES. Universal grammar is the axion by which we humans communicate. I believe you will get it. Hey, sounds like a disciple, right? I hate religion.

Cheers,

Wick

--
Posted via http://dbforums.com