Please excuse the incompleteness of these expressions, they will get more precise with your criticism and contribution. I began to try to tie these ideas together in the fall of 1995, when I gave a guest lecture at UT for Koko Kishi's course on Multimedia. I had served on Ms. Kishi's Master's Thesis committee and she and I have talked about these ideas for some years. As have Dr. Tim Rowe an many of my other friends who a patient enough to listen. This lecture gave me the incentive to tie some of this material together into a consideration of the future of the area of study which we refer to as Multimedia. As an introduction, I used a discussion of the timeline of related developments and innovations. (rather incomplete, but it conveys the ideas)
This area of endevor may at first seem strange for an engineer and product planner to be involved in, but it isn't. Today our ability to develop products has outstripped our ability to think about why we need these products. More and more, product planning is becoming "applied philosophy". In the past, philosophers have been insulated from the embarrasment of actually implementing their philosophies and watching them fail. Today, the philosophy results directly in a product and if "the dogs won't eat it" the philosophy is wrong. Computer products are at a juncture, the "Language Problem" must be solved if we are going to interface with the rest of the marketplace. By the "Language Problem", I mean the ability of the computer to truly understand the language of human beings, as would another human. The very basis of human language seems to be embedded within our senses and is intimately involved with perception. Logic and reason are learned skills which are not innate in human behavior. The computer is very successful at tasks which involve these skills but cannot effective communicate with it's human operator. The efforts of the last thirty years to build artifically intelligent machines has had limited results. Many practicioners in the field have taken up the study of real intelligent systems, ie. people and animals. Cognitive Science holds out the hope enough understanding of the human system to allow us to implement a system in silicon. (the Wright brothers watched birds) I believe that much of the knowledge we seek is already resident in the minds of those people who have been studing people and their humanity for centuries. It is time for engineers to talk to humanists, artists, poets writers and creative people who use their humanity to proceed. The WEB seems to be an optimum medium to pursue this study.
For most of my adult life I have been trying to figure out how to extend Shannon's Information Theory to human language. Rather than finding solutions, I have found the problem expanding in directions which I did not anticipate. Therefor this effort is directed toward defining the problem.
Shannon Theory states that the information content of a message transmitted through a channel is inversely proportional to the natural log of the product of the probabilities of oach of the characters contained in the message. (or something like it). The simplest case is a one character message from a two character alphabet with equal probabilities. The amount of information in such a message is defined to be a "bit".
Human language, as differentiated from any of our character sets, is extensible. Human language, any language is not a closed set. The nature of human language allows us to add words, to make up words, and to change the meaning of the old words. Our brains seem to store information is story form, actors, stages, relationships, motivations, hopes, desires, intentions, etc. whether experienced directly or through the related experience of a story teller. Words become short hand tokens for identifying shared experiences. The teacher is a story teller who gradually gives meaning and history and experience to the words that are used by the students. The classroom is the shared experience. TV becomes an experience sharing event, with an entire watching audience keying a word or phrase to an event or an episode.
Language seems to be linked to experience, specifically, sensory experience of a human speaker. The underlying basis of meaning is so tied up with our mechanics of perception that no fundamental breakthroughs of computer understanding of language can be made without somehow coupling language and perception through experience. Humans also, like computers, are capable of using words without understand what they mean. This is explified by students who learn only the "buzz words" of a subject with learning the underlying physical meaning.
At first glance, the study of the Humanities would seem to be the antithisis of technology. Yet it is within the Humanities that the accumulated knowledge of our kind, man kind, has been distilled from thousands of generations of human experience. It is from this body of knowledge that the seeds of the next great technical breakthroughs will come.
Logic is a discipline that is learned. Nature provides her wards (us) with innate abilities to perceive but not necessarily to reason. Reason and logic are a hard learned, external capability that we forget is not natural. Our machines are logical, but not natural in their reasoning. Logic is what allows us to accumulate knowledge of the real world, but it limits our ability to deal with our inner world.
We all use the word Knowledge and seem to know what it means. Yet, we have no generally available quantitative definition to work with. We know when we, or others don't have it. We use it when we do have it. But it has been very tricky to tie down to a defination from which we can make measurements which both sides of any argument would agree represents knowledge.
Even less, can we measure Wisdom. So, we shall undertake to examine and try to develop measure of :
Human Information: measured in "hits"
Knowledge: measured in "kits"
Wisdom: measured in "wits"
(We must, of course, keep out "wits" about us, in
case we score a "hit" with our "kit".)
To WIK Home Page
To bbott Home Page
To AusBComp Home Page
email the Bill Bottorff