Creg wrote:
In class we discussed this idea of whether or not computers can have knowledge or do they just have information. The example was a calculator adding 2 + 2: is that knowledge of addition? I'm still not sure what I think here. My general reaction is that this is the processing of information but I'm not sure how that differs from knowledge. Does the ability to follow a procedure or execute an algoritm require knowledge? What about IBM's Big Blue Chess computer? Does that have knowledge of the game of chess?
I personally like the idea that knowledge requires human interaction - it requires cognition, context, experience, etc. However, I'm not certain that this is true. Nonetheless, I think it is this type of knowledge that is more interesting and has more value within organizations. That is dynamic knowledge vs. static knowledge.
[Greg Harmeyer's KM Weblog]
Yes the answer is that yes a calc does have knowledge because the calc has also knowledge of the (+) that is the algorithm, which has the knowledge encoded it in much like a manual that instructs the computer what to do. Algorithm comes from the Greek Algos which mean "labor (pain)", therefore an algorithm is supposed to be mechanical and manual and repetative, however that procedure
NOTICE NO DOUBLE LOOP LEARNING in the 2+2 THIS IS HARD TO DO IN COMPUTERS
(I think you refer to Dynamic K as the double loop learning, humans call one aspect of the double loop learning their mental model, BIG blue has that mental model encoded again on an aspect of its algorithm or optimization technique, it would be interesting to point out that these algorithms or techniques are also very hard to describe or formulate a-priori therefore in the computer field they are called heuristic algorithms, same as an ad-hoc human approach that we know works.)
However the BIG BLUE not only has algorithms that encode mathematical knowledge in order to evaluate moves but also has Exprert K( database of previous games and analysis) on top of this it also had double loop learning (adaptive learning or AI techniques). And Guess what the adaptive learning is also algorithm(s) (or optimization techniques which I will treat the same) that AI experts have encoded in a program and Is that all information or is it knowledge?
In the KM world, seems that we should only talk about information that encodes K and ALLLLLL information encodes knowledge. The information that does not encode K is just noise. In the case of big blue as in the case of the class case people could and wrote the manual or algorithm (to much use of labor) to beat Casparov or make sure RAs will know what to do with HS students.
Following a procedure does require some K ( for practical reasons I will assume it does not since not everyone is an automaton, for computers yes)
My thoughts.
- All information has K encoded in it
- Information with out K is noise and should not be the study of KM
- Humans only affect the way information and K is encoded and managed.
8:37:33 PM 
|
