>I'm sure all the people who build knowledge bases would be very
>interested
>to know, from Eugene Taurman, that:
>
>>Knowledge can not be stored, only information can be stored. Knowledge
>>has to do with ability and willingness to use information.
>
>While I support the idea that there can be useful distinctions made
>between knowledge and information, I'm not sure that there is anything
>like a consensus to support Gene's statement.
Perhaps knowledge base builders *should* pay attention.
It wasn't too long ago when people agreed that we should just code some
knowledge and put it in a symbol processor to get artificial intelligence.
And robotics people agreed that we should simply embed a model of the
world in the robot's "mind" so it could reason its way around.
Now it's recognized that there's something special about embodied
perception-action systems that "grounds" them to environment for
"intelligent" interaction. These systems ("people" included) learn through
interaction and experience, not by storing symbolic representations and
reasoning with them.
So we should ask what that stuff is in knowledge bases. At best, it
provides a potential information flow to experience, interact with, and
learn from. If people are tuned up to "pay attention" and "see" what's
there (which is where - I expect - the "willingness to use" part of
Taurman's definition could come from), then we say they can learn and
become more knowledgeable. But the "knowledge" really isn't what people
say they know, it's more related to the ability to say it.
Regards,
John Hanna
--"John Hanna" <jh@hanna.connix.com>
Learning-org -- Hosted by Rick Karash <rkarash@karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>