Re: Talk on Concept Mapping
Peter,
Intent is in the eye of the user:
> I'm not an expert on Concept Maps, but my understanding
> is that they are deliberately meant to be informal; their
> primary motivation is not just to convey information,
> its to help groups collaboratively mentally organize ideas
> together.
Novak and Gowin originally designed them as teaching and
learning aids, which proved to be very useful for aiding
the instructor to spot misunderstandings by the students
and to evaluate their progress.
At IHMC, where Novak is now a professor emeritus, they
have developed software, which does indeed facilitate
collaboration. But that could be called one development
that grew out of the original idea. However, IHMC has
also developed a suite of tools for using the same Cmap
displays to represent RDF and OWL. That is a different
kind of development, and you could consider collaboration
between humans and machines to be a worthy goal for an
Institute for Human and Machine Cognition.
> I'm just cautioning you against assuming "poor information
> conveyance to people" from "formal ambiguity". If that were
> true, we'd all be talking in predicate logic in the first
> place.
But natural languages have very highly developed conventions
for expressing information precisely. I have always maintained
that NLs are the *ultimate* knowledge representation languages.
They have the capability of expressing every step of the analysis
from a vague initial statement to the most detailed elaboration.
They can state anything that can be stated in any version of
predicate calculus -- with the same level of detail *and*
precision. As I said in those slides, no version of logic
can cover that span. Predicate calculus forces you to make
premature commitments before it's clear what areas are worth
making precise and what are to be discarded.
Just as we can critique some uses of English as badly written,
we can critique the Cmap for kayaking as badly drawn. Any
reasonable set of guidelines for using Cmaps should be able
to state why it was badly drawn.
John