Re: [topicmaps-comment] Challenge Part 1 - Part 2

["Thomas B. Passin" <tpassin@home.com>]

[Piotr Kaminski]
>
> Can you expand further on why you think the "fundamentalist" approach is
> unsustainable?  In my opinion, a fundamentalist model will provide a
> flexible basis on which to build different presentations and will allow
> for information interchange outside of the context in which it was first
> captured.
>
> To provide a (somewhat strained) example, suppose I was doing a study on
> the social aspects of making predictions.  Your example statement would be
> one of the "data points" in my study, along with as many others as I could
> find.  If I am interested in finding patterns in the levels of certainty
> expressed, or in the percentages quoted (or a correlation between them?),
> then those pieces of information become primary topics within my context.
>

A close analogy would be the function of a particular word in a sentence.
The same basic word or phrase could be used as the subject, modifier, verb,
etc.  The subject would be analagous to a topic.  Thus the choice of
modeling something with a topic or not would depend on your understanding of
its place in the topic map, or its place in your understanding, or what you
want to focus on or express.  Just because the topic map might be
interpreted by a machine does not, I think, change that.

On the other hand, it is asking a lot for a given program to be able to
"understand" the various approaches and idioms that might be used to
construct a map.  This makes me think that there will need to be something
like topic map patterns, standardized approaches that would be tuned for
different kinds of modeling.  Programs could be written that are especially
effective for one or a few patterns.  This would be more useful than
constantly devising new programs for each new application, and also more
useful than general purpose topic map programs.

Cheers,

Tom P