> ...Watch nearly anyone trying to
> learn to construct, or for that matter, read this type of
> representation. They will get thrown by the "double work" of not only
> trying to chop up discourse into nodes and give those nodes types, but
> also trying to determine what types the links should be; where do the
> semantics go? Too many choices need to be made. ...
I have to agree that this a weakness of the system. In effect, it asks
people to learn algebra in order to do what they normally do, but do it
better. Symbolic logic was intended as a step in that direction, but it
failed to capture important nuances in useful ways. Our hope with
computerized systems is that we can do a better job of making important
relationships apparent to the beholder (or ourselves, when working on a
hard problem) but it is not clear that we have the mental cycles to do
that well, even as the beholder -- much less as the author.
Fundamentally, we're looking for a good model of collaboratiove design
making, trying to make it function better using automation, with the
goal of using it to capture the *next* great model. (We have two
questions therefore. One, what is the best design/discussion model a
priori, and second, what kind of model can we implement and make useful.
Combining the two questions gives us: "What is the best model we can
implement in reasonable time, and use to further our cause".
> One strategy that, to some degree, gets around this difficulty, is to
> use Questions as the semantic glue between nodes, but that's a matter
> for another posting.
Please follow up on that concept. It feels right, somehow. There was
some discussion earlier on question-based mechanisms, and I note that my
design work proceeds as a series of answers to questions, with the
meta-question always being -- "what questions need to be resolved
together, and which options answer the most questions", so I'm intrigued
by this possibility.
Post message: unrev-II@onelist.com
List owner: unrev-IIemail@example.com
Shortcut URL to this page:
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
This archive was generated by hypermail 2.0.0 : Mon Nov 05 2001 - 10:18:37 PST