John Paul extols the merit of graphics. I certainly agree.
Since we have been using graphics of the type he mentions in the
Interactive Management system for about 20 years or more, we have had lots
of experience with using and interpreting graphics. These graphics are
NOT the type extolled by Tufte, whom many graphics people like. This is
not to say that Tufte's book isn't good--it just doesn't deal with that
type of graphic.
Here are some of the things we have learned from repeated experience with
these graphics:
o From a cognitive or interpretive perspective, the number of words in a
graphics box on a page that has many such boxes, should be approximately
eight. Too few, and there are not enough to convey anything but ambiguity
of definition. Too many, and you've got an overload of cognitive cues,
also conveying ambiguity. This, of course, doesn't say that eight won't
also convey ambiguity, but research shows that it tends to minimize the
ambiguity. This is important when you're trying to deal with a set of
ideas in a common context.
o Many people who see these graphics for the first time tend to think that
every such graphic is a PERT chart. They really don't know how to read
them, and should either have some training in how to read such diagrams,
or should have someone experienced with them to read and interpret for
them (after some "soak time" has elapsed).
o If one tries to capture the content of such a diagram under the
constraint that only English prose is allowed, you will find that this
task is comparable to writing out in English prose the musical contents of
Beethoven's Fifth (or other) Symphony. This is not an attempt to single
out Beethoven.
o Prose is inherently defective in conveying patterns of ideas involving
relationships among a set of ideas. This is why students who study
English for 16 years still can't write these or dissertations about
complex issues. We are looking to the year 2020 as the time when this
idea will finally have made it into the perception of English Departments.
This is not an attempt to single out English.
o People can ultimately comprehend the patterns that the structures
convey. This is like "chunking" a la Miller/Simon, except that it is
chunking of graphic patterns. There is no known way to achieve this
without using the graphic type, formally known as a mapping that is
isomorphic to digraphs. There is no way to become expert in this domain
without learning set theory and the theory of relations.
o Such patterns use the sense commonly called "vision", as opposed to the
sense commonly called "hearing". Hearing is the mode preferred by the U.
S. Congress, by politicians everywhere, by university faculty, etc.
o Vision is the mode preferred by musicians, who are able to map vision
into sound, thus also catering to the mode called "hearing", but in a way
that takes advantage of vision.
-- John N. Warfield JOHNWFIELD@aol.com