[From Bill Powers (931210.1430 MST)]
Tom Bourbon (9312??) --
Re suggestions about publishing the Little Man article in Closed
Loop. I'm certainly for that, but perhaps Martin's projected
issue of International Journal of Man-Machine Systems would be a
venue with a wider circulation. I had mentioned the idea of
reducing the scope of the article to include just the kinesthetic
control systems, because the visual control systems are not
really worked out as well as I would like. The match with the
real human data will be just as good. Also, it will give an
opportunity to show that the mass-spring model is actually
embedded in the PCT model as its output function, and to explain
how the feedback model takes care of the dynamics without
requiring calculation of inverse dynamics.
By the way, Martin has been wondering why so few people have
responded with offers of articles for that special issue. The
editor has opened the door to articles of a wide variety of
kinds, not just man-machine issues. Martin, how would a deadline
of next August 1st sound?
···
--------------------------------------------------------------
Best,
Bill P.
From Tom Bourbon [931213.0910]
[From Bill Powers (931210.1430 MST)]
Tom Bourbon (9312??) -- [That was a direct post to Bill]
Re suggestions about publishing the Little Man article in Closed
Loop. I'm certainly for that, but perhaps Martin's projected
issue of International Journal of Man-Machine Systems would be a
venue with a wider circulation.
Of course, you are right. JMMS *will* have a wider circulation, far beyond
the PCT ghetto reached by Closed Loop (a statement which in no way implies
Closed Loop is inferior to JMMS). My suggestion about publishing in CL
was selfish: I want to cite the paper in a manuscript I am preparing for
rejection.
I had mentioned the idea of
reducing the scope of the article to include just the kinesthetic
control systems, because the visual control systems are not
really worked out as well as I would like.
That might be a good idea. A number of people to whom I have demonstrated
Little Man were more interested on the visual control systems than the
kinesthetic. On the other hand (no pun intended), to what degree would you
eliminate the visual control systems? Aren't they essential to the
demonstration -- the image of the target must be kept on the "fovea," must it
not? Am I misinterpreting your ideas about how to alter the model for
JMMS?
The match with the
real human data will be just as good. Also, it will give an
opportunity to show that the mass-spring model is actually
embedded in the PCT model as its output function, and to explain
how the feedback model takes care of the dynamics without
requiring calculation of inverse dynamics.
Whatever you do with the visual part of the model, these two points are
essential. I believe the Little Man paper will be a modern classic, in
that it will put to rest the near universal misconceptions in which (1) the
mass-spring model is assumed to explain far more than it does and (2) the
calculation of inverse dynamics is assumed to be a prerequisite to the
preparation of motor "commands."
Sorry if I seemed to be pressing for hasty publication of Little Man --
count that as evidence of widespread demand for the article.
Until later,
Tom
[Martin Taylor 931213 1615]
(Bill Powers 931210.1430)
By the way, Martin has been wondering why so few people have
responded with offers of articles for that special issue.
Yes, I have. In fact, nobody has proposed a paper except for those
I approached directly. I'd particularly like proposals for papers,
long or short, theoretical or based on experience, that deal with
the use or prospective use of PCT in human factors work, whether it
be in direct hands-on situations such as flying a complex modern aircraft,
or in software design, or in database management or ... But any proposal
relating to the exposition, development or (un)successful application
of PCT, or the contrast of PCT to other approaches, will be happily
considered. One-page notes are permissible. The criterion for selection
will be quality for length, not fit-to-a-mould.
The
editor has opened the door to articles of a wide variety of
kinds, not just man-machine issues. Martin, how would a deadline
of next August 1st sound?
I'd prefer to have first drafts available for circulation among the authors
before the Durango meeting, so that the final papers can have substantial
cross referencing. That means that we should shoot for 1 July rather
than 1 August. But for final drafts to be sent to external referees, given
summer holidays and the like, I think 1 October would be easier to satisfy.
I haven't seriously given the question of deadlines much thought. The
Durango meeting has always seemed to me to be a good staging point, at
which the overall form of the issue should be more or less crystallized.
Proposals, please!
Martin