>So far these issues haven't been discussed very widely, and we would
>be very interested to know what people think.
...these issues have been discussed more widely than you think...
I've read the Cherniak paper and have visited your web site... and wish
to comment a bit...
my impression is that Cherniak is trying to make a case with respect to
something that he picked up on outside of his own work... I tend to
agree with the position youve taken with respect to what's in the
Cherniak paper, but not with respect to the larger considerations...
...a strong case can be made for "optimal placement"... and a strong
case can be made for minimal circuit lengths... but what's minimized is
not 3D circuit lengths, but, rather, 4D circuit lengths... (for those
who have the "Automation of Knowing..." ms, as I discussed in another
online forum several years back, AoK's use of 5D stuff has now been
reworked as 4D 3 space, and 1 "interconnectedness", demensions that are
used to describe "momentary" "energy" "states"...
...I don't know about the horse example you use at your web site, but I
agree that it's not possible to evaluate Cherniak's C. Elegans
findings... I have worked things through re the human CNS... it's my
position that it's a straight-forward, if a bit arduous (because
there's a =lot= that needs to be integrated), task to verify that CNS
optimization is maximized if what I refer to as the 4D neural circuit
lengths are minimized... (note, the only thing that this says about
=individual= nervous system instances is that, if they are to be
competitive, they must approach such minimal 4D wiring... and to the
degree that they do so, they are, in fact, competitive...)
...the "np-completeness" stuff does not hold with respect to the CNS
because it can be demonstrated that everything that occurs within CNSs
reduces to the one thing that is what's described by the 2nd Law of
Thermodynamics (WDB2T)... the nervous system problem can be solved
all-at-once... all-of-a-piece... one does not have to check off every
node... ken
