I read with interest the recent discussions regarding the Optacon
from Dr. Bliss, Mr. Cranmer and Mr. Roderick. In general, I agree
with Dr. Bliss and strongly disagree with Mr. Cranmer and Mr.
Roderick.
I think that Dr. Bliss is correct in asserting that an analog
threshhold control or something very much like it is of immense
value when using the Optacon. As he says, one can "tease" out
information by judicious use of such a control that otherwise
might remain hidden. I also think that he may well be correct in
asserting that the larger array as seen on the Optacon Model R1D
is of value when viewing printed text in the English language. As
I stated in a previous message, however, of more concern to me
than either of these considerations was the fact that the Optacon
II's array did not respond quickly enough to a changing camera
image (as, for example, when rapidly scanning print or changing
from one printed line to the next). Under these conditions, the
array became extremely "mushy" -- a fact which I found very
disconcerting. The aray on the Optacon R1D responded as quickly
as I could move the camera. Useful information concerning format
can be gleaned at high camera speeds even when one cannot read
individual letters or words.
I strongly beg to differ with Mr. Cranmer. To my mind, signal
processing is exactly what is *not* needed for a new Optacon. The
human brain, given adequate tools, makes a fine signal processor
and I would prefer to let it do its thing unmediated by machine-
processing of an image -- processing which I would have little
control over. It was signal processing (albeit in a primitive
form) that was the downfall of the Optacon II. Let's get the
computers out of the loop and go back to analog machines!
Furthermore, I do not believe that an array designed for multiple
fingers would prove useful. Although Braille can be read with
mutliple fingers on each hand (I do not do so but some like it),
Braille has the distinct advantage of uniform character size and
spacing -- something print lacks. It would involve signal-
processing (which I've excoriated above) to make sure that
distinct print characters appeared on each finger - something I
would find essential to increase print reading speed substantially
over that currently possible with the OPtacon.
Mr. Roderick's idea of having pins simply protrude as opposed to
vibrating is an interesting one. However, I suspect that the
changing image would make less sense to the fingers under these
conditions than it does conveyed by vibrating pins. Furthermore,
processing a full line of text would require considerable image
analysis-- analysis that's not possible with the low resolutio of
the Optacon camera (remember the ill-fated project to make the
Optacon talk?). Also, many images one looks at with an Optacon do
not consist of characters marching uniformly across the page.
Flow charts are one example of this.
No, I stick to my original contension: the Optacon that should be
built is the R1D. The only change I would make would be to
incorporate the Optacon II's external (detachable) battery pack.
Otherwise, give me the R1D! Better yet, set up a facility that
will service those now in existence!
Mike Freeman
This archive was generated by hypermail 2b29 : Sun Dec 02 2012 - 01:30:03 PST