Page 31 - Analog Circuit Design Art, Science, and Personalities

Analogs Yesterday, Today, and Tomorrow

between the two ideologies referring for example to instruments for interplanetary
exploration. It is preferred here to avoid an opinion of judgment on this very impor-
tant issue, but it is suggested that others similarly withhold judgment. At all costs
we must not go down the wrong road. There are quite powerful and rational and
experienced brains in which the reliability vote would be cast for analog, or at least
against the exclusion of continuous variability. We must cooperate in a dispassionate
but devoted study to determine the likeliest facts and fancies in this affair. If one
believes that Nature is ahead in reliability, and there would appear to be justification
for this belief in recognition of the redundancy, repairability, and adaptability of
animal organisms, then conclusions may follow which are based on how one views
such organisms. It has been standard practice to view the details of animal nervous
systems as evidence that they are digital, but there are major reasons to question
this.' The central nervous system itself seems digital to digital men, and analog to
analog men. If it is both, then it is more intimately and profoundly intermingled
hybrid than any of the artificial structures which have come to light. One thing is
pretty sure, and that is that the brain builds models. We are in good company.
Back on reliability, at least in the sense of predictability, there is a duality to be
noted in the relation between analog and digital techniques. If one must predictably
manipulate an imperfectly accessible entity, he may proceed by arranging a discrete
set of states for that entity, then transmit a prearranged number of command signals
to it. Alternatively, with a nonquantitized feedback indicating the state of the entity.
one commands changes outwardly by whatever means until the desired state is
shown to have been attained. What one achieves by quantitizing, the other does by
feedback. This is oversimplified, and does not immediately enable an evaluation of
reliability. For the moment, it is only a point in (practical) philosophy, but as with
many other continuous/discrete instrumental relations it is reminiscent of the
wave-particle dualism.

Auguries

It has been predicted above that the analog-digital struggle will persist, and this
will mean some wear and tear as the proponents contend, but on balance such con-
tention will probably be beneficial since it will assure that the maximum potential
of each technique will be realized. As to some mixtures, all the obvious ones will
soon be seen somewhere. More intimate mixtures, which might offer something
approaching universal applicability, will depend on the appearance of new instru-
mental tools. But also note that urgent needs provide as potent a force for develop-
ment as does the availability of new and startling techniques. Hasty prediction from
either angle would be hazardous; certainly anything specific on our part would be
irresponsible as well as foolhardy. There do seem to be possibilities, however, in
recognition of the ability of continuous analog instruments to operate quickly and
smoothly in closing feedback loops, plus the abitrary accuracy and permanency of
discrete processes. Graphical computation may give a clue of sorts here, since any-
one who deals with geometrical plots is prone to appeal alternately to continuous
criteria and to numerical coincidences in calibration. Coordinates in general may
have both of these meanings simultaneously. Are they any better than we are?
As to analogs themselves, it is evident that some forms of instrument, though not
all, will become progressively smaller and handier in solid slate incarnations. It is

1. R.W. Jones, Science 140,3566 (1963). See also the companion article hy J. S. Gray.

12