|
[ Pobierz całość w formacie PDF ]
.For the obvious question where does it all start? is readilytems.It is so common that we usually take it for granted, but he answered.As biologists, our fundamental fact (S.10/3) is that thewho would learn how to regulate the very large system must earth has now existed for a long time, that selection has actedbecome fully aware of it.throughout this time, and that selection favours the appearance cTo get some idea of how much the world we live in shows regulators (S.10/5).These facts alone are sufficient to account forreducibility, compare its ordinary behaviour with what would the presence on the earth today of many good regulators.And n~happen if, suddenly, the reducibility were lost, i.e.if every varia- further explanation is necessary if it should be found that some oble had an effect, immediate or delayed, on every other variable.these regulators have as goal the bringing of some mechanism toThe turning over of a page of this book, instead of being just that standard form, even if the standard form is that of a regulatorand nothing more, might cause the lights to change, the table to (with goal, of course, distinct from that of the first).The scientiststart moving, the clock to change its rate, and so on throughout the would merely be mildly curious as to why something that couldroom.Were the world really to be irreducible, regulation would be done directly, in one stage, is actually done indirectly, in two.be so difficult as to be impossible, and no organised form of life We can thus answer this section s question by saying that a reg-could persist (S.7/17).ulator can be selected from some general set of mechanisms (man,The subject must be left now, but what was said in Design & on non- regulatory) only by being either the survivor of some process Iterated systems , and in the chapters that followed, expands the of natural selection or by being made (another process of selectionthesis.Meanwhile we can draw the conclusion that if a responsi- by another regulator.ble entity &! (S.13/10) is to design (i.e.select) a machine to act asregulator to a very large system, so that the regulator itself is13/23.Is not this making of the desired regulator by two stagessomewhat large, the achieving of the necessary selection within awasteful? That it should be arrived at in two stages suggests thatreasonably short time is likely to depend much on whether thethe problem of getting a regulator always has to be solved beforeregulator can be made in reducible form.it can be tackled!Again, what does this imply when the very large system to be13/22.Whence the Regulator? Now at last we can answer the regulated is the social and economic world and the responsiblequestion that has been latent throughout Part III: how is the entity &! is some set, of sociologists perhaps, whose capacity, as adesired regulator to be brought into being? The question was regulator, is limited to that available to the members of the speciesraised in S.13/10, but since then we have explored a variety of top- Homo? Does this imply that no advance in regulation is possibleics, which had to be discussed before the threads could be pulled (for the regulator will have to be built by members of the species)?together.Let us now survey the position.It does not; for when regulation is achieved in stages when aThe process of arriving eventually at a particular machine with regulator R1 acts so as to bring into existence a regulator R2 thedesired properties implies selection, and it also implies that the capacity of R2 is not bounded by that of R1.The possibility arises262 263 AN INTRODUCTION TO CYBERNETICSthat R2 may be of capacity greater than R1, so that an amplificationoccurs.This possibility is studied in the next chapter, where weshall see that, apart from being necessarily wasteful, the methodof regulation by stages opens up some remarkable possibilities.264 Chapter 14AMPLIFYING REGULATION14/1.What is an amplifier? An amplifier, in general, is a devicethat, if given a little of something, will emit a lot of it.A soundamplifier, if given a little sound (into a microphone) will emit alot of sound.A power-amplifier, such as the one described inS.12/21, if given a little power (enough to move L) will emit a lotof power (from H).And a money-amplifier would be a devicethat, if given a little money, would emit a lot.Such devices work by having available a generous reservoir ofwhat is to be emitted, and then using the input to act as controllerto the flow from the reservoir.Rarely an amplifier acts by directlymagnifying the input, as does the cine-projectionist s lens; butmore commonly it works by supplementation.Thus thepower-amplifier has some source that will provide power abun-dantly (the compressed air at A in Fig.12/21/1), and it is thissource that provides most of the power in the output, the inputcontributing little or nothing towards the output.Similarly, thework performed by the cranedriver on the control-handle doesnothing directly towards lifting the main weight, for the whole ofhis work is expended in moving electrical or other switch gear.It will be seen that in the power amplifier (e.g.that of Fig.12/21/1) the whole process that of lifting a heavy weight at H, by aforce at L goes in two stages, by two coupled systems.It is thisseparation into two stages that makes power-amplification possi-ble, for otherwise, i.e.in one stage, the law of conservation ofenergy would make any simple and direct amplification of powerimpossible.Stage 1 consists of the movement, by the operator, ofthe point L against the friction at K and the pressure at V; over thisstage energy, or power, is conserved strictly.Stage 2 consists of themovement of compressed air into or out of B and the lifting of P,G and H; over this stage, also, energy is conserved; for the energyused when the weight at H is lifted is derived from the expansionof the compressed air.Thus the whole system can be regarded ascomposed of two systems, within each of which energy is con-served strictly, and so coupled that forces of 0, 1, 2& dynes at Lcorrespond respectively to forces of 0, 1000, 2000,& dynes (or265 AN INTRODUCTION TO CYBERNETICS AMPLIFYING REGULATIONsome other multiple) at H.It is the division into two stages that Obviously, both points of view are right.Designers of cranesenables a power-amplifier to be built in spite of the law of conser- should be well aware that they are not really amplifiers, but thevation of energy, the point being that the energy supplied to the users of them should think of them as if they were.input in stage 1 can be supplemented to give the output in stage 2.Sometimes the proportionality is important, as in the radio 14/3.We can now see how we should view the question of ampli-amplifier.Then the machine has to be made so that the ratio has fying regulation.During the designing (in this chapter) we shallthe same value all along the scale.In other cases the exact value have to be clearly aware that the designer is really achieving onlyof the ratio is of little importance, as in the crane, the essential a supplementation, by robbing some easily available and abun-point in it being that the input values shall all be within some dant source of it
[ Pobierz całość w formacie PDF ]
zanotowane.pldoc.pisz.plpdf.pisz.plhanula1950.keep.pl
|