Chapter 10 Chapter 9: The First Industrial Revolution and the Second Industrial Revolution

The first few chapters of this book dealt with the problem of man as a communicating organism.But, as we have seen, a machine can also be a communication organism.This chapter will discuss the impact of the communication characteristics of humans and machines, and will also try to determine the future direction of the development of machines and the resulting impact on human society. At one time in history, machines have impacted human culture and had a great impact on it.The impact of machines on human culture is called the Industrial Revolution, and the machines involved at that time were all used as substitutes for human muscles.In order to examine the current crisis of what we shall call the Second Industrial Revolution, it may be wise to discuss the history of the last crisis as something to draw upon.

The first industrial revolution had its roots in the intellectual turmoil of the eighteenth century, when the scientific method of Newton and Huygens was well developed but hardly applicable beyond astronomy.By that time, however, all progressive scientists had recognized that these new technologies were signals of profound changes to come in other branches of science.The two fields that were first influenced by Newton's spirit were navigation and watchmaking. Seamanship is an ancient art, but until the 1730s it suffered from a marked weakness.The problem of determining latitude has always been straightforward, even in ancient Greece.It's just a matter of determining the height of the celestial pole.Taking the Polaris as the actual celestial pole, the height can be roughly determined. If the center of the apparent circumference of the Polaris is further calculated, the latitude can be accurately obtained.In contrast, the problem of determining longitude has always been more difficult.Since there was no geodetic method at that time, this problem could only be solved by comparing the local time with a standard time (such as Greenwich Mean Time).For this reason, it is necessary to carry a timepiece calibrated according to GMT when sailing, or to find a certain celestial body other than the sun as a timepiece to replace it.

Until these two methods were employed by the practical navigator, seamanship was very limited.Generally, he sails along the coast until he reaches the latitude he wants to reach.Then he set out a course parallel to the parallel parallel to the east or west until he came across land.He could not tell how far he had traveled along his course, except by an approximate estimate of the distance, a question which, however, was of the utmost importance, for he should not be ignorant of whether the ship was approaching a dangerous shore or not.When approaching land, the ship sailed along the coast until it reached the intended place.It will be seen that, under these circumstances, every voyage involves great peril.But nonetheless, it was the nautical model for many centuries.

The routes of Columbus, the routes of the Silver Fleet, and the routes of the galleons of Acapulco were opened up in this way. This slow and dangerous method of navigation was not satisfactory to the Admiralties of various countries in the eighteenth century. First of all, the overseas interests of Britain and France are different from those of Spain. They are all distributed in high latitude areas. In this regard, it is obvious that direct flights in a large circle are superior to east-west flights along latitudes.Secondly, there was a fierce competition between the two northern powers for supremacy of the seas, so the superiority of seamanship was of serious significance.It is no wonder that the governments of both countries have offered huge rewards for accurate methods of measuring longitude.

The history of winning these prizes is complex and uninstructive.Many talented people were deprived of the victories they deserved, and Gu's family went bankrupt.In the end, both countries awarded cash prizes for two completely different achievements.One was the design of an accurate nautical chronometer, a well-made and accurately-traveled clock capable of telling time accurately to within a few seconds during a voyage in which a ship was subjected to constant and severe vibrations.The other was the fabrication of a precise numerical table of the moon's motion, which enabled navigators to use it as a clock to check the apparent motion of the sun.These two methods dominated the whole of navigation until the recent invention of radio and radar.

Thus, in the industrial revolution, the vanguard of artisans included two groups of people, the watchmakers, who used Newton's new mathematics to design pendulums and balance wheels, and the artisans who made optical instruments, who built sextants. and binoculars. These two industries have a lot in common.They both make exact circles and exact straight lines, and divide them into degrees or inches.Their tools are lathes and indexing machines.These mechanical tools for precision work were the forerunners of our present machine-building industry. It's worth recalling: Every tool has its own family tree, descended from the tools that made it.Through a very clear historical chain composed of intermediary tools, the turning bed of the watchmaker in the eighteenth century produced the huge turning turning bed of today.Perhaps this chain can be shortened, some unnecessary stages omitted, but it must have a minimum length.In making a gigantic revolving lathe, it is evident that we cannot use hands to cast the metal, and to place the castings on the machine for processing, let alone as the power required to machine them.

These jobs have to be done by machines, which in turn have to be built from other machines.Only through these many stages can one go back to the original hand-cranked or foot-operated turning bed of the eighteenth century. It is therefore quite natural that those who are going to make new inventions, if they are not watchmakers, are makers of scientific instruments, or who call upon the aid of artisans of these trades.Watt, for example, was a scientific instrument maker.But even a man like Watt had to wait for the time to come before he could apply the sophistication of clockmaking to a larger enterprise.To prove this, we must remember that, as I have said before, Watt's criterion of the fit of a piston to a cylinder was to see whether a thin sixpence of copper could fit just between the two.

We must therefore see navigation and the instruments it required as the trigger for the industrial revolution that preceded a full-blown industrial revolution.The full industrial revolution started with the invention of the steam engine.The first form of the steam engine was the crude and uneconomical Newcomen engine, which was used to pump water from mines.In the middle of the eighteenth century, some people tried to use a steam engine to generate power, but failed. Their method was to use a steam engine to pump water into a high-level reservoir, and then use the falling water to drive the water wheel.This clumsy device fell into disuse when the perfect Watt-machine was introduced, and Watt-machines were at once employed in factories for as many purposes as they were used for pumping mines.At the end of the eighteenth century, steam engines had been widely used in industry, and the appearance of steamboats on rivers and steam traction locomotives on land was just around the corner.

The place where the steam engine was first used was to replace the cruelest form of human or animal labor: pumping water out of mines.In the best cases, the work was performed by livestock, crude machinery pushed by horses.In the worst cases, such as the silver mines of New Spain, the work was done with slave labor.As long as the mine doesn't collapse, the work will be endless without stopping.The replacement of this slave labor by the steam engine should certainly be regarded as a great humanitarian advance. But slaves did more than pump water from mines.They also towed ships full of cargo upstream.

The second great victory of the steam engine was the invention of the steamboat, especially the river steamboat.As for the seas, the steam engine was for many years a dubious addition to the sails of seagoing ships; yet it was the use of steam engines for transport on the Mississippi that opened up the American hinterland.Like the steamboat, the steam locomotive began to appear as a means of hauling heavy loads where it is being phased out today. The second place where the industrial revolution appeared is the textile industry. The revolution in this field may be later than the heavy miner's labor, but it is carried out at the same time as the revolution in the transportation industry.The textile industry at that time was already full of problems.Even before the invention of the mechanical spindle and loom, the working conditions of weavers left much to be desired.What they were able to accomplish lagged far behind demand at the time.Therefore, it is difficult to imagine that mechanization will make the working conditions of textile workers worse, but mechanization has indeed made their working conditions worse.

The beginning of the development of textile machines can be traced back to the era of steam engines.Hand-operated knitting frames have been around since the days of Queen Elizabeth.For the first time the spinning machine became necessary to carry the warp for the handloom.It was not until the beginning of the nineteenth century that the textile industry realized full-scale mechanization, including both spinning and weaving.The first looms were operated by hand, but horsepower and water power were soon used.Unlike the Newcomen machine, part of what prompted the development of the Watt machine was the desire to provide the textile industry with the power to turn the machine. The textile industry provides a model for almost the entire process of industrial mechanization.Socially, the mechanization of the textile industry initiated the transition of workers from the home to the factory and from the countryside to the city.The labor exploitation of child labor and female labor at that time was so brutal if we forgot about the diamond mines in South Africa and the new industrialization of China and India and the general situation of laborers on plantations in almost every country , which we cannot imagine today.This happened primarily due to the fact that new technologies brought new responsibilities to people, yet there were no laws and regulations to oversee them.However, there is one situation in which the technical significance is greater than the moral significance.By this I mean that many of the disastrous consequences and situations of the early days of the Industrial Revolution were not all due to the immorality or wrongdoing of the people concerned at the time, but to a number of technical features which were Unavoidable in the early stages of industrialization, they more or less disappeared in the subsequent history of technological development.These features that determined the direction of technological development at the beginning of the Industrial Revolution were in the very nature of early steam power and its method of delivery.The fuel used by steam engines was very uneconomical by modern standards, but this problem is not so important when one considers the fact that there were no newer types of steam engines to compete with them.But even with this steam-engine it is much more economical to use it on a large scale than on a small scale.Compared with prime movers, textile machines (whether looms or spinning machines) are light machines that consume little power.It is therefore necessary, for economical purposes, to concentrate these machines in one great factory, and to have many looms and spindles driven by one steam engine. Back then, the only effective means of delivering power were mechanical tools.One of the earliest tools used was the transmission shaft system with interlocking belts and pulleys.Even as late as my childhood, the typical look of a factory was that of a large shed, with long transmission shafts suspended from beams, and belts connecting the pulleys to the various machines. Such factories still exist, although in many cases they have been replaced by modern enterprises in which the machines are individually driven by electricity. In fact, the second scenario is typical at the moment.The machinist industry has been completely transformed.This is an important fact that concerns the entire history of invention.It was these artisans who made the inventions of mechanics and other new trades of the machine age that laid the foundation for our patent system.In fact, the mechanical connection of machines involves very serious difficulties which cannot be easily summed up in a simple mathematical formula. First of all, if the long shafting is not well assembled row by row, then simple coupling methods (for example, universal coupling or parallel coupling) must be used to ensure a certain degree of convenience in the work.Second, the power consumed to support the long bearings of these transmission shafts is enormous.In a machine, the rotating parts and the advancing and retreating parts have to be subject to the same requirement, that is, the same stability is required; these parts are also subject to the requirement of reducing the number of bearings as much as possible, so as to reduce power consumption and obtain production efficiency. simplify.These requirements were not easily fulfilled by general formulas, and this provided ample opportunity for the inventiveness and innovative skills of the old-fashioned artisans. It is due to this fact that the change in engineering technology from mechanical to electrical linkages has had such a huge impact.The distribution of power provided by electric motors makes it very convenient for us to make small motors for each machine for its own use. The transmission loss of the factory circuit is relatively low, and the efficiency of the motor is relatively high.The connection of the motor to its circuit does not have to be fixed, nor does it have to consist of many parts.At present, considerations of the convenience of transportation and equipment may oblige us, as usual, to combine the different machines of an industrial process in one factory; but the necessity of connecting all the machines to a single source of power It is no longer an important reason for the concentration of locations. In other words, we're in a situation where we're going back to a cottage industry that can be built anywhere. I do not want to maintain that the necessity of mechanical conveyance is the sole cause of those warehouse-like factories and the resulting moral corruption.In fact, the establishment of the factory system preceded the establishment of the machine system, and its purpose was to establish discipline in the undisciplined cottage industry of individual workers, so that the product could maintain a certain standard.It is true that these non-mechanized factories were quickly replaced by mechanized factories, and the social consequences of the sharp increase in the urban population and the sharp decrease in the rural population may be caused by the mechanization of factories.Further, even if we had small horsepower motors at the outset, even if such motors could increase the productivity of domestic workers, it is difficult to conclude that large-scale production could be established in those domestic industries, such as textiles. Organization and discipline required. If we wish to proceed in this way, a single machine may contain several engines, each dedicated to powering a particular part.This relieves the designer much of the burden of inventing those mechanical designs which he would otherwise be obliged to invent.In the design of the motor, if the connection problem of each part is only considered, there will not be too many difficulties that are difficult to deal with using simple mathematical formulas and mathematical solutions.The inventor of the conveyor system has now been replaced by the circuit calculator.This is an example of how the art of invention depends on the actual conditions of the time. When electric motors were first used in industry between the 1950s and 1970s, it was initially thought of as just another device that made the industrial technology of the day work.It may not have been foreseen at the time that its final result would be a new concept of the factory. Besides the electric motor, another great electrical invention is the vacuum tube, which has a similar history.Before the invention of the vacuum tube, we needed many discrete mechanisms to regulate high power systems.In fact, most regulating mechanisms themselves require considerable power.There are individual exceptions, but only in special areas, such as the steering of ships. As recently as 1915, I crossed the Atlantic on an old American steamer.It was a transitional steamship, still with sails, and a sprig on its pointed prow.On the deck not far aft of the main superstructure was mounted a gigantic machinery consisting of four or five wheels six feet in diameter with handles.These wheels are intended to be used to steer the ship should the autopilot fail.In a storm, it would take the full strength of a dozen men to keep the big ship on its course. This is not the usual way of steering a ship, but a substitute in an emergency, or as sailors call it, a "back-up wheel."In normal steering, the ship has a steering gear, which converts the small effort of the helmsman at the helm into the motion of a large and heavy rudder.Thus, some progress has been made in solving the problem of force or torque amplification, even on a purely mechanical basis.But even so, this solution of the amplification problem was not at that time capable of maintaining a very large difference between the input and output quantities, and it did not manifest itself in instruments of the neat general type. The most ingenious general-purpose instrument for amplifying small power to high power is the vacuum or electron tube.Its history is interesting, but it would take too long to discuss it here.However, it is interesting to recall the fact that the electron tube was Edison's greatest scientific discovery, and perhaps his only one not listed as an invention. Edison noticed that if an electrode is placed in the lamp and it has a positive potential to the filament, when the filament is hot, there will be a current flowing between the electrode and the filament, but not vice versa.After a series of inventions by others, this discovery led to a method of controlling high currents with small voltages, which was more effective than any previous method.This method is the basis of the modern radio industry, but the tube is also a tool that has found widespread use in many new sectors of industry.Therefore, the process of controlling high power no longer needs to use the kind of machinery in which the important control parts also need to use the same high power to work.It is entirely possible for us to make a certain behavior pattern that requires very low power, even far lower than the power required by those behavior patterns in ordinary radio devices, and then use a series of amplifier tubes to go through such an instrument Take control of a heavy machine, for example, to control a rolling mill.The identification and behavioral patterning to achieve this control is done under the following conditions: the power consumption is insignificant, however the final application of this identification process can reach arbitrarily high power level. It can be seen that this invention can bring about a fundamental change in industrial conditions, which is no less important than the use of small electric motors to transmit and distribute energy.The study of behavior patterns is left to the specific part of the control instrument, where the question of energy saving is trivial.Consequently, the ingenious designs and devices previously employed to ensure that mechanical coupling systems consist of as few components as possible, and those designed to ensure a reduction in friction and loss of motion, have now largely lost their value.The design of machines for which the above-mentioned parts are to be used has passed from the hands of the skilled workshop worker to the scientific researcher in the laboratory; On the one hand, we strive to renovate.The invention of past meanings has been supplanted by the exploitation of certain natural laws.The distance between the laws of nature and their utilization has been shortened a hundredfold. As I said before, when an invention is proposed, it usually takes a long time before people understand its full significance.The full impact of the invention of the airplane on international relations and the human condition of life was not understood until much later.The impact of atomic energy on mankind and its future is yet to be estimated, although many observers insist that it is nothing more than a new weapon, like all old weapons. The same was true of the vacuum tube, which at first was regarded only as an auxiliary tool to improve the old telephone communication technology.Electrical engineers were initially so ignorant of its true value that they for a long time regarded it as a special component in a communication network.This component is used to interface with other traditionally limited so-called inert circuit elements - resistors, capacitors and inductors.It wasn't until the war that engineers began to use vacuum tubes freely, connecting them where they were needed, just as they used to use the above three inert components. Vacuum tubes were originally used as replacements for existing components in long-distance telephone lines and wireless telegraphy.But it wasn't long before wireless telephony had advanced to the level of wireless telegraphy and radio broadcasting became a reality, and people's uses for it became clear.We should not, however, be blinded by the fact that this great triumph of inventive thought is at present mainly at the service of "soap operas" and vulgar singers, from seeing what people have done in the course of this invention. The excellent work done, does not see its great possibilities in spreading culture, although these possibilities have been misused as a national drug exhibition window. Although vacuum tubes have already tried their hand in the communication industry, the boundaries and scope of this sector of the communication industry have not been fully understood for a long time.The vacuum tube and its sister invention, the photoelectric cell, have been used piecemeal to inspect industrial products; for example, to adjust the thickness of paper coming out of a paper machine, or to inspect pineapples in a pineapple can color etc.These applications have not hitherto resulted in a reasonable new technology, and engineers have had in mind the connection of the vacuum tube to its other function, that of communication. All this changed during the war.One of the few gains we have gained from the great war has been the rapid growth of the enterprise of invention, stimulated by objective necessity and unrestricted funding, and, in particular, by the new growth of the industrial research department. In the early days of the war our greatest task was to keep Britain from being overwhelmed by an extremely heavy air attack.Anti-aircraft guns were therefore one of the first objects of scientific research in wartime, especially when anti-aircraft guns were combined with radar devices or ultra-high-frequency Hertzian wave devices for reconnaissance aircraft.Apart from the various inventions of radar itself, radar technology is used in the same way as old radio technology.Therefore, we naturally regard radar as a branch of communication theory. In addition to finding the plane with radar, it was also necessary to shoot it down.This involves the control of artillery fire.The speed of the aircraft is extremely fast, so it is necessary to use machines to calculate various parameters of the trajectory of the anti-aircraft shells.It is also necessary to make the predictive machine itself have those communication functions that are originally performed by humans.Thus, the problem of control of anti-aircraft fire acquainted a new generation of engineers with the notion of communications aimed at machines rather than at men.In our chapter on languages ​​we have already mentioned another field, that of automatic hydroelectric power stations, where the idea has long been familiar to a certain number of engineers. On the eve of World War II, other uses for vacuum tubes were discovered, all directly related to machines and not to human labor.One of the most widely used is in the computer.In computers, the idea of ​​a large computer like that developed by V. Bush was originally purely mechanical in nature.The integration is done with rolling discs which meet in a frictional manner; the exchange of output and input between the discs is done with a series of old-fashioned shafts and gears. These early computer ideas, in their origin, predate Bush's work by a long time.In some ways it dates back to the work of Babbage in the early nineteenth century.Burbage already had the astonishing idea of ​​a modern computer, but all he could do was mechanical, which was far from satisfying his ambitions.The first difficulty he encountered and could not overcome was that the movement of long series of gears required considerable power, so that the output force and torque soon became too small to move the machine. other parts.Bush saw this difficulty too, and overcame it in a very ingenious way.In addition to electric amplifiers made of vacuum tubes and similar devices, there are several mechanical torque amplifiers, such as you often see when unloading ships.The cargo is lifted by stevedores on the wharf by hanging it on the beam of a crane or the drum of a winch.Using this method, the mechanical force applied by the stevedores is amplified by a proportionality factor which increases rapidly with the contact angle between the sling and the drum.Therefore, one person can lift many tons of goods. Such devices are basically force amplifiers or torque amplifiers.By means of an ingenious design, Bush incorporated these mechanical amplifiers into the various stages of the computer, and thereby was able to perform efficiently the kind of work which for Bibbage could only dream of. I became interested in problems of partial differential equations early in my career at Bush, before there were any high-speed automatic controls in factories.Bush's work involved ordinary differential equations, in which the independent variable was time, and the machine simulated the process of the phenomenon it analyzed in the course of time, although at different speeds.In partial differential equations, instead of time variables are quantities that vary in space.I once suggested to Bush that, since television scanning technology was developing rapidly at the time, we must consider this technology ourselves, and use it to describe multivariate, say, spatial variables as opposed to univariate time .The computer thus designed must work extremely fast, so that, in my mind, mechanical processes cannot be considered, which forces us to still consider electronic processes. Furthermore, in such a machine all data must be written, read or erased at a rate commensurate with the other actions of the machine.In addition to including an arithmetic mechanism, such a machine must also include a logic mechanism capable of solving programming problems on a purely logical and automatic basis.The idea of ​​programming, already familiar in the factory from the work of Taylor and Gilbreths on the calibration of man-hours, was ripe for transfer to machines.The solution of this problem presents considerable difficulties in detail, but not great difficulties in principle.I therefore believed as early as 1940 that the establishment of an automated factory was in sight and told Bush about it.Around the publication of the first edition of this book, automation developed, which convinced me that my judgment was correct, and that this development would be one of the huge factors determining the future social and technical life. The fuse of the industrial revolution. Among the early forms of the Bush differential analyzer was an analyzer that performed all the important functions of amplification.Electricity is used only to deliver energy to the engine to start the whole machine.The computer in this state is an intermediate and transitional computer.It was soon clear that electric amplifiers connected by wires rather than pivots were much cheaper and more flexible than mechanical amplifiers and mechanical connections.Consequently, later forms of the Bush machine used vacuum tube arrangements.These devices were then applied to all computers, whether they are analog computers (which work mainly by measuring physical quantities) or digital computers (which mainly work by counting and arithmetic operations) as they are called today. After the war, these computers developed very rapidly.In most fields where calculations are required, these computers appear to be much faster and more accurate than calculators.Their speed has long since reached such a level that it is impossible to carry out any human intervention in the middle of their calculations.Thus, as we saw in the control instruments for anti-aircraft guns, these computers also required the replacement of human power by machine power.The parts of the machine must use an appropriate language to talk to each other, and they speak to no one, and listen to no one, except in the initial and final stages of the process. This is again one of the arguments in favor of the generalization of the concept of communication to machines, which is agreed upon. In such conversations that take place between the parts of a machine, we often have to know what the machine has said.Here, we have to talk about the feedback principle, which we have discussed before.The steering gear on a boat is an example of the feedback principle, but there are older examples than steering gears.In fact, the governor that regulates the speed of a Watt steam engine is an application of this principle.This governor prevents the locomotive from going too fast when its load is lightened.If the locomotive begins to go too fast, the governor ball will go up by centrifugal action, and the lever will go up with it, thus blocking a portion of the steam from entering the machine.Thus, the tendency to increase the rotational speed causes a tendency to partially compensate, thereby reducing the rotational speed. In 1869, Maxwell made a thorough mathematical analysis of this adjustment method. Here, feedback is used to regulate the speed of the machine.In the ship's steering gear, it is used to adjust the position of the cabin.The driver operates a lightweight transmission system, using chains or hydraulic transmissions to push a cab-mounted member.The distance between this member and the tiller is indicated by a special instrument, and from this distance one controls the amount of steam entering the valve of the steam steering gear, or, in the case of an electric steering gear, the admission of electrical energy.Regardless of the specific method of coupling, the changing profile of the incoming energy is always such that it keeps the components actuated by the tiller and steering wheel in coordinated motion.Therefore, one person at the helm can easily complete the task that an old-fashioned human steering wheel requires a whole team to do with great effort. So far we have only mentioned a few mainly mechanical feedback processes.However, a series of operations of the same structure can also be accomplished by electrical means or even vacuum tubes.These approaches have the potential to become standard methods for designing control instruments in the future. The trend to automate factories and machines has been around for a long time.Except for some special purposes, no one wants to produce screws with ordinary lathes, because the machinist with such lathes must watch the progress of the turning tool and adjust it by hand.Now, a large number of screws can be produced without much human intervention, which is the usual work of ordinary screw machines.While this machine does not exclusively use a feedback process, nor does it exclusively use vacuum tubes, it serves roughly the same purpose.What feedback and vacuum tubes can do is not the piecemeal design of individual automatic devices, but our general guidelines for making automatic machines of all kinds.In this regard, our research on new communication theories plays a catalytic role.Our theory fully considers the possibility of machine-to-machine communication.It is the combination of these circumstances that is now making possible the advent of the new era of automation. The current state of industrial technology includes all the fruits of the first industrial revolution and many of the inventions that we see today as the heralds of the second industrial revolution.As for the question of the strict boundary between the two revolutions, it is too early to talk about it.In terms of potential, the vacuum tube certainly belonged to a revolution different from the Industrial Revolution of the Power Age.However, it is only now that the true significance of the invention of the vacuum tube is fully understood to make it possible for us to lead the present era into a new, second industrial revolution. Until now, we've been talking about the status quo of things.We have only scratched the surface of various aspects of the last industrial revolution.We haven't mentioned airplanes, or bulldozers and other mechanical construction tools, or automobiles, and we haven't even mentioned a tenth of those factors that make modern life so different.But it may be justly said that, except for a considerable number of isolated instances, the Industrial Revolution has hitherto only changed the face of man and beast as sources of power, and has not shown any significant influence on the other functions of man.Today, if a worker uses a pick and a shovel to earn a living, the best he can do is follow a bulldozer and clear the ground.In all work of any importance, a man who has nothing else to sell but his physical strength will sell nothing worth anyone's money. Let us now look at a more complete picture of the age of automation.For example, let's imagine what the car factory of the future will look like, and in particular, imagine the assembly department, the most labor-intensive part of a car factory. First, the operating program will be controlled by something similar to a modern high-speed computer.我在本书认及其他地方常常说到，高速计算机基本上是一部逻辑机，它把不同的命题拿来互相考较，并从它们的结论中作出选择。它能把全部数学归结为一系到纯逻辑任务的运演。如果把这种数学表示体现到机器中，则这种机器使是通常的计算机。但是，这种计算机除了解决通常的数学任务外，还能够担负起给机器传达一系列有关数学演算的指令的逻辑任务。因此，它至少要包括一大堆进行逻辑运算的设备在内，目前的高速计算机事实上正是这样的。 给予这种机器的指令（我这里仍然是谈目前的实际情况提由我们称之为程序带这个地方发出的。爱给机器的命令可以由完全预定的程序带馈进的。机器在执行自己的任务时所遇到的意外情况也可以被用作进一步调节机器自身所制定的新控制带的基础，或者作为修改旧控制带的基础。我已经解释过，我认为这些过程都同学习的过程有关。 也许有人认为，目前计算机的价格太高，无法把它们利用到工业过程中来；而且，制造这些机器的工作过于精细，机器的职能又是多种多样的，因而无法进行大量生产。 这些看法都是不正确的。第一，目前用来进行极复杂的数学工作的大型计算机，其价格大概是数十万美元。即使是这样的价格，一个真正大工厂也不会拒绝采用它作为控制机器的，但是这个价格还是太贵了一些。目前的计算机发展得如此之快，以致实际造出的每部计算机都是新式的。换句话说，在这些显然过高的费用中，大部分都是花在新的设计工作上和制造新的零件上，因为生产这些零件要求有十分精巧的劳动和十分昂贵的设备。因此，如果这些计算机之一在价格和型式方面确定下来了，并且是十架、二十架地采用了，那它的价格是否会超过一万元就很值得怀疑了。一架类似的功率较小的机器虽然不适于解决最困难的计算问题，但却完全适于工厂控制之用的，而这种机器对于任何一种中等规模的生产说来，其价格都可能不超过几千美元。 让我们现在再来考虑一下大量生产计算机的问题。如果大量生产仅仅是指大量生产各种型式的整部机器的话，那末，十分清楚，在相当长的时间以内，我们至多只能进行中等规模的生产。但是，每架机器的零件，基本上都是无数次重复生产的。不管我们考虑的是记忆装置，是逻辑装置，抑是其中的算术运算设备，都是如此。因此，只要有少量机器的生产，实际上就意味着大量零件的生产，因而在经济上也就具有大量生产的优点。 也许还有人认为，机器的专门化必然意味着每件不同的工作需要有一种新的特殊模型的机器。这个看法也是错误的。即便机器的数学部分和逻辑部分所需的操作类型不尽相同，机器总任务的完成也是由程序带至少是由原初的程序带来调节的。给这种机器编制程序带，对于高明的专家说来，是一件很复杂的工作，但它大半是或者完全是一劳永逸的工作，当机器为了用于新的工业装备而有所改变时，这种工作只需部分地改变。因此，花费在这种精巧技术上的费用可以分摊到大量产品上去，并不真的会对机器的采用发生重大的影响。 计算机是自动化工厂的中心，但它决不等于整个工厂。另一方面，它是从那些带有感官性质的仪器那里取得详细的指示的，这些仪器如光电管、测量纸张厚度的电容器、温度计、氢离子浓度计以及现时各仪器公司制造出来的并利用人手来控制工业过程的各种仪器。这些仪器已经制成到这样的地步：能够借助电力把命令传达给遥远的工作站。 为了使这些仪器能够把自己的信息进入自动化的高速计算机中，只需要有一个读数装置，把位置或刻度译成一连串数字的模式就行了。这种装置已经做出来了，无论在原理上或在制造细节上都没有太大的困难。感觉器官的问题不是新问题，而是一个已经有效地解决了的问题。 除了这些感觉器官外，控制系统还必须包括效应器官或作用于外界的构件在内。其中有些效应器的类型已经是大家所熟悉的了，例如，回转阀电动机、电离合器等。有的还需要发明，以便更准确地模仿人手的功能，用作人眼功能的补充。我们对汽车车架进行机械加工时，完全可以在磨光的车架表面上留下几处金属突起作为参考点。为了使工作工具（不论是钻孔机、铆钉钉接器或其他必要的工具准够自动找到这些参考点起见，我们可以采用光电机械，例如用油漆光点引动的光电机械。最后固定下来的位置可以使工具和参考点紧密接触，但又没有紧密到使这些地方遭受破坏的程度。这只是做这类工作的一种方法。任何有能力的工程师都能够再想出成打的其他方法来的。 当然，我们假定这类带有感官性质的仪器不仅可以把工作的初始状况记录下来，而且还可以把以前所有过程的结果记录下来。因此，机器所能完成的反馈操作，除了现今已经完全了解的那种简单类型的反馈操作外，还有由中央控制系统（例如逻辑装置或数学装置）来调节的包括比较复杂的选择过程在内的反馈操作。换句话说，整个控制装置相当于一个具有感觉器官、效应器官和本体感受器的完整动物，而不是相当于一个孤立的脑，象超速计算机那样，其经验和有效性要取决于我们对之参预的程度。 工业中可能采用这些新装置的速度，将因工业部门的不同而有很大的不同。自动机已经广泛地应用到实行流水作业的工业部门中了，例如罐头厂、轧钢厂，尤其是电线厂和白铁制造厂。这类机器同我们这里所讲的可能不完全相同，但其功能大体一样。造纸厂中也常常见到这类机器，同样是采用流水作业法生产。另外一种必须使用自动机的工厂就是在生产中具有太大危险性的工厂，在这种工厂里，大量工人冒着生命危险去操纵机器，而且，这种工厂一出事故就可能很严重，损失很大，所以事故的可能性应当预先得到警告，不应当从现场中某人仓皇作出的判断为根据。如果预先能够考虑到行动方案，那就可以把它记到程序带上，以便按照仪器的读数控制其后的行动。换句话说，这种工厂应当按照一定制度进行工作，就象铁路信号塔的连锁信号和开关工作的制度那样。这种制度已经在炼油厂和其他许多化学企业以及国利用原子能而出现的各种危险物质的处理工作中建立起来了。 我们曾经提到的装配部门就是应用这类技术的部门。在装配部门，例如在化学工厂或流水作业的造纸厂中的装配部门，必须对产品的质量进行某种统计性质的监督。这种监督是靠抽样过程来进行的。这个过程经过瓦尔德（Wald）等人的改进现在已经提出一种所谓连续分析的技术方法了。依据这种方法，抽样不再是集总进行的，而是同生产一道进行的连续过程。因此，凡是能够用相当标准化的技术来完成并可以交给一个不懂得其中所含的逻辑关系的统计员去管理的那些过程，都可以交给计算机去做。换句话说，除了最高水平的工作外，机器既能照顾到生产过程，又能照顾到日常的统计监督工作。 一般说来，工厂都有会计手续，它与生产无关，但是，既然帐簿上出现的数据都是来自机器或装配线，那它们也就能够直接送到计算机中。其他数据可以通过人手运算随时送到计算机中，但是，大部分的书写工作可以用机械方法来完成，只剩下少数的细目，例如对外通讯，才需要由人来担当。但是，即使是对外通讯，也可以大部分变为收取对方进来的穿孔卡片或者是通过极为简易的劳动把它打印到穿孔卡片上的。从这一阶段开始，一切工作都可以由机器去完成。这种机械化的方法同样适用于工业企业图书馆和档案处的绝大部分的工作上面。 换句话说，机器既不偏爱体力劳动，也不偏爱文牍式的劳动。因此，新的工业革命所能渗透进去的领域就会非常广泛，包括执行不太用脑筋的一切劳动在内，情况和上次工业革命在人力的各个方面都有被机器排挤掉的现象极为类似。当然，也会有些行当是新工业革命所不能插手的，因为新的控制机器对于小规模工业会是不经济的，承担不起大量的基本投资，或者，由于这些工业的工作变化多端，以致每一特定的工作几乎都要有一个新的程序带。我不能想象一部代替我们去做判断的那种类型的自动机会被小杂货铺或出租汽车行所采用，虽然我能够清楚地想象到，杂货批发商或汽车制造商会采用它。 农业工人虽然在生产中也会受到自动机的排挤，但由于他所耕作的土地面积的规模、他所种植的作物的多样性和气候条件的特殊性以及他所面临的其他情况等等，他还不至于感受到它的全部压力。但即使在农业中，大规模经营的农场或种植园的资本家也开始愈来愈依赖于摘棉机和锄草机了，例如种小麦的农场主早就用上麦克考尔密克（McCormick）收割机了。凡是可以使用这些机器的地方都不是不可能在一定程度上使用机器来作判断的。 当然，这些新装置的采用与否以及可能被采用的时间主要是经济方面的问题，而在这个问题上，我不是行家。据我粗略的估计，如果没有剧烈的政治变动，或者发生另一次大战的话，那末，新机器需要一、二十年才能占据应有的地位。战争会使这一切在一夜之间发生变化的。假如我们同大国发生战争，例如，同俄国发生战争，以致迫切需要大量的步兵，因而需耍大量的人力时，我们就很难维持我们的工业生产了。在此情况下，采用其他方式来代替人力生产的问题也许就是民族存亡的问题。我们现在在发展自动控制机器的统一体系方面所面临的处境就象我们在1939年发展雷达技术时的处境一样。正是由于“英国战役”这一紧急事变，使得我们有必要大规模地去研究雷达问题，有必要使这个领域的自然发展过程加快起来，从而使它提前几十年成熟；同样，在一次新战争中，代替劳动力的必要性也可能给我们带来同样的影响。由熟练的无线电爱好者和数学家、物理学家这批人员所组成的那支技术队伍，过去曾经很快地变成了熟练的从事雷达设计的电气工程师，今天，他们仍然可以在自动机器的设计方面做出类似工作的。还有，由他们训练出来的新的熟练的一代也正在成长中。 在这些条件下，自动化工厂发展起来的时间很难超过两年左右，这相当于过去使雷达在战场上发挥高度效能所需的时间。在这样一次战争结束时，建立这类工厂所需的“专门技能”就成为人所共知的事情了。那时甚至还有剩余下来的政府所制造的大批设备，可以卖给或交给工业家使用。因此，一次新的战争几乎免不了会在不到五年的时间内掀起一个自动化的高潮。 我曾经谈到这种新的可能的真实性和迫切性。我们从中能够预期到什么样的经济后果和社会后果呢，首先，我们可以预期，那种进行纯粹重复工作的工厂会突然降低劳动力的需要，最后变成完全不需耍。归根到底，这种极其乏味的重复劳动解除以后，也许会带来好处，这是人类文化能够得到充分发展所必需的闲暇时间之来源。但是，它也可以在文化领域里产生毫无价值的和有害的结果，就象目前在无线电广播和电影中所得到的大部分结果一样。 不管怎样，采用新方法的中介时期，特别是，通过一次新战争而可以期望它迅速来临的话，那就会立即出现一个充满灾难性混乱的过渡时期。我们有许多经验来说明工业家对待新工业潜力的态度。他们的全部宣传就是要达到下述的目的：新技术的采用不应当看成是政府的事，而应当交给愿意在这项技术上投资的企业家自由掌握。我们也知道，当事情牵涉到攫取工业中全部能够攫取到的利润，然后让公众也捡到一些残羹剩饭的时候，企业家是很难克制自己的。伐木工业和采矿工业的历史就是如此，这也是我在另一章中叫做传统美国的关于进步的哲学的一个部分。 在这些条件下，工业中采用新机器就会到达这样的程度：只要眼前有利可图，就不管它们在以后可能带来怎样的危害。我们将看到一个和原子能发展的过程相类似的发展过程。原子能被用来制造炸弹。这就妨碍了未来利用原子能以代替石油和煤的蕴藏量，而这却是我们极其必要的潜力，因为石油和煤在几百年内（如果不是几十年的话）即将耗竭。要注意，原子弹的生产是不同动力生产作竞争的。 让我们记住，不管我们对于自动机之有无感情问题采取什么样的想法，它在经济上完全和奴隶的劳动相当。而任何一种同奴隶劳动竞争的劳动都必须接受奴隶劳动的经济条件的。十分清楚，自动机的采用会带来失业现象，它同目前的工业萧条甚至三十年代的危机相较，后者只不过是儿戏而已。这种危机会给许多工业部门带来危害，甚至也可能给那些利用新潜力的工业部门带来危害。另一方面，我们的工业传统决不妨碍工业家去攫取迅速取得而又稳当可靠的利润，并且在他个人行将破产之前溜之大吉。 因此，新工业革命是一把双刃刀，它可以用来为人类造福，但是，仅当人类生存的时间足够长时，我们才有可能进入这个为人类造福的时期。新工业革命也可以毁灭人类，如果我们不去理智地利用它，它就有可能很快地发展到这个地步的。然而，目前已经呈现出一些有希望的迹象。自从本书初版发行以来，我曾经参加过两次大型的实业家代表会议，我很高兴地看到，绝大部分的与会者已经意识到新技术给社会带来的威胁，已经意识到自己在经营管理上应尽的社会义务，那就是要关心利用新技术来为人类造福，减少人的劳动时间，丰富人的精神生活，而不是仅仅为了获得利润和把机器当作新的偶像来崇拜。我们面前还有许多危险，但是善良愿望的种子也在生根发芽，所以我现在不象本书初版的时候那样地感到完全悲观失望了。