ELECTRONICS — ZWORYKIN 239 



the computation process. The human operator only enters in the 

 setting up of the machine for a given calcuhation and the reading of 

 the final results. 



To perform its function effectively the electronic digital computer 

 hence requires, in addition to a group of electronic computing units, a 

 set of "fast" electronic memories, in which an intermediate result can 

 be stored until needed for a later operation, as well as a command sys- 

 tem on which the complete sequence of operations to be performed by 

 the machine may be recorded. Additional "slow" memories, in the 

 form of magnetic or punched tape, can be employed, furthermore, for 

 the storage of large quantities of information, such as tables of func- 

 tions, which may be fed into the fast electronic memories on order. 



Much ingenuity has been expended in devising satisfactory elec- 

 tronic memories. The essential feature of all electronic memory de- 

 vices is the presence of elements with two alternative stable states 

 which are maintained indefinitely unless subjected to new electrical 

 "writing" impulses. The Eccles-Jordan trigger circuit, containing 

 two vacuum tubes so connected that, in equilibrium, one tube conducts 

 while the other is cut off, constitutes such a single element of memory. 

 More recent devices, such as the cathode-ray tube memory of F. C. 

 Williams (3) and the SB-256 tube of Rajchman (4) by comparison, 

 utilize the properties of the secondary emission of insulators to pro- 

 vide several hundred such elements (corresponding to the storage of 

 an equal number of binary digits) in a single envelope. 



The last few years have seen great advances in the magnitude, scope, 

 and organization of digital computing machines. These will un- 

 questionably continue and further expand the role of electronic meth- 

 ods in science and in our economic organization. It may well be, 

 however, that electronic computation will achieve its greatest triumphs 

 through the integration of analogue computing devices, with their 

 ability to short-circuit extensive numerical calculations in specific 

 applications, in digital computing systems. In fact, it seems scarcely 

 thinkable that effective headway can be made with extremely complex 

 problems such as are encountered in weather prediction unless all 

 available techniques are combined to best advantage. 



Possibly the most striking demonstration of the power of electronics 

 to the average man has occurred in the field of television. The de- 

 velopment of this new industry has far outstripped the predictions of 

 the most extravagant prophets of a few years ago. It requires little 

 imagination today to foresee for the television set a universality com- 

 parable to that of the kitchen range or the refrigerator. Numerous 

 articles have dealt with the probable effect of this new factor on 

 education, social patterns, and the economy, and I shall make no at- 

 tempt to add anything to their conclusions. Teclmically, also, the 



