CODE MAKING MACHINE

Southland Times, Issue 18968, 15 June 1923, Page 15

CODE MAKING MACHINE

WRITES, TRANSMITS, AND DECODES. A WONDERFUL CONTRIVANCE. A machine which looks like a combination typewriter and phonograph, and which creates, transmits, and decodes in automatic code capable of more than eleven million combinations, has been invented by a man in Oakland, Cal., and is now under test by agents of the United States Government and of five foreign Governments, states the Christian ■ Science Monitor. The inventor is Edward < H. Hebern, and he has been at work on l the machine for twelve years, his first ; patents, received about five years ago, • being for an instrument about six feet in height and three feet square. The present i and perfected machine, however, is 8 x 10 x 6 inches and weighs a trifle more than six pounds, about the weight of a portable . typewriter.

Externally, the invention consists of a rectangular box, with a universal typewriter keyboard covering the lower half of the front face, and three rows of small dials, each containing a letter of the alphabet on the upper half of the front. Projecting from the top is a roller, not unlike the roller on the early phonographs which operated by means of a wax cylinder record. This machine, however, has no records, and keeps no record of the messages which pass through it. In the middle of this cylindrical projection at the top is a small removable solid wheel, about the size of the metal spools used to carry the ribbon on a typewriter. Ranged at equal intervals around the rim of this wheel or spool, are 26 apertures, each containing a letter of the alphabet. Each and all of these letter terminals are wired to a central post at the cere of the wheel. This wheel is Hebern’s secret. By the internal wiring of these letters to the central connecting-post—the machine being operated by electricity—he can make 11,881,376 combinations, resulting in 403,303,146,321,064,400,000 entirely different and distinct codes. There is no key, number, word, or other device by which any one of these ccd.s can be read. Agents of the United States -jcvernment learned this when they worked for 15 months trying to decipher one message, and were unable to ascertain even one letter in that message.

While Hebern’s master-wheel is a secret the operation of the machine with the wheel in it is open to inspection by anyone, and appears to be very simple. The sending operator writes the message he is to send in plain English on the keys on the lower half of th 3 front of the machine, which operates either over wires or by wireless. ihe striking of these key’s, operating on the masterwheel, causes it to send out a message in code which it forms. The receiving machine, in which has oeen inserted a wheel synchronised to the same combination of wiring and letters as the wheel on the sending machine,, receives this code message, and prints it, letter by letter, in English again, on the small illuminated, letter-marked dials on the upper half of the machine. Thence the receiving operator writes it down in English again, just as it was ticked out on the keyboard of the sending machine. The only feature necessary to the operation of the code-making machine is that both sender and receiver use wheels which are duplicates of each other as to their internal wiring. An interloper who could very easily pick the code letter combinations out of the air with a radio receiving set, or from the wires with an ordinary Morse receiver, would be unable to decode them; indeed, the man who invented the machine would be unable to decode them, because there is no key letter or word. For example, Mr Hebern, seated at one machine, tapped out on the keyboard the word “noon”—a difficult code word because it contains only two letters, two “o’s” and two “n’s” and is apt to reveal the code letter or word for those letters. This word passed out of the sending machine, through some 50 miles of wire looped round the room, and appeared almost instantly on the illuminated dial board of the receiving machine, spelled out slowly, so as to give the receiving operator time to write it down. This, of course, demonstrated nothing more than has been done with the telegraphtypewriter. But here we cut in on the transmitting wire with a Morse receiver, and we got these four letters, U-F-E-E.

Then we tried it with nine pairs of other synchronised master-wheels, and here are the nine combinations of letters, each of which spelled out the word noon in English on the receiving board after n-o-o-n had been sent from the sending machine: YZFD, RQZK, LJSM, EFHW, VYDT, GCKG, MEZE, HKMZ, and BUFH. If there is any human decoder who can make “noon” out of any of those combinations—and they are only 10 out of nearly 12,000,000 possible combinations and - 403,000,000,000,000.000,000,000. of codes a miracle would be child’s play to him. Here is another one: PWKHUBHLIH, IMMWGHZJMVM, SZBOCVRFGUK. That is the name made on three differently synchronised and wired wheels for one of the best-known rivers in the United States. Every code w T ord contains the same number of letters as the word it represents, but no two combinations of letters in the 403,000,000,000,000,000,000, 000 of codes is the same in any two instances fcr the same word in English. Should one wheel be lost, or fall into enemy hands, all that is necessary is to stop using its mate, and use two other wheels. A field operator can carry a dozen of the wheels easily in one pocket, and by prearranged signal inform the receiving operator which wheel he is using, so that the receiver can drop into his machine the mate to it. INCREASING THE COMBINATIONS.

Statisticians and others interested in figures from a mathematical point of view solely, will see that the 12,000,000 combinations can be increased an infinite number of times by the use of a code within a code. That is to say, if the English words ticked out on the keyboard of the sending machine already are in code, then they will come back in the original English code from the receiving end. Thus, only one man, the commanding officer, if need be, knows what he has sent, though the sender may think he knows the simple English message he has tapped off to the ether machine 100 or 1000 miles away. If my figures arc correct, the first of these code-within-code combinations would result in 8,832,000.000 forms. By reversals, niter-combinations and ether tricks known to code-makers, this could be in creased to a number which would repre sent infinity.

Hebren estimates that should the air or the wires be tapped by any intruder, that intruder would have one chance in, 11.881, 376 of deciphering the code, and that it would require the full attention of a code expert, working 12 hours a day, 26 days a month, for 100 years to decode one word. It should be understood, in connection with this, that even were the intruder posses sed of the machine that Hebern has invented, it would be useless to him unless he had all the wheels with all the wiring com binations possible, and, cn top of ail this, knew which wheel to use, for no one of the nearly 12,000.000 wheel-combinations is able to decode the message from any other wheel except that wheel to which it is the synchronised mate. The inventor, of course, is net disclosing the secret of the wiring of the wheels, but he claims that, even were this fully understood, it could not be used to solve the code sent by any other wheel, since the time required would be so great that the decoded result of the message woufd be useless by the time it had been obtained.

This brings out the reason that Hebren discarded the idea of a machine

which would keep a written record, < which would show the actual code liters as sent from the master-wheel one machine to the master-wheel of m other. The operators of t-he machine especially in time cf war, never wcu! have in their machines or on their pc: tons —except in the ease of the receivir operator, who probably would be in t place of safety, far from the danger z< : .n —any written record of what has passed through the code-mak.ng instrument. 1is impossible to get, frem the machine the code letters as they pass in or out: the only way they can be picked up i< by tapping either wire or ether wit I an ordinary receiving set. Hebern be lieves that eventually he will overcome this in radio by a combination cf wav-: lengths which will baffle the present re ceiving instruments.