Word processing

The pioneers of computing sought solutions in software rather than hardware. The statute to encode each instruction in the binary language of the internal logic of the computer — “yes” or “no”, 1 or 0 — they searched for a more convenient way to communicate with the machine. The first result of their efforts in this direction were new codes made up of letters and short words taken from the English language. Maurice Wilkes, who worked at the University of Cambridge, is considered the creator of one of the so-called assembly languages. In 1949, he developed such a language for the Edsac EDSAC machine, the first active computer whose programs were stored in its internal memory. Around the same time, researchers at MIT were developing a more advanced programming language for their still unfinished computer Whirlwind .

Usually an assembly language consists of commands with easy-to-remember characters. For example, the letters SU denote the Subtract subtraction command, TS is the transfer storage entry following the data element instruction, etc. Each such mnemonic designation replaced a long chain of zeros and ones in the program. These characters are automatically converted into binary codes of machine language, understandable to the computer, with the help of a special program called assembler.

Communication with the system in such a, even if rudimentary, human language, replacing the language of zeros and ones, was the first step towards creating a high-level programming language - and a very important step towards the use of computer processing of texts. Further progress was made in 1955 at the MIT Lincoln Laboratory when the TX-0 computer came into operation, with which students began experimenting several years later. TX-0 was truly a programmer’s dream of the time. In addition to the large by the then standards of memory - 65536 bytes - it had the most modern input-output devices: a CRT display and a photosensitive device, called a light pen, which made it possible to change the image on the screen. In addition, the advanced flexorizer of this computer worked on-line, that is, it was directly connected to the computer, and all data typed on its keyboard was immediately entered into the machine, bypassing the stuffing step and the subsequent reading of punched tape.

When young aviation engineer Jack Gilmore researcher Jack Gilmore, who served in naval aviation during the Korean War, returned to MIT, he was already awaited by the new TX-0 computer. While still a graduate student, Gilmore wrote in 1951 the first assembler program for the computer "Whirlwind". Now, colleagues suggested that he work with the TX-0 computer, saying, "I wonder what you can do with this thing."

Gilmore responded to this challenge by writing primarily a series of utility programs — utilities. Their purpose, according to Gilmore, was “to help us do the dirty work”; they allowed the programmer to modify and debug (i.e., correct the detected errors) their programs when they were already in the computer's memory. As a method of communication between the programmer and the machine, Gilmour’s programs were the “ancestors” of modern operating systems, i.e., sets of programs that perform, among other things, functions such as tracking the location of data in memory, on disk, or magnetic tape. One program called FIND (find) gave the user the opportunity to look for specific words or instructions in another program. The operator could also quickly print a program or data by typing the key word PRINT.

The Gilmore Utility System could even ask questions, thus becoming a harbinger of the type of human-machine interaction that is now called the interactive mode of operation. For example, when executing a data output routine for punched tape - this routine Gilmore called PUNCHY (from punch - punched) - the computer asked: “Will there be a header?” If the programmer answered “yes”, the program asked: “What will the header be?” then the user typed a name on the keyboard.

In the process of working on his next project, Gilmore came close to computer-based text processing in the modern sense of the term. Intending to design a more advanced input device for the TX-2, an upgraded version of the TX-0 (the TX-1 was not built), Gilmore wrote in 1957 an experimental program that allowed the alphanumeric characters on the screen to be reproduced with a light pen. CRT. A screen measuring 20 x 20 cm contained 512 x 512 picture elements (dots). “We filled the screen with dots,” Gilmore later recalled, “and then, using a light pen, selectively removed some of the dots, thereby drawing the symbol.”

Gilmore and his colleagues created the so-called program editor, which made it possible to manipulate these characters. The program displayed on the lower half of the screen an image of a keyboard with 200 keys - approximately three times more than that of Flexoraiter. Each key was represented by a dot on the screen. When one of these points was touched by a light pen on the upper half of the screen, which is a printed page, a corresponding letter or number appeared, for example 8. Using this input device, you could even put upper and lower indices.

Perhaps the most interesting feature of the editor was the presence of so-called function keys - an indispensable component of a modern computer keyboard. Here were the keys with functions similar to carriage return at the typewriter: tabs, vertical and horizontal omissions.

The program also made it possible to quickly edit and correct the text that the light pen had already captured on the modeled page. The desired place in the text could be approached using a special pointer, a cursor that moved around the screen, so that when the key was touched with a light pen, the corresponding letter occupied the position indicated by the cursor. Using a light pen, it was also possible to destroy a symbol, and in a later version of the program, move words or whole paragraphs of text across the screen.

Some of the tools of the Gilmore Experimental Program two years later were incorporated into an input system developed for the TX-2 computer. The new device was an advanced model of Flexoriter with an expanded keyboard, which included letters of the Greek alphabet, mathematical symbols, superscripts and subscripts. The device could support and direct communication with the computer, and display characters on a punched tape.

Hardly aware of this, Gilmore and his colleagues laid the foundation for fast word processing , starting from which Stephen Peiner took another step to write a program called “Dear Typewriter”. Researchers from MIT, as well as many other computing centers, were mainly engaged in computational tasks, that is, in the processing of numerical data. They invented new input devices not for processing textual information, but in order to “improve the language in which a person“ talks to the machine, ”as Gilmore noted in one of the reports in 1959.

In this sense, scientists have definitely achieved success. But although the progress in software was not in doubt, practical applications of the method of computer processing of texts followed only another innovation in hardware.

This innovation was a kind of hybrid of a new computer equipment and a good old typewriter, about which we have already told more than once. However, this time the authors of the innovation were not university researchers, but "firm". IBM, a leader in the field of computer data processing, several years earlier was engaged in the production of electric typewriters, buying a small company Electromatic Electromatic during the economic crisis. IBM engineers even invented a perforation I / O system, on the basis of which other specialists later created a flexorayter.

1961. On IBM's electric typewriter, conventional levers with letters were replaced with a spinning ball, which revolutionized printing devices.

In 1961, IBM released the Selectric typewriter , which not only fundamentally changed the design of the electric typewriter, but also became an essential component of the new word processing system. The most obvious innovation in the Selectric typewriter was its printing mechanism. The letters of all letters, numbers and signs, usually attached to individual levers, were assembled here on a single printing ball of the size of a golf ball. Moreover, this spherical element quickly moved across the page, while the other components of the machine remained stationary. Prior to this, in almost all models of typewriters the moving part was the carriage. (The only notable exception was the first primitive electric machine patented in 1872 by Thomas Edison.)

With the advent of IBM Selectric, a new standard was established for office printers, and soon, based on it, terminals were created for working with computers. As in the input device, this typewriter used simple combinations of codes, according to which each time you pressed a key, the printing element made certain rotations. As an output device, this machine had two properties that distinguished it favorably: firstly, its typing speed reached 15 characters per second, which was twice the speed of an ordinary typewriter, and, secondly, it had a fixed carriage, due to which the paper was pulled into the machine more smoothly.

Even before the start of the widespread commercial production, IBM Selectric demonstrated its capabilities as a printer in the system of a powerful IBM-7030 computer, called Stretch . This machine was developed in the late 50s to perform calculations on the instructions of the Atomic Energy Commission of the United States, as well as to serve highly secret encryption systems of the State Security Service. At that time, it was the fastest computer in the world, and its very name Stretch (stretch - to exert maximum strength) suggests that, creating this machine, IBM tried to exceed the level of computing equipment.

Although Stretch was designed to solve top-secret government assignments, IBM arranged for it to be publicly demonstrated in one of its laboratories in Paukipsi, NY, several months before the official presentation of the Selectric typewriter. To temporarily keep a new spherical printing element secret, IBM employees covered the open part of the typewriter of the Stretch system with a regular piece of cardboard. A few visitors, although they looked under the cardboard, obviously did not understand the essence of the innovation that they had before their eyes.

Commercially, the Stretch machine failed, bringing the company $ 20 million in losses. However, her printer, Selectric, was in great demand as a typewriter. And in 1964, three years after the creation of this typewriter, IBM engineers used a Selectric printing element in the system, which marked the most significant step forward (after Scholes invention) in the automation of word processing: they combined Selectric with a tape drive for data storage. This tape was essentially an electronic analogue of a punched tape used in the old flexorayter, but instead of holes, the presence or absence of magnetized zones on a plastic tape covered with a magnetic material corresponded to binary digits (units and zeros). A combination of 7 bits of electronic information encoded one or another symbol: a letter, number or sign.

The output from the Selectric keyboard was simultaneously printed on paper and recorded on a magnetic tape. If an incorrect character was typed, the printing element simply went back one position and wrote the correct one over the wrong character. Although the overlay of characters was noticeable on paper, only the correct character remained on the magnetic tape, because when you return the printing device back, the old character is automatically erased.

Thus, the magnetic storage device allowed the printing element to work faster without fear of errors that could be easily corrected. The corrected text was then imprinted on the Selectric or stored on a magnetic tape. Later, IBM improved this system by replacing the tape with magnetic cards that are more convenient to use. Each map contained an axis of 5000 characters — a little more than a full printed page printed at one interval.

The first time IBM employees, when speaking about the operation of this device, used the term “automated printing”, and in 1965 one of the employees of the German branch of the company Ulrich Steinhilfer used the term more accurately describing the electronic process of manipulating the words: textverarbeitung, which means “ word processing, ”and the term caught on.

Selectric provided IBM with a leading position right up until the early 70s. But by that time computer technology began to move forward by leaps and bounds. As the hardware became more compact and cheaper due to the mass production of integrated circuits, more and more advanced text processing systems began to emerge and quickly replace each other. They were usually produced by small firms, which, unlike IBM, had never before been involved in the production of office equipment. In 1971, Lexitron proposed the first commercial word processing system with a CRT display. In 1972, Vydec took two steps forward at once by offering the first integrated system, including a CRT screen, electric printers and floppy disk drives for storing programs and data. The plastic disk with a special coating, which was invented by IBM for large computers, had a large information capacity and a shorter access time (i.e., data retrieval) compared to magnetic cards or tape.

IBM 7030 Stretch Computer Presentation Left to right: Sid Fernbach - Head of Department, Clarence Badger Programmer and Richard von Holdt Deputy Head of Department.