June 3, 2019 at 14:29Van Vivar Bush: “As We May Think”. Part 2
- Transfer
It will be about the essay by American engineer Vanivar Bush, “As We May Think,” published in The Atlantic magazine in 1945 . In it, Bush predicted the advent of the information age and the emergence of some of its manifestations, for example, personal computers, the Internet. The article describes the hypertext implemented in hardware.
This work inspired and was a guide for the pioneers of information technology Joseph Liklider (computer network, time sharing), Douglas Engelbart (mouse, NLS, GUI, proof ), Ted Nelson (hypertext, Xanadu), etc.
In 1940, Van Vivar Bush was appointed chairman of the National Defense Defense Research Committee of the United States, and from 1941 to 1947 led the organization of the successor committee - the Bureau of Scientific Research and Development, which coordinated the efforts of the scientific community (6000 leading scientists of the country) for military defense, developing nuclear weapons and the Manhattan project. Biography on Wikipedia .
- Presidential Roosevelt Science Advisor.
- He initiated the development of a differential analyzer , an analog computer that could solve differential equations with 18 independent variables.
- The supervisor is Claude Shannon (founder of the theory of information) and Frederick Terman (the "father" of Silicon Valley).
Parts 1-3
(thanks for the translation to Alexei Vorsin)
As Director of the Office of Research and Improvement, Dr. Vanivar Bush coordinated the activities of about six thousand leading American scientists in the field of applied military research. In this momentous article he invested an incentive for scientists, in case the struggle ceases. He urged the people of science to pay attention to the enormous task of turning into a more accessible, dumbfounded repository of our accumulated knowledge. Over the years, our discoveries have expanded human physical capabilities, as well as the capabilities of the mind. Falling hammers, which were the continuation of the blacksmith’s hands, microscopes that hone the eye, and machines of destruction and detection, are the latest results, but the results mark the end of modern science. Now, says Dr. Bush, “the tools are in the hands, which, if sufficiently developed, give a person access and endurance to manage the knowledge accumulated over the centuries. The perfection of such peaceful instruments should be the first goal of our scientists when they are freed from their military work. Like Emerson’s famous message in the American Scientist in 1837, this article by Dr. Bush calls for a new relationship between the thinking person and our body of knowledge.
It was not a war of scientists, it was a war in which everyone took part. Scientists who buried their old professional differences, for the sake of a common cause, shared a lot together and learned a lot. It was intoxicating, working in effective partnerships. Now, for many, this is drawing to a close.
What will scientists do next?
For biologists, and partly for medical scientists, there may be some indecision in the fact that their military work harshly dictated to them to abandon the old approaches. Many, of course, will be able to continue their military research in their good old peacetime laboratories. Their goals remain almost the same.
These are physicists, those who were most cruelly astray, who left academic careers to create strange destructive devices, who had to develop new methods for unforeseen tasks. They did their part of the work on devices that made possible the defeat of the enemy. They worked in concert with the physicists of our allies. They felt a thirst for accomplishment. They were part of a great team. Now that the world has been reached, some are asking where they can take the tasks on their shoulder.
What can be the lasting benefits of a person using the achievements of science and the new tools that his research has created? Firstly, these tools have strengthened human control over the material world. Man's clothing, his food, his shelter have improved; these tools secured man, partially freed him from the shackles of the struggle for survival. They gave him advanced knowledge about his own biological processes, so that he received an ever-increasing freedom from disease and increased life expectancy. They elucidated the relationship between psychological and physiological functions, thereby portending better mental health.
Science has accelerated communication between people; made it possible to write down ideas and the ability to manipulate this information, to extract the necessary information, thus, knowledge develops and is maintained throughout the life of the people, not the individual.
This is a growing mountain of exploration. BUT there is growing evidence that we are plunging into the swamp, due to the growing specialization. The researcher becomes indecisive among hundreds of other people's finds and conclusions, which he does not have enough time to realize, much less time to remember than these finds and conclusions exist. However, specialization is becoming increasingly necessary for progress, and the effort to build bridges between disciplines is apparently apparent.
Professionally, our methods of transmitting and reviewing research results arose several generations ago and are completely inadequate for their purpose. If the total time spent on writing scientific papers and reading them could be calculated, then the difference between these times would be striking. Those who honestly will try to keep up with the flow of thought, even in a limited area, reading in a concentrated and continuous manner, will most likely run away in horror, familiarizing themselves with the calculations of how much effort previously applied will need to be repeated again. Mendel’s ideas on the laws of genetics were inaccessible for a whole generation because his publications did not reach the few people who were able to capture and develop them; and this kind of disaster will undoubtedly happen to us again,
The problem is seen not so much in the fact that we are creating an excessive number of publications on the topics of today, in all their diversity and volume, but in the fact that the publications themselves have far gone beyond our ability to benefit from them. The totality of human experience increases with exorbitant speed, and the ways in which we advance in this maze to what we need are the same as in the days of sailboats.
The emergence of powerful new tools is a harbinger of future change. “Seeing” photocells are an advanced photography technique that can be used to photograph what is visible and what is not, thermionic tubes capable of controlling significant energies with a force lower than that required by a mosquito on a wing flap, cathode ray tubes, reproducing an image with such a frequency that in comparison with it a microsecond is a long time, relay circuits that can control complex sequences of actions are more reliable than any person and a thousand times faster - an abundance of technical solutions, helping transform science records.
Two centuries ago, Leibniz invented a calculating machine that has the majority of the required attributes of already existing devices with a keyboard, but not yet embodied in reality until now. The economic aspect of the situation is against it: the efforts that go into the development of such a machine, until the machine goes into mass use, exceed the efforts that will be saved thanks to it, until we can do with pen and paper. Moreover, frequent failures of such machines will be inevitable, so we cannot rely on them; at the moment and for a long time to come, complexity and unreliability are synonyms.
Babbage, even with incredibly generous support for his time, could not build his great arithmetic machine. His idea sounded convincing, but the creation and maintenance of those days was unbearable. If the pharaoh had all the drawings necessary for building a car, and even if he knew perfectly what he was doing, it would take the resources of the whole Egyptian kingdom to make all these hundreds of parts for one car, and such a car would surely break down on his first trip to the pyramids.
Machines with interchangeable parts can now be built with significant cost savings. In spite of their high complexity, they work very reliably. Evidence for this: typewriter, movie camera, car. Electrical contacts stopped sticking us with wheels in the wheel, as soon as they became clear.
Let me remind you about automated telephone exchanges in which there are hundreds of thousands of such contacts and they are still reliable. Metal weaves, enclosed in a thin glass container, a wire heated to a dazzling radiance in a thermo-cathode lamp from the radio - hundreds of millions of such devices, divided into groups, included in the connectors and it all works! The finest elements, the creation of which required incredible accuracy of work, and on which the best guild masters would work for months, are now made for thirty cents. The world has entered an era of cheap and complex, fault-tolerant devices; and something will certainly come next.
Recording, in order to be useful for science, must continue uninterruptedly, it must be kept, and above all this, it must be possible to access. At the moment, we record traditionally using letters and photographs, followed by printing; In addition, we record on film, on plates and on magnetic wire. Even if new methods are not added, existing ones are undoubtedly in the process of modification and development.
Indeed, progress in photography is not going to stop. Faster materials and lenses, more automated cameras, fine-grained sensitive compounds that allow developing the idea of mini-cameras are inevitable. Let me predict the logical, if not inevitable, continuation of this trend. The amateur photographer of the future wears on his forehead a tubercle a little larger than a walnut. This device captures images of three square millimeters, which can then be projected or enlarged, which, in the end, only 10 times exceeds our existing capabilities. Universal focus lenses, at any distance that the naked eye is capable of, simply due to the short focal length. There is a built-in photocell in the nut, the same one exists in at least one automatic camera now and allows you to automatically adjust the exposure in a wide range of illumination. There is enough film in the nut for hundreds of frames and a spring that moves the shutter and scrolls the film starts up only once when we insert the cartridge. This device produces full color photographs. This device can very well be stereoscopic and record with the help of spaced apart glass eyes, since the amazing successes in stereoscopic technology are already on the way.
The cord that pulls the bolt can go down the sleeve so that a person can easily use it. Short press and the picture is ready. On ordinary glasses, a square of thin lines, almost at the upper edge of the lenses, outside the usual viewing angle, where we do not notice these lines. When we see an object through this square, the lines indicate the boundaries of the future image. When the scientist of the future, in the laboratory or field, sees something worth being captured, it is enough for him to pull the shutter: he will do all the work, silently. Sounds like a fantasy? The only fantastic thing about this is the idea of showing all the resulting photos.
Will a dry photo appear? It already exists in two forms. When Brady took his photographs of the Civil War, the photographic plate was supposed to be wet all the time. Now the film should be wet only during development. In the future, it may not be necessary to wet it at all. For a long time, there was a film interspersed with a diazotizing dye, which is a kind of photograph without manifestation, so the technology of dry photography has been here since the cameras came into use. Exposure to ammonia gas destroys the unexposed dye, and the photo can be extracted and checked. The process is currently slow, but can be accelerated, and there are no particular difficulties, such as those currently occupied by photography researchers.
Now another process is being used, also slow and awkward. For fifty years, paper has been used that darkens at the place where it is touched by electrical contact, due to the chemical reaction produced by the iodide compound included in the paper. This technology can be used to record using a pointer that leaves a trail. The electric potential of the pointer can change, making the line lighter or darker.
This scheme is now used in facsimile. The pointer draws a set of closely spaced lines across the entire paper, one on top of the other. As the pointer moves, its electrical potential changes, depending on the alternating current coming from the remote station, where the photocell scans the image. In each segment, the line is as dark as the dot in the image that the photocell scans. When the whole image is transferred, the replica will appear in the output window.
The image will be read by the photocell as well, line by line, as photographed earlier. This whole apparatus is a camera with an attached device that allows you to transfer an image, which allows you to print it remotely. Also, this will give another way to dry photography, in which the photograph will be ready at the same moment as it was taken.
You need to be a brave person to claim that this process will always remain awkward, slow and defective. Television equipment today transmits sixteen reliably good images per second, and yet in its device there are only two fundamental differences from the process described above. Firstly, the recording is made by a beam of moving electrons, instead of a pointer, for the reason that the electrons move faster. Secondly, the point is the screen, which starts to glow instantly when an electron hits it, unlike film or paper, which, moreover, change forever. This speed is needed in television, as the image moves.
Using a chemically impregnated film instead of a luminous screen, the device will be able to transfer one picture instead of a series, and a high-speed camera will allow you to get a dry photo.
(“Use chemically treated film in place of the glowing screen, allow the apparatus to transmit one picture rather than a succession, and a rapid camera for dry photography results.”)
The impregnated film should act much faster (needs to be far faster in action) than existing instances, but this is achievable. A more serious drawback in this scheme is that the film will need to be placed in a vacuum chamber, because the electron beam behaves normally only in a rarefied medium. This difficulty can be avoided by forcing the electron beam to move from one side of the partition, and pressurizing the film against the other side of the partition, if such a partition allows the electrons to pass perpendicularly through the surface without scattering the beam. Such partitions, in rough form, can undoubtedly be made, and are unlikely to be able to stop the development of the idea as a whole.
Microphotographs, like dry photographs, have a great future. The main idea is to reduce the size of the record and read it through the projection, instead of reading directly, and this idea opens up too many possibilities to be ignored. The combination of optical projection and downsizing is already yielding results such as scientific microfilms, and the potentialities of this technology are extremely attractive. At the moment, the microfilm, reduced by 20 times, gives the same clear image as the material of the original size. The possibilities are limited by the film grain, the perfection of the optical system and the efficiency of the used light detectors. All this is developing rapidly.
Imagine a hundredfold decrease in the future. Imagine a film the thickness of paper, no doubt, and a thinner film can be used. Even if these conditions are met, we get a ten thousandth difference in size between the original book and recorded on microfilm. The British Encyclopedia will fit in a matchbox, a library with millions of volumes on half the desktop. All that mankind has written down since learning to write: all magazines, newspapers, books, treatises, advertising leaflets, correspondence, has a volume comparable to a billion books, and all this volume, properly arranged and compressed, can easily fit in a mobile van. Simple compression, of course, is not enough; it is required not only to create and store a record, but also to access it, we will consider this aspect of the issue later.
Compression is important when it comes to cost. The material for microfilm with the British Encyclopedia will cost 5 cents (nickel - nickel), and can be sent by mail for a cent. How much will a million copies cost? Printing a newspaper sheet takes a small part of the cent. All British Encyclopedia material in microfilm will fit on a sheet eight and a half by eleven inches in size. As soon as this becomes possible, with the help of photographic technologies of the future, the cost of a copy in good quality will be less than a cent. Preparing the original for copying? This brings us to the discussion of the next aspect.
To record, we take a pen or knock on the keys of a typewriter. Then comes the turn of correction and cleaning of the text, entailing an intricate process of typographic typing, printing and distribution of finished products. What if we assume that in the future the writer will be spared the need to write or print, directly slandering the text for writing? He acts very indirectly, telling a stenographer or a wax cylinder; but all the elements necessary for him to directly record the voice in the form of printed text already exist. All he needs to do is use the capabilities of existing mechanisms and change his language.
At the last World Exhibition, the Voder machine was introduced. The girl pressed the keys and the car reproduced a recognizable speech. No human speech was introduced into the procedure, at any stage, the keys simply combined the vibrations produced by the electric current, which were then output to the loudspeaker. The Bell lab has a remake of this car called the Vocoder. The loudspeaker is replaced by a microphone that picks up sound. You speak into it, and certain keys move. This may become one of the elements of the above system.
We find another element in the stenotypic machine (stenotype - a typewriter that prints in whole words), the confusing device that is usually found at public events. The girl bores her keys bored, with an annoying gaze, drills the hall and lecturer. A typewriter appeared from this typewriter, recording in phonetically simplified language what the speaker was saying. Later, this strip is reprinted into ordinary speech, since it itself is understood only by the initiates. Combine these two elements, we get a machine that will print what we say.
Existing languages are not adapted for this kind of mechanization, this is obvious. It is strange that researchers of universal languages have not yet understood the idea of creating such a language that would be convenient for speech transmission and recording technologies. Mechanization at the moment affects the results negatively, especially in the scientific community; it is worth recalling that scientific jargon still does not become more understandable to the common man.
Let us introduce the researcher of the future in his laboratory. His hands are free, he himself is not attached to the place. Going about his business, watching, he simultaneously takes pictures and leaves comments. Recording time is automatically written to bind them sequentially. If he goes into the field, he can be connected via radio to a recording device. When he parses his notes in the evening, he can also make comments on already recorded notes. His written notes, like photographs, are miniature, so he uses a projector to study them.
Even more than collecting data and observations, we need to extract similar data from existing records, and introduce new material into the body of an existing record. There is no mechanical replacement for mature thought. But creative thought and essentially repeating thought are very different things. For the latter, mechanical aids are already possible and already exist.
Adding a column of numerical data is a repeating thought process, and it was long ago entrusted to the machine. Undoubtedly, the machine is sometimes controlled by the keyboard: we reflect on what to enter when reading the data, and sort out the necessary keys, but even this can be avoided. Already created machines that will read printed data with the help of photocells and press the corresponding key; this is a combination of photocells for scanning, electrical circuits for sorting the results, and relay circuits for translating the results into the movements of the solenoids pushing the keys.
All these difficulties are necessary because of the awkward way that we learned to record our data. If we correctly position them on the cards and record them with simple sets of dots, the automatic reading mechanism will be relatively simple. In fact, if the dots are replaced with holes, then we get a long-built punch card machine for the census of Herman Hollerith, now used in business. The activities of some types of large enterprises would be very difficult without such machines.
Adding is just one of the operations. To make arithmetic calculations, including subtraction, multiplication and division, and, in addition, some methods of temporarily storing the results, deleting these results from the repository, for future calculations and printing the final results.
Now there are two types of machines for solving these problems: machines with a keyboard for counting and all that, manual control of data entry, and automation, usually the greater, the more sequences of operations are used; as well as machines on punch cards, in which different operations are usually delegated to several machines and punch cards are actually transferred from one to another (machine). Both types of machines are very convenient, but, in the light of interest in complex calculations, they are still in their infancy.
Fast, electricity-based notation will appear shortly after physicists show interest in calculating cosmic radiation. For their own needs, physicists quickly created equipment based on thermo-cathode lamps, capable of counting electrical impulses with a frequency of up to 100,000 per second. Advanced arithmetic machines of the future will be fundamentally electric and operate at speeds hundreds of times greater than existing ones.
Moreover, these machines will be much more versatile than existing commercial machines, and will be easily customizable for a wide range of tasks. They will be controlled by a control card or film, they will select and process the data on their own, in accordance with the instructions entered, they will perform complex calculations at extremely high speeds, and will make notes in a form that will allow further manipulation of the data and its distribution . These cars will have a monstrous appetite. Even one of them will be able to receive instructions from a whole room of keyboardists and give out sheets of counting results every few minutes. There will always be something to count in the relationships between millions of people engaged in complex business.
To be continued
(Who wants to help with the translation of the second part - contact magisterludi )
Editor's note.
As Director of the Office of Research and Improvement, Dr. Vanivar Bush coordinated the activities of about six thousand leading American scientists in the field of applied military research. In this momentous article he invested an incentive for scientists, in case the struggle ceases. He urged the people of science to pay attention to the enormous task of turning into a more accessible, dumbfounded repository of our accumulated knowledge. Over the years, our discoveries have expanded human physical capabilities, as well as the capabilities of the mind. Falling hammers, which were the continuation of the blacksmith’s hands, microscopes that hone the eye, and machines of destruction and detection, are the latest results, but the results mark the end of modern science. Now, says Dr. Bush, “the tools are in the hands, which, if sufficiently developed, give a person access and endurance to manage the knowledge accumulated over the centuries. The perfection of such peaceful instruments should be the first goal of our scientists when they are freed from their military work. Like Emerson’s famous message in the American Scientist in 1837, this article by Dr. Bush calls for a new relationship between the thinking person and our body of knowledge.
How can we think
It was not a war of scientists, it was a war in which everyone took part. Scientists who buried their old professional differences, for the sake of a common cause, shared a lot together and learned a lot. It was intoxicating, working in effective partnerships. Now, for many, this is drawing to a close.
What will scientists do next?
For biologists, and partly for medical scientists, there may be some indecision in the fact that their military work harshly dictated to them to abandon the old approaches. Many, of course, will be able to continue their military research in their good old peacetime laboratories. Their goals remain almost the same.
These are physicists, those who were most cruelly astray, who left academic careers to create strange destructive devices, who had to develop new methods for unforeseen tasks. They did their part of the work on devices that made possible the defeat of the enemy. They worked in concert with the physicists of our allies. They felt a thirst for accomplishment. They were part of a great team. Now that the world has been reached, some are asking where they can take the tasks on their shoulder.
1
What can be the lasting benefits of a person using the achievements of science and the new tools that his research has created? Firstly, these tools have strengthened human control over the material world. Man's clothing, his food, his shelter have improved; these tools secured man, partially freed him from the shackles of the struggle for survival. They gave him advanced knowledge about his own biological processes, so that he received an ever-increasing freedom from disease and increased life expectancy. They elucidated the relationship between psychological and physiological functions, thereby portending better mental health.
Science has accelerated communication between people; made it possible to write down ideas and the ability to manipulate this information, to extract the necessary information, thus, knowledge develops and is maintained throughout the life of the people, not the individual.
This is a growing mountain of exploration. BUT there is growing evidence that we are plunging into the swamp, due to the growing specialization. The researcher becomes indecisive among hundreds of other people's finds and conclusions, which he does not have enough time to realize, much less time to remember than these finds and conclusions exist. However, specialization is becoming increasingly necessary for progress, and the effort to build bridges between disciplines is apparently apparent.
Professionally, our methods of transmitting and reviewing research results arose several generations ago and are completely inadequate for their purpose. If the total time spent on writing scientific papers and reading them could be calculated, then the difference between these times would be striking. Those who honestly will try to keep up with the flow of thought, even in a limited area, reading in a concentrated and continuous manner, will most likely run away in horror, familiarizing themselves with the calculations of how much effort previously applied will need to be repeated again. Mendel’s ideas on the laws of genetics were inaccessible for a whole generation because his publications did not reach the few people who were able to capture and develop them; and this kind of disaster will undoubtedly happen to us again,
The problem is seen not so much in the fact that we are creating an excessive number of publications on the topics of today, in all their diversity and volume, but in the fact that the publications themselves have far gone beyond our ability to benefit from them. The totality of human experience increases with exorbitant speed, and the ways in which we advance in this maze to what we need are the same as in the days of sailboats.
The emergence of powerful new tools is a harbinger of future change. “Seeing” photocells are an advanced photography technique that can be used to photograph what is visible and what is not, thermionic tubes capable of controlling significant energies with a force lower than that required by a mosquito on a wing flap, cathode ray tubes, reproducing an image with such a frequency that in comparison with it a microsecond is a long time, relay circuits that can control complex sequences of actions are more reliable than any person and a thousand times faster - an abundance of technical solutions, helping transform science records.
Two centuries ago, Leibniz invented a calculating machine that has the majority of the required attributes of already existing devices with a keyboard, but not yet embodied in reality until now. The economic aspect of the situation is against it: the efforts that go into the development of such a machine, until the machine goes into mass use, exceed the efforts that will be saved thanks to it, until we can do with pen and paper. Moreover, frequent failures of such machines will be inevitable, so we cannot rely on them; at the moment and for a long time to come, complexity and unreliability are synonyms.
Babbage, even with incredibly generous support for his time, could not build his great arithmetic machine. His idea sounded convincing, but the creation and maintenance of those days was unbearable. If the pharaoh had all the drawings necessary for building a car, and even if he knew perfectly what he was doing, it would take the resources of the whole Egyptian kingdom to make all these hundreds of parts for one car, and such a car would surely break down on his first trip to the pyramids.
Machines with interchangeable parts can now be built with significant cost savings. In spite of their high complexity, they work very reliably. Evidence for this: typewriter, movie camera, car. Electrical contacts stopped sticking us with wheels in the wheel, as soon as they became clear.
Let me remind you about automated telephone exchanges in which there are hundreds of thousands of such contacts and they are still reliable. Metal weaves, enclosed in a thin glass container, a wire heated to a dazzling radiance in a thermo-cathode lamp from the radio - hundreds of millions of such devices, divided into groups, included in the connectors and it all works! The finest elements, the creation of which required incredible accuracy of work, and on which the best guild masters would work for months, are now made for thirty cents. The world has entered an era of cheap and complex, fault-tolerant devices; and something will certainly come next.
2
Recording, in order to be useful for science, must continue uninterruptedly, it must be kept, and above all this, it must be possible to access. At the moment, we record traditionally using letters and photographs, followed by printing; In addition, we record on film, on plates and on magnetic wire. Even if new methods are not added, existing ones are undoubtedly in the process of modification and development.
Indeed, progress in photography is not going to stop. Faster materials and lenses, more automated cameras, fine-grained sensitive compounds that allow developing the idea of mini-cameras are inevitable. Let me predict the logical, if not inevitable, continuation of this trend. The amateur photographer of the future wears on his forehead a tubercle a little larger than a walnut. This device captures images of three square millimeters, which can then be projected or enlarged, which, in the end, only 10 times exceeds our existing capabilities. Universal focus lenses, at any distance that the naked eye is capable of, simply due to the short focal length. There is a built-in photocell in the nut, the same one exists in at least one automatic camera now and allows you to automatically adjust the exposure in a wide range of illumination. There is enough film in the nut for hundreds of frames and a spring that moves the shutter and scrolls the film starts up only once when we insert the cartridge. This device produces full color photographs. This device can very well be stereoscopic and record with the help of spaced apart glass eyes, since the amazing successes in stereoscopic technology are already on the way.
The cord that pulls the bolt can go down the sleeve so that a person can easily use it. Short press and the picture is ready. On ordinary glasses, a square of thin lines, almost at the upper edge of the lenses, outside the usual viewing angle, where we do not notice these lines. When we see an object through this square, the lines indicate the boundaries of the future image. When the scientist of the future, in the laboratory or field, sees something worth being captured, it is enough for him to pull the shutter: he will do all the work, silently. Sounds like a fantasy? The only fantastic thing about this is the idea of showing all the resulting photos.
Will a dry photo appear? It already exists in two forms. When Brady took his photographs of the Civil War, the photographic plate was supposed to be wet all the time. Now the film should be wet only during development. In the future, it may not be necessary to wet it at all. For a long time, there was a film interspersed with a diazotizing dye, which is a kind of photograph without manifestation, so the technology of dry photography has been here since the cameras came into use. Exposure to ammonia gas destroys the unexposed dye, and the photo can be extracted and checked. The process is currently slow, but can be accelerated, and there are no particular difficulties, such as those currently occupied by photography researchers.
Now another process is being used, also slow and awkward. For fifty years, paper has been used that darkens at the place where it is touched by electrical contact, due to the chemical reaction produced by the iodide compound included in the paper. This technology can be used to record using a pointer that leaves a trail. The electric potential of the pointer can change, making the line lighter or darker.
This scheme is now used in facsimile. The pointer draws a set of closely spaced lines across the entire paper, one on top of the other. As the pointer moves, its electrical potential changes, depending on the alternating current coming from the remote station, where the photocell scans the image. In each segment, the line is as dark as the dot in the image that the photocell scans. When the whole image is transferred, the replica will appear in the output window.
The image will be read by the photocell as well, line by line, as photographed earlier. This whole apparatus is a camera with an attached device that allows you to transfer an image, which allows you to print it remotely. Also, this will give another way to dry photography, in which the photograph will be ready at the same moment as it was taken.
You need to be a brave person to claim that this process will always remain awkward, slow and defective. Television equipment today transmits sixteen reliably good images per second, and yet in its device there are only two fundamental differences from the process described above. Firstly, the recording is made by a beam of moving electrons, instead of a pointer, for the reason that the electrons move faster. Secondly, the point is the screen, which starts to glow instantly when an electron hits it, unlike film or paper, which, moreover, change forever. This speed is needed in television, as the image moves.
Using a chemically impregnated film instead of a luminous screen, the device will be able to transfer one picture instead of a series, and a high-speed camera will allow you to get a dry photo.
(“Use chemically treated film in place of the glowing screen, allow the apparatus to transmit one picture rather than a succession, and a rapid camera for dry photography results.”)
The impregnated film should act much faster (needs to be far faster in action) than existing instances, but this is achievable. A more serious drawback in this scheme is that the film will need to be placed in a vacuum chamber, because the electron beam behaves normally only in a rarefied medium. This difficulty can be avoided by forcing the electron beam to move from one side of the partition, and pressurizing the film against the other side of the partition, if such a partition allows the electrons to pass perpendicularly through the surface without scattering the beam. Such partitions, in rough form, can undoubtedly be made, and are unlikely to be able to stop the development of the idea as a whole.
Microphotographs, like dry photographs, have a great future. The main idea is to reduce the size of the record and read it through the projection, instead of reading directly, and this idea opens up too many possibilities to be ignored. The combination of optical projection and downsizing is already yielding results such as scientific microfilms, and the potentialities of this technology are extremely attractive. At the moment, the microfilm, reduced by 20 times, gives the same clear image as the material of the original size. The possibilities are limited by the film grain, the perfection of the optical system and the efficiency of the used light detectors. All this is developing rapidly.
Imagine a hundredfold decrease in the future. Imagine a film the thickness of paper, no doubt, and a thinner film can be used. Even if these conditions are met, we get a ten thousandth difference in size between the original book and recorded on microfilm. The British Encyclopedia will fit in a matchbox, a library with millions of volumes on half the desktop. All that mankind has written down since learning to write: all magazines, newspapers, books, treatises, advertising leaflets, correspondence, has a volume comparable to a billion books, and all this volume, properly arranged and compressed, can easily fit in a mobile van. Simple compression, of course, is not enough; it is required not only to create and store a record, but also to access it, we will consider this aspect of the issue later.
Compression is important when it comes to cost. The material for microfilm with the British Encyclopedia will cost 5 cents (nickel - nickel), and can be sent by mail for a cent. How much will a million copies cost? Printing a newspaper sheet takes a small part of the cent. All British Encyclopedia material in microfilm will fit on a sheet eight and a half by eleven inches in size. As soon as this becomes possible, with the help of photographic technologies of the future, the cost of a copy in good quality will be less than a cent. Preparing the original for copying? This brings us to the discussion of the next aspect.
3
To record, we take a pen or knock on the keys of a typewriter. Then comes the turn of correction and cleaning of the text, entailing an intricate process of typographic typing, printing and distribution of finished products. What if we assume that in the future the writer will be spared the need to write or print, directly slandering the text for writing? He acts very indirectly, telling a stenographer or a wax cylinder; but all the elements necessary for him to directly record the voice in the form of printed text already exist. All he needs to do is use the capabilities of existing mechanisms and change his language.
At the last World Exhibition, the Voder machine was introduced. The girl pressed the keys and the car reproduced a recognizable speech. No human speech was introduced into the procedure, at any stage, the keys simply combined the vibrations produced by the electric current, which were then output to the loudspeaker. The Bell lab has a remake of this car called the Vocoder. The loudspeaker is replaced by a microphone that picks up sound. You speak into it, and certain keys move. This may become one of the elements of the above system.
We find another element in the stenotypic machine (stenotype - a typewriter that prints in whole words), the confusing device that is usually found at public events. The girl bores her keys bored, with an annoying gaze, drills the hall and lecturer. A typewriter appeared from this typewriter, recording in phonetically simplified language what the speaker was saying. Later, this strip is reprinted into ordinary speech, since it itself is understood only by the initiates. Combine these two elements, we get a machine that will print what we say.
Existing languages are not adapted for this kind of mechanization, this is obvious. It is strange that researchers of universal languages have not yet understood the idea of creating such a language that would be convenient for speech transmission and recording technologies. Mechanization at the moment affects the results negatively, especially in the scientific community; it is worth recalling that scientific jargon still does not become more understandable to the common man.
Let us introduce the researcher of the future in his laboratory. His hands are free, he himself is not attached to the place. Going about his business, watching, he simultaneously takes pictures and leaves comments. Recording time is automatically written to bind them sequentially. If he goes into the field, he can be connected via radio to a recording device. When he parses his notes in the evening, he can also make comments on already recorded notes. His written notes, like photographs, are miniature, so he uses a projector to study them.
Even more than collecting data and observations, we need to extract similar data from existing records, and introduce new material into the body of an existing record. There is no mechanical replacement for mature thought. But creative thought and essentially repeating thought are very different things. For the latter, mechanical aids are already possible and already exist.
Adding a column of numerical data is a repeating thought process, and it was long ago entrusted to the machine. Undoubtedly, the machine is sometimes controlled by the keyboard: we reflect on what to enter when reading the data, and sort out the necessary keys, but even this can be avoided. Already created machines that will read printed data with the help of photocells and press the corresponding key; this is a combination of photocells for scanning, electrical circuits for sorting the results, and relay circuits for translating the results into the movements of the solenoids pushing the keys.
All these difficulties are necessary because of the awkward way that we learned to record our data. If we correctly position them on the cards and record them with simple sets of dots, the automatic reading mechanism will be relatively simple. In fact, if the dots are replaced with holes, then we get a long-built punch card machine for the census of Herman Hollerith, now used in business. The activities of some types of large enterprises would be very difficult without such machines.
Adding is just one of the operations. To make arithmetic calculations, including subtraction, multiplication and division, and, in addition, some methods of temporarily storing the results, deleting these results from the repository, for future calculations and printing the final results.
Now there are two types of machines for solving these problems: machines with a keyboard for counting and all that, manual control of data entry, and automation, usually the greater, the more sequences of operations are used; as well as machines on punch cards, in which different operations are usually delegated to several machines and punch cards are actually transferred from one to another (machine). Both types of machines are very convenient, but, in the light of interest in complex calculations, they are still in their infancy.
Fast, electricity-based notation will appear shortly after physicists show interest in calculating cosmic radiation. For their own needs, physicists quickly created equipment based on thermo-cathode lamps, capable of counting electrical impulses with a frequency of up to 100,000 per second. Advanced arithmetic machines of the future will be fundamentally electric and operate at speeds hundreds of times greater than existing ones.
Moreover, these machines will be much more versatile than existing commercial machines, and will be easily customizable for a wide range of tasks. They will be controlled by a control card or film, they will select and process the data on their own, in accordance with the instructions entered, they will perform complex calculations at extremely high speeds, and will make notes in a form that will allow further manipulation of the data and its distribution . These cars will have a monstrous appetite. Even one of them will be able to receive instructions from a whole room of keyboardists and give out sheets of counting results every few minutes. There will always be something to count in the relationships between millions of people engaged in complex business.
To be continued
(Who wants to help with the translation of the second part - contact magisterludi )
4
However, repetitive thought processes are not limited to questions of arithmetic and statistics. In fact, every time you combine and record facts in accordance with established logical processes, the creative aspect of thinking concerns only the choice of data and the process used, and the subsequent manipulation is repetitive in nature and, therefore, this task will be assigned to the machine. Not much has been done in these areas, beyond arithmetic, from what can be done, primarily due to the economic situation. The needs of the business and the vast market, clearly awaiting, guaranteed the emergence of mass-produced arithmetic machines as soon as the production methods were sufficiently advanced.
With machines for advanced analysis, this situation did not occur; since there was not and there is no vast market; users of advanced data manipulation techniques make up a very small fraction of the population. However, there are machines for solving differential equations, as well as functional and integral equations. There are many special machines, such as a harmonic synthesizer that predicts tides. There will be many more such machines, appearing naturally in the hands of a scientist and in small quantities.
If scientific reasoning was limited to the logical processes of arithmetic, we should not go far in our understanding of the physical world. You can just as well try to understand the game of poker exclusively using probability theory. Abacus, with its beads strung on parallel wires, led the Arabs to a positional number system and the concept of zero several hundred years before the rest of the world; and it was a useful tool, so useful that it still exists.
Abacus didn’t stand next to him with a modern keyboard. This will be the equivalent step to the arithmetic machine of the future. But even this new machine will not tell the scientist in which direction he should move. Any help should also be protected from painstaking detailed manipulations with higher mathematics if its users want to free their brain for something more than repeated detailed transformations in accordance with established rules. A mathematician is not a person who can easily manipulate numbers; often he cannot. He is not even the kind of person who can easily carry out transformations of equations using calculus. He is mainly an individual who knows how to use symbolic logic at a high level, and first of all he is a person of intuitive judgment in the choice of manipulative processes that he uses.
He must be able to turn to his mechanism as confidently as he turns to the engine of his car, to the tangled mechanism under the hood. Only then will mathematics be practically effective in bringing the growing knowledge of atomics to a useful solution to the advanced problems of chemistry, metallurgy, and biology. For this reason, there are even more machines for processing advanced mathematics for a scientist. Some of them will be bizarre enough to satisfy the most fastidious connoisseur of the real artifacts of civilization.
5
The scientist, however, is not the only person who manipulates data and views the world around him using logical processes, although he sometimes retains this skill, taking into account anyone who becomes logical, much like the British leader of the knights. Whenever logical processes of thinking are used, that is, whenever a thought passes for some time along the usual path, an opportunity arises for the machine. Formal logic was usually a favorite tool in the hands of a teacher who tested the souls of students. It is easy to build a machine that will manipulate the theses in accordance with formal logic, just wisely using relay circuits. Put the set of abstracts in such a device and turn the knob, and it will easily give you one conclusion after another,
Logic can become extremely complex, and it would certainly be nice to provide more confidence in its use. Higher-level analysis machines were usually devices for solving equations. Ideas for equation transformers are beginning to appear, which will change the ratio expressed by the equation in accordance with strict and fairly advanced logic. Progress is inhibited in an extremely rude way by which mathematicians express their relationships. They use a symbol system that has grown like Topsy and has a small sequence; strange fact in this very logical field.
The new system of symbols, probably positional, should apparently precede the reduction of mathematical transformations to machine processes. Then, going beyond the strict logic of mathematics, we find the application of logic in everyday affairs. We can one day mechanically calculate the arguments on a machine with the same confidence with which we now enter sales into the cash register. But the logic machine will not look like a cash register, even its optimized model.
Too much is required to manipulate ideas and insert them into the record. Although we seem to have done worse so far than before, now we can significantly expand the record; but even in its current volume, we are unlikely to cope. This is much more than simply extracting data for research purposes; it includes the whole process by which a person benefits from his inheritance of acquired knowledge. The first action of use is selection, and here we are actually inhibited. There can be millions of wonderful thoughts and the amount of experience on which they are based; all this is enclosed in stone walls of an acceptable architectural form; but if a scientist can get the necessary information only once a week thanks to a thorough search, his syntheses are unlikely to match the current scene.
Selection, in the broad sense of the word, is a stone that is in the hands of a joiner. However, in the narrow sense and in other areas, something has already been done mechanically by selection. A human resources officer throws a stack of several thousand employee cards into the car for selection, sets the code in accordance with the established agreement, and receives in a short time a list of all employees who live in Trenton and know Spanish. Even such devices are too slow when it comes to, for example, matching a fingerprint set to one in five million files. In this case, high-speed selection devices will soon be accelerated in comparison with their current data processing rates by several hundred minutes.
This process, however, is a simple selection: it proceeds to the next step, checking in turn one element from a large set of objects and highlighting those that have certain distinctive characteristics. There is another form of choice, best illustrated by ATS. You dial the number, and the machine selects and connects only one of the millions of possible stations. She does not run them all. She pays attention only to the class specified by the first digit, and then only to the subclass of this class specified by the second digit, and so on; and thus quickly and almost unmistakably moves to the selected station. It takes a few seconds to make a choice, although the process could be speeded up if savings could be guaranteed. If necessary, everything could be done very quickly, replacing the switching of the thermionic tube for mechanical switching, so that a complete choice could be made in one hundredth of a second. No one would like to spend the money necessary to make such a change in the telephone system, but the general idea is applicable elsewhere.
Take the prosaic issue of a large department store. Each time a credit sale is made, there are a number of things to do. It is necessary to review the inventory, the seller must obtain a loan for the purchase, it is necessary to make basic calculations, and, most importantly, the client must be serviced. A central recording device has been developed in which most of this work is done conveniently. The seller puts on the stand an identification card of the client, his card and a card taken from the sold cards. When he pulls the lever, the contacts go through the holes, the machines at the central point make the necessary calculations and notes, and the proper receipt is printed for the seller so that he can pass it on to the customer.
But the store may have ten thousand customers who will have any business with the store, and before the completion of the full operation, someone must choose the right card and insert it into the central office. Now, quick selection can instantly shift exactly the right card in place or return it back. However, another problem arises. Someone must read the total amount on the card so that the machine can add its calculated element to it. Apparently, the cards may be of the type of dry photograph already described by me earlier. Existing totals can then be read using a photocell, and a new total quantity is introduced by the electron beam.
Maps can be in miniature so they take up little space. They must move fast. They should not be transported far, but only to the position in which the photocell and tape recorder could work with them. Positional points can enter data. At the end of the month, you can easily read and print a regular invoice. With the so-called tube selection, in which no mechanical parts are involved in the switches, it takes little time to use the right card - a second should be enough for the whole operation. The entire recording on the card can be made with magnetic dots on a steel sheet, if you wish, instead of points that should be observed optically, following the pattern by which Pulsen long ago recorded speech on magnetic wire. The advantage of this method is its unpretentiousness and ease of erasing. When using photography, however, it is possible to organize the projection of the recording in an enlarged view and at a distance using the general process in television equipment.
One can consider the quick selection of this shape and the remote projection for other purposes.
To be able to enter one sheet of a million before the operator does it in a second or two, with the ability to add notes to them, which leads to reflection in many ways. It may even be useful in libraries, but that's another story. In any case, some interesting combinations are now possible. For example, you can talk with a microphone, just as described in connection with a typewriter with speech control, and thus make your selection. Thus, the machine would definitely outperform the performance of an ordinary clerk.
6
However, the essence of the selection issue is much deeper than the lag in the adoption of mechanisms by libraries or the lack of development of devices for their use. Our inability to get a record is largely due to the artificiality of indexing systems. When data of any type is placed in the store, they are submitted in alphabetical order or numerically, and the information is (when it is) by tracking it from subclass to subclass. It can only be in one place if duplicates are not used; you need to have rules regarding which path it appears in, and these rules are cumbersome. Moreover, having found one element, you must log out of the system and enter the new path again.
The human mind does not work this way. He acts by combining. One object in his mind instantly becomes attached to the next, which is proposed by the method of association, in accordance with some complex chain of neural pathways reproduced by brain cells. Of course, he has other characteristics; paths that are not often followed are prone to fading, objects are not permanent, memory is temporary. Nevertheless, the speed of action, the complexity of the paths, the details of mental images, are impressive, among other things, that exists in nature.
A person cannot hope for a complete artificial duplication of this mental process, but he, of course, must be able to learn from it. In minor cases, he can even improve this scheme, since his records have relative constancy. The basic idea, however, should be taken from the analogy regarding choice. Associative choice, in contrast to indexing, can be mechanized. Thus, one cannot hope for equal speed and mobility with which the mind follows an associative path, but one must be able to defeat the consciousness with regard to the constancy and clarity of the objects that we get from the store.
Imagine the device of the future for personal use, which is a mechanized personal file and library. He needs a name and, choosing on a random, let it be “memex”. Memex is a device in which the user stores all his books, notes and conversations, and since the device is mechanized, it provides the necessary information with sufficient speed and flexibility. This is an enhanced personal addition to our memory.
Memex consists of a desktop, and although it can be used from a distance, it is primarily a piece of furniture that can be worked on. Above are translucent screens at an angle, on which material can be projected for easy reading. There is a keyboard, sets of buttons and levers. Otherwise, it all looks like a regular table.
Material is stored at one end. Improved microfilm copes with large volumes of data. Only a small part of the interior of Memex is reserved for storage, the rest is directly the mechanism itself. However, even if the user inserts 5,000 pages of material per day, it will take him hundreds of years to fill the repository and consider himself a spender who is wasting material.
Most of the Memex content obtained from microfilm is ready for administration. Books of all kinds, drawings, current periodicals, newspapers are thus obtained and abandoned inside the device. Business correspondence is sent there too. And there is a margin for direct introduction. At the top of Memex is a transparent window. It contains handwritten notes, photographs, notes, and more. When someone is near the device, pressing the lever forces the device to photograph the object and record it on the next empty spot in the memex film section. It uses a dry photo.
There is, of course, a provision on working with records using the usual indexing scheme. If the user wants to access a specific book, he clicks its code on the keyboard, and the title page of the book immediately appears in front of him, projecting onto one of the viewing positions. Frequently used codes are mnemonic, so the user rarely accesses his code book; but when he does, a single keystroke projects it for use. Moreover, he has additional leverage. Turning one of these levers to the right, he runs through the book with his eyes, each page in turn is projected at a speed that simply allows the gaze to catch on each page. If he continues to swing the lever to the right, he will flip 10 pages of the book at a time; reject even more - immediately scroll 100 pages.
A special button takes it directly to the first page of content. Thus, any book from his library can be called up and viewed with much greater capabilities than if it had been taken from a shelf. Since it has several projection positions, it can leave one element in the position while it calls another. He can add notes and comments on the margins using one of the possible types of dry photography, and you can even make it possible for the user to do this with the stylus, which is now used in the teleautograph, which can be seen in railway waiting rooms. You can do everything that we could do if we had a physical page in front of us.
7
All this is conditional, with the exception of the projection of modern mechanisms and gadgets. Thus, one can immediately start associative indexing, the main idea of which is the position in which any item can be called immediately if desired and another item can be automatically selected. This is an essential feature of Memex. The process of linking two objects is an important part of his work.
When the user creates his own path, he gives it a name, inserts it into his codebook and clicks on the keyboard. Before it are two objects that need to be connected, which are projected onto adjacent viewing positions. At the bottom of each of them there are several empty areas of code, and the pointer is set in such a way as to point to one of them when working with each element. The user presses one key and the elements are constantly connected. The set of points for viewing the photocell is out of our sight, but is part of the code; and on each element, these points in their positions indicate the serial number of another element.
After that, at any time when one of these elements is in viewing mode, the other object can be immediately called up by a simple click of a button under the corresponding code space. In addition, after numerous objects were connected together in such a way as to form trajectories, they could be viewed in turn, quickly or slowly, deflecting a lever similar to that used to turn pages of a book. Just as if physical objects were collected from widely separated sources and combined into a new book. Moreover, for any subject, you can combine multiple trajectories.
The owner of Memex, let's say, is interested in the origin and properties of the bow and arrow. In particular, he studies why the short Turkish bow was clearly superior to the English long bow in battles during the Crusades. He has dozens of related books and articles at Memex. First, he flips through the encyclopedia, finds an interesting, but fragmentary article, leaves it designed. Then, in history, he finds another suitable item and ties them together. Thus, he works by building a trace of many objects. From time to time, he inserts his own comment, either by linking it to the main path, or by attaching it with a third-party path to a specific subject. When it becomes apparent that the elastic properties of the materials available are of great relevance to the bow, it forks a side path, which now shows him elasticity textbooks and tables of physical constants. He inserts a page of his own analysis. In this way, he paves the way for his interest through the labyrinth of materials available to him.
And its trajectories do not disappear. A few years later, his conversation with a friend turns into a discussion of the strange ways that people use to resist innovation, even vital ones. He has an example that indignant Europeans still could not accept the Turkish bow. In fact, it has a trajectory containing this example. Touch calls the codebook. Pressing a few keys projects the main path folder. A lever runs through it at will, stopping at interesting objects, periodically causing third-party materials. This is an interesting trajectory that relates to the discussion. In this way, he puts the reproducer into action, photographs all the material of the path and transfers it to his friend for insertion into his own Memex, in order to connect the materials with a more general path.
8
There will be completely new forms of encyclopedias, with a freshly compiled network of associative trajectories ready to get into Memex. The lawyer has at his disposal relevant opinions and decisions that he made in his life, as well as the experience of friends and authoritative personalities. The patent attorney has at his disposal millions of patents granted, with similar elements regarding each point of interest of his client. The doctor, puzzled by the patient’s reactions, follows the path established during the earlier study of such a case, and quickly looks through similar case histories with side references to sources containing information about anatomy and histology. A chemist struggling with the synthesis of an organic compound has at his disposal all the chemical literature in the laboratory, where the trajectories are similar to the compounds
The historian, with an extensive chronological description of the people, draws a parallel with the trajectory, which stops only on significant objects, and can at any time go along modern trajectories that will lead him throughout civilization in a certain era. A new profession has emerged for trajectory seekers, people who enjoy creating useful trajectories by working with a huge number of common records. The legacy of the master is not only his additions to world records, but also the vast amount of materials on the basis of which they were made.
Thus, science can realize the ways in which a person will create and store records of mankind and access them if necessary. It can be strikingly more vividly describe the tools of the future, rather than take a close look at the methods and elements already known and rapidly developing, as was done before. Of course, technical difficulties of all kinds were ignored, but still unknown means that could come any day to accelerate technological progress as quickly as the thermionic tube appeared were also ignored. In order for the picture not to be too ordinary, due to the fact that it adhered to modern models, it would be nice to mention an important thing - not to prophesy, but simply to assume that a prophecy based on the expansion of something known has an essence, while prophecy,
All our steps for creating or perceiving recording material go through different feelings: tactile when we touch the keys, verbal when we speak or listen, visual when we read. Is it hard to imagine that someday the trajectory can be set more directly?
We know that when the eye sees, all subsequent information is transmitted to the brain through electrical vibrations in the optic nerve canal. This is an exact analogy with the electrical vibrations that occur in the cable of the TV: they transmit the image from the photocells to the radio transmitter from which it is broadcast. We also know that if we can approach this cable with the appropriate tools, we do not need to touch it; we can pick up these vibrations with the help of electric induction and thus detect and reproduce a scene that is transmitted, just like a telephone wire can listen to messages.
Impulses entering the nerves of the typist’s hands transmit the translated information to her, which reaches her eyes or ears so that her fingers can hit the corresponding keys. Could these flows be intercepted, either in the original form, in which information is transmitted to the brain, or in a surprisingly metamorphosed form, in which they then pass to the hand?
Through bone conduction, we are already introducing sounds: for example, into the nerve channels of the deaf so that they can hear. Is it hard to imagine that we can learn to introduce them without the cumbersomeness of the first conversion of electrical vibrations into mechanical ones, which the human mechanism quickly converts back into electrical form? With a pair of electrodes on the skull, the Encephalograph now produces traces of ink that are somewhat related to electrical phenomena occurring in the brain itself. True, the record is illegible, except that it indicates a certain gross malfunction of the brain mechanism; but who will now determine what this may lead to?
In the outside world, all forms of intelligence, whether sound or sight, have been converted into the form of alternating currents in an electrical circuit so that they can be transmitted. Inside the human body, the exact same process takes place. Should we always move on to mechanical movements in order to move from one electrical phenomenon to another? This is a thought-provoking thought, but it is unlikely to justify the prediction, without losing touch with reality and spontaneity.
Presumably, a person’s spirit should be exalted if he can better examine his dark past and more fully and objectively analyze his current problems. He built a civilization so complex that he needs to mechanize his notes more fully if he wants to push his experiment to its logical conclusion, and not just partially get bogged down, overloading his limited memory. His adventures may be more enjoyable if he can regain the privilege of forgetting about various things that he does not need to have at hand, with some certainty that he will be able to find them again if they prove to be important.
Science has given man a well-to-do home, and teaches him to live in it being healthy. She allowed people to attack each other with brutal weapons. She can still allow him to truly embrace a great story and grow in the wisdom of tracking experience. It can die in conflict before a person learns to use this record for his true good. However, in applying science to the needs and desires of a person, this may seem like an extremely unfortunate stage for ending the process or losing hope for a result.
Translation by Diana Sheremiev.
Finished translations
- The Art of Science and Engineering by Richard Hamming
- Complete Translation of Community Building: Social Architecture
- "Hackers and Artists." How we translated Paul Graham’s RuNet for 13 years
- Translation of the book “Skunk Works. Personal Memoirs of My Work at Lockheed
- Translation of Andrew Un's book, Passion for Machine Learning
- The course of lectures "Startup". Peter Thiel. Stanford 2012