Toward Distant Suns
by T. A. Heppenheimer
Copyright 1979, 2007 by T. A. Heppenheimer, reproduced with permission
Chapter 7: Robots and Other Space Workers
In the science-fiction movies of recent years, among the most appealing characters has been the robot Artoo-Deetoo (R2D2) of Star Wars. It is hard not to refer to such a robot as "he," and in his diminutive way (he looked rather like an overgrown canister-type vacuum cleaner), he performed prodigies of exertion in the service of the beautiful Princess Leia, leader of the revolt against the galactic overlord Darth Vader. But there was one thing wrong. Star Wars was set amid an interstellar civilization technically in advance of our own; but to anyone conversant with modern computers, Artoo-Deetoo would be lamentably out of date.
For Artoo-Deetoo could not speak. He communicated by means of chirps and whistles, a fact that immediately dates him. [Author's footnote: In fairness, his robot friend See-Threepio (C3P0) could not only speak but could, the occasion demanding, wax eloquent.] Computer speech is already a reality, and what is more, has been so for some time. Indeed, it is likely that by now most Americans have had the experience of hearing a computer talk, even if few have been aware that that is what they have heard.
Pick up the phone and dial a number that is out of service. In many cities, doing this will produce a message: "The number you have reached, 123-4567, is not in service. Please check the number and dial again. " The message will sound like a recording, but there's a catch; it specifies the number dialed. The phone company can't keep thousands of separate recordings of out-of-service numbers, not to mention keeping them up to date. The answer is that the message is not a recording. It is a computer synthesis of human speech.
How can a computer synthesize human speech? The technique is actually no more marvelous than the means whereby, via ink marks on paper, literate people can learn of things they never experienced. The "alphabet" of computer speech is straightforward. It is no more than the sounds, or phonemes, which make up human speech. There are about forty-five of them: long and short vowels, hard and soft consonants, and the like; they are listed in any dictionary. In ordinary speech or conversation, one hears or produces ten to twenty such phonemes per second.
Each of these phonemes can be synthesized from a combination of only a few standard sounds: tones of different frequencies, which can be added or combined by means of electronic filters, together with hissing sounds. A hissing or sibilant sound, known as white noise, is essential for phonemes like s, t, th, b, and others. Then, by means of a method of computer programming known as LPC (Linear Predictive Coding), one can very efficiently specify the way the electronic filter is to combine the various tones and the sibilance, while allowing for changes of tone, so that computer speech can sound like a man or a woman, with natural-sounding rises and falls of inflection. A flat-sounding monotone that grates on the ears is thus avoided.
To most adults, the idea of hearing a computer speak may seem bizarre, even frightening. But the next generation of children may well grow up regarding it as part of their everyday worlds. Texas Instruments, a leading manufacturer of advanced electronics, recently announced a new voice synthesizer which uses only three integrated-circuit chips, giving it roughly the complexity of the familiar pocket calculator. One of its uses is in a new $50 childrens' game, Speak and Spell.
Speak and Spell has a keyboard for letters and numbers, a calculatorlike display—and the synthesizer. For instance, it will say to a child: "Spell CAT. " The child responds by pushing buttons on the keyboard. There is a pause. Then, if the child is right, the box says, "You are right," and goes on to the next word. If the answer is wrong, the box says, "No, that's incorrect." Then the display flashes the right spelling, and the synthesizer spells it out verbally as well.
If computers can talk, can they also listen? Experiments in teaching computers to recognize human speech go back at least twenty years but have had mixed success to date. The problem is one of having a computer match a pattern of frequencies and overtones, representing a word spoken to it, with one of a large number of patterns stored in its memory. If it is to understand ordinary speech, it must be able to match such patterns very rapidly, searching over many possible matches each second. With the fast computing times and large memories of modern computers, this can indeed be done. Bell Labs has a machine today that will recognize spoken words with up to 98 percent accuracy. To do this, a person first speaks each of several thousand words, a dozen or so times each, and the computer records what it hears in its memory. The subsequent word-recognition is keyed to the speech of that particular individual, so that it cannot understand the speech of a variety of different people.
This is in any case rather a brute-force procedure, since it ignores such clues as syntax and grammar. Still, it can go further than this. By the early 1980s, as memories improve, computers will be storing patterns for up to half a million words to be recognized when spoken. Each word will be stored as twenty or thirty general patterns or variations for a vocabulary larger than most people possess. Because the patterns will not be keyed to any one person, the machine should be able to do a creditable job of understanding anyone's ordinary speech. In any case, if the machine fails to catch a word or phrase, it can do what anyone else would do in that situation. It will simply say, "Excuse me, I didn't catch that.''
With these abilities computers will be well along in their use of artificial intelligence, or AI. AI has intrigued computer scientists for quite a few years. As early as 1965 an AI program allowed a computer to play championship-level checkers. More recently chess-playing computers have come
Chess is an excellent problem for AI since there are many more possible games (and moves) than there are electrons in the Universe. Hence a chess-playing computer must be able to recognize patterns and eliminate huge numbers of inappropriate possibilities. This attribute, of course, is what is needed in order to recognize ordinary human speech. To be sure, computers can do rather more than play chess. As early as a decade ago, Joseph Weitzenbaum of MIT devised a program called ELIZA, with which one could converse with a computer in English, using a keyboard to type in statements. He also gave the computer another program, DOCTOR, designed to permit it to play the role of a psychotherapist. In his book, Computer Power and Human Reason, he records the following exchange:
When a computer does psychoanalysis, or when it plays a good game of chess, there is obviously the question: Can it match the thinking ability of a human being? The important feature of human intelligence is not that ordinary people can match a computer when it comes to number-crunching. The mathematician Karl Friedrich Gauss committed logarithm tables to memory, the better to aid his studies, and there have been occasional calculating prodigies who could multiply twenty-digit numbers in their heads. But people usually outshine a computer at dealing with patterns and symbols, particularly when the patterns are complex. Why? Because the amount of memory available to an ordinary human is something like a trillion bits, or twenty thousand books the size of the Bible. (The King James Version has 773,693 words.)
Until recently there was no prospect that a computer could match this. True, such quantities of memory could be stored on magnetic tapes; but it would be a slow, inefficient process to mount tapes and have the computer search them through. Also, pattern-recognition programs have not been powerful enough to allow computers to make effective use of such reserves of memory. But the video disk has brought new hope. To electronics buffs, the video disk is already familiar; it is about the size of an ordinary long-playing record but records and plays not music but rather TV shows. A single such disk can store up to four hours of TV viewing. Used as a means of storing data, such a video disk can readily store a trillion bits. With appropriate coding and organization, any part of this information can be made available to a computer in less than a tenth of a second.
There is little doubt that AI programs will grow smarter, more subtle. Already MIT has a program whereby a computer can learn about its world as would a child. This concept-learning program can infer the idea of, say, an arch simply by being shown a set of pictures of arches and non-arches. A common-sense-reasoning program written at Stanford Research Institute learned to extend its initial solution to a particular problem to a more general solution applicable to a range of problems.
One trend in AI will be to develop "knowledge about knowledge"; that is, models of what a computer knows, so that it can select or anticipate an appropriate response. To a degree, BORIS does this in chess when it predicts correctly an opponent's move and flashes, "I expected that. " In programs such as ELIZA, which deal with written or spoken English, a knowledge of grammar and syntax will obviously permit advances over the brute-force recognition or matching-up of word patterns. Even more subtlety will be available through the programming of rule-making rules. Thus, a computer may start with a set of basic rules for understanding conversation or English text. However, with these rules as examples, it will be able to formulate new rules and to try different sets till it gets the best response and can interpret its material most fluently.
Within a decade all these trends may culminate in a computer that can read the texts of thousands of books and interpret them. It will be able to talk and to listen. One could thus converse with it as with the brightest and most well-read of companions. It will be the Speak and Spell game, but at an unimaginably richer and more complex level.
By the time AI reaches this level, it will be possible to anticipate a day when robots and computers will fulfill all the functions and perform all activities for which we now believe humans will be needed in space. Such computers will even check out one another's performance, carry out maintenance, or make repairs, removing faulty electronics units and replacing them with good ones. This will actually be a continuation of trends of past decades.
In the 1950s it would scarcely have been believed that the exploration of Mars would fall to such craft as Viking. Even in the 1970s there was much pointing with pride at the lunar explorations of astronauts, who could select geological samples of interest. Today this is ancient history. When anticipating similar explorations of Mars, it is generally agreed that there will be mobile robot vehicles resembling Viking spacecraft with tank treads. Ranging widely over the Martian surface for months or years, they will stop frequently to make measurements and chemical analyses. The most interesting rock samples will be loaded aboard rockets and returned to Earth under automatic control. From a control center in Pasadena, the world's best geologists will learn more about Mars than the Apollo teams learned by actually visiting the Moon.
As in exploring Mars, so one day will sophisticated robots assemble entire power satellites; but that day still appears far off. If the power satellite becomes a major project before century's end, almost certainly the automated systems used will be far less sophisticated. Like power shovels and other road-building equipment, there will be the beam-builders and cranes or manipulators described in Chapter 6, each requiring control by an operator. Other space workers will operate the boom-supported cherry pickers or remote work stations. The highest degree of sophistication will be in the remotely controlled robots known as teleoperators.
The Space Spider will be one such teleoperator. Under control of an internal computer, it will lay down metal structures by following along the periphery of a shape already built. It will do this until it is commanded to stop or its rolls of sheet metal run out. A more interesting type of robot will be the free-flying teleoperator. It will be equipped with manipulator arms and "end effectors "—robot hands, if you will. It will also have TV cameras and lights and a rocket motor with propellant. Its operator will use its TV cameras to see what it sees, and by remote control the operator will steer it to a particular work station. There the robot will be put to work using tools or other equipment it may carry, while the operator sits in the comfort of his control center, projecting his skills through a radio link.
Once there is good understanding of the construction equipment to be used, it is a straightforward matter to determine how large a work force must be supported in space in order to build power satellites. Even without robots of extraordinary sophistication, even with beam-builders and similar items that call for the attention of skilled human operators, the construction rates can be astonishing. This is because of the ease with which large structures can flow from a source, being very light in design and requiring no elaborate bracing.
The Space Needle in Seattle, a relic of the 1962 Worlds Fair, is a good example of the difference between space and earth construction. Leaving aside the revolving restaurant on top, it is 150 meters high. It has a massive concrete foundation, and with the foundation but without the restaurant, it weighs 8,660 tons. It took six months to build with a construction crew of fifty. By contrast, a similar-size space structure built of beams would be 200 meters long by 20 wide. Such a structure, indeed, would be a section of twenty-meter beam, described in Chapter 6 as the main structural element of a powersat. It would weigh 1.5 tons and take three hours to build using a crew of two.
So the work force will not be large, by standards of even a modest earthside construction project. Still, it will dwarf all previous notions of the size of space crews. It will be, indeed, the first large-scale use of people in space. There will be about five hundred workers in a variety of categories. At Boeing, where Eldon Davis and Keith Miller have studied the problems of power satellite construction, the space jobs to be filled are believed to include
These jobs are similar to those in present-day industry. The work of a crane/manipulator operator would resemble that of manipulator operators in the nuclear industry, where radioactive fuel rods must be handled remotely, behind thick shielding. The operators for automated equipment like the beam machines and deployment machines will have work similar to that of operators of automated wing riveter machines in the aerospace industry or of paper-making or bottle-making machines. The people who do maintenance, cargo handling, communications, and traffic control will likely have done very nearly the same things at airports or air-freight terminals. The management, supervisory, and staff jobs will be like those found in any earthside manufacturing plant.
When these jobs open up, applicants will not have to be superior people, like astronauts. They will be chosen for being adventurous, ambitious, hard-working, intelligent, and having a strong commitment to excellence in their work. Requirements on their physical condition will be minimal, most notably that they be in good health, capable of tolerating several g's of acceleration, minimally susceptible to motion sickness, and not too far from the general population in height and weight. There will be psychological tests, like those used to select crews of submarines, in order to detect applicants who could not stand the cramped quarters and unfamiliar environments of space.
Equal opportunity laws will be used in hiring; there will be both men and women in the space workers' employment line. There will probably be at least two complete crews, each numbering some five hundred: one to work in space and the other to be on Earth between trips to space. If powersat construction rates are stepped up, there will be more crews as well, and there will probably be tens of thousands of applicants for these prized jobs.
The training of selectees will start soon after they are hired and will call for an enormous investment in facilities and simulators. As in any complex project, the size of the training crews will be at least as large as that of the crews they are training. There will be classroom studies and plenty of practice sessions with the equipment they will use, ranging from the most advanced teleoperators to the stew pots used by the cooks. The most realistic forms of training, and therefore the most important, will involve simulating actual on-the-job work activities and weightlessness.
For the on-the-job simulations, the equipment-operator crews and their supervisors will have control cabs, cherry pickers, and instrument panels just as they would have in space. Within the crew members' fields of view will be, not actual space hardware under construction, but rather full-color TV screens, seven feet across or even larger. These will be linked to a powerful computer capable of generating TV images of work in progress. The computer will produce a continuously updated representation of the construction and will generate the TV displays that will show each worker what in fact would be seen if actual construction were taking place.
The actions of individual workers will not produce real beams or solar-cell arrays. Instead, they will generate signals to the computer, which will interpret them as actions that would have produced so many meters of beam or array. So far as concerns actual production, the trainees will be like the tailors who stitched the Emperor's New Clothes in the Hans Christian Andersen tale, whose needles held no thread, whose looms bore no draperies. But when a beam-machine operator switches his controls, he will have the satisfaction of seeing lengths of beam energe on his TV screen. If someone else is to attach the edge of a solar-cell array to the length of the beam, there it will be on the TV screen, as expected. When production equipment would run low on supplies because of long use, teleoperator controllers will simulate the reloading of magazines and stores.
This type of large-scale simulation will not only be a most effective way of training; it will help assure program managers that the powersats will in fact be built according to schedule. For weeks or months on end, crews of trainees can follow their work schedules, as major sections of power satellite emerge and take shape entirely within the computer. Someone may fall sick and be absent; a replacement will have to carry on. Groups of people will be rotated off the job, as if to return to Earth; the new crew members will pick up where they left off. Equipment can be made to shut down or malfunction; managers and foremen will have to devise means of working around the problems to keep production on schedule. People will grow bored, fatigued, irritable; they will demand coffee breaks or times when they can relax and stretch their legs. Inevitably, some otherwise highly qualified people will find the job of a space worker is not for them. It will be much easier to let them go when this involves walking out a door into the Texas sunshine, rather than being returned from earth orbit.
Most of these simulations will be in normal gravity, but training for weightlessness will be very important, especially for maintenance people and for those whose work will require great skill and care, in contrast to the more routine activities of the equipment operators. Training for weightlessness, of course, dates back to the 1950s. For at least that long there have been jet aircraft flying special maneuvers to give their passengers thirty seconds or so of the real thing. Since the mid-1960s, would-be astronauts have trained in a much more convenient and better way: under water.
At NASA's Marshall Space Flight Center is a large swimming pool, seventy-five feet in diameter and forty feet deep, containing over a million gallons of water. This is the Neutral Buoyancy Facility. Properly weighted, an astronaut under water will float freely within it, as if in zero-g. What's more, the tools and equipment he handles can also be made buoyant. Aluminum beams can have small blocks of styrofoam attached at their ends; the beams then behave very much as if in space. These underwater exercises can go on for three hours or longer.
The large size of the tank allows people to work with full-scale structures, just as there would be in orbit. Most astronauts training in it have worn complete space suits, which double as diving suits when weighted with lead. However, the upcoming Spacelab flights will have astronauts working in shirtsleeves, and underwater training without pressure suits has been introduced as part of this program. In these "shirtsleeve simulations," people use only ordinary scuba equipment with a face mask but no flippers. It would be considered bad form to actually swim under water; instead, people move about by grasping hand-holds or climbing along the ladderlike beams of space structures.
The resulting training is indeed realistic. The Neutral Buoyancy Facility saw extensive use during the first Skylab flight in 1973. There the crew leaned the procedures that enabled them to save their damaged space station. When Aviation Week Editor Craig Covault took part in underwater training, he found few clues other than bubbles from other divers that the activities were indeed being conducted under water and not in space.
In time all training exercises will come to an end, and the space workers will be sent to begin the jobs for which they were hired. To get to space, there will be what in the era of the power satellite will count as a golden oldie—the space shuttle. With an advanced booster giving lower cost, the shuttle will be just the right size to carry seventy-five to one hundred crew members to and from orbit. For this, the shuttle's payload bay will hold a passenger module, somewhat resembling a section of tourist-class seating from an old-style airliner like the 707 or DC-8. The seating will likely be five abreast, and there may be no windows, stewardesses, or choice of hot dinners. The passenger module will be like a sealed and enclosed Greyhound bus, loaded by crane aboard an aircraft to be delivered as air freight. The passengers' first experience with space flight thus would present few amenities beyond what would be found in a crowded subway stalled underground, but mercifully the flight will be brief. And there will at least be the opportunity to install a TV screen to show views of the world outside and of the construction base as the spaceliner approaches.
The construction base will be an immense open structure of beams extending for miles. Here and there the discerning eye will pick out the rocket transport centers, perhaps with spacecraft moored beside them. There will be the centers for receiving the great cargo rockets from Earth, for unloading and storing their massive payloads, or for serving as terminals for passenger traffic. The power plants for the construction base with their purple solar arrays will be visible, as will the clusters of modules used for crew quarters. But only close in, when the base appears as a fathomless immensity of beams, will anyone recognize the tiny and widely dispersed work stations scattered amid the vastness.
At the construction base the people will live in crew modules somewhat like the Skylab space station. These will have been built and fitted out on the ground and carried to orbit by means of the large cargo craft. They will be larger than Skylab—say, fifty-five feet in diameter rather than twenty-two, with seven levels with ceilings at little more than seven feet. But whereas Skylab held only three people, each of these units will hold something like a hundred. Three levels will be personal quarters. Two others will serve for storage and for the heating, cooling, oxygen, and waste-control systems. An entire level will be fitted out for the galley and the cafeteria or dining hall, and the seventh level will serve as a zero-g gymnasium.
How will these people live? This question distinguishes the power satellite enterprise from science fiction for science fiction writers rarely have to worry about how their heroes get their laundry done or wash their dishes. The experience of Skylab and the space shuttle will be valuable here. Skylab carried a fold-up shower system, which its astronauts much appreciated. The shuttle carries a sit-down toilet, which can be used in space by both men and women in privacy. In both these items, currents of air mix with sprays of water as a substitute for gravity. There will be a water-reclamation system to recycle the waste water from showers and toilets, as well as from sinks and kitchen galleys, since this water would not be for drinking.
The senior managers will have private suites, since rank hath its privileges even in space; but most of the people will sleep dormitory-style. They will zip themselves into "sleep restraints," comfortably padded, loose-fitting affairs mounted to the bulkheads. It will be possible to provide privacy by means of curtains, and there will be ample drawer or cabinet space for the clothes and belongings of each crew member. When the people get up in the morning, the resemblance to a college dorm will be evident: people lining up to use the shower or to brush their teeth at a sink, or rubbing the sleep from their eyes as they prepare to shave. (The electric razors will come with tiny vacuum cleaners to vacuum up the bits of hair.) Morning coffee will be served in the dining area.
The care and feeding of the crews will receive careful attention, since appropriate concern for the inner man will be one of the ways to maintain morale. Certainly a hot, appetizing, home-cooked meal will be welcomed by tired construction workers at the end of a shift, just as on Earth. The food will include plenty of fresh water, vegetables, meats, milk and eggs, and much else, all delivered fresh from Earth. The frequent cargo flights, one or two per day, will carry five hundred tons of cargo each, so that fresh food will be a minor item on the cargo manifests. Indeed, since the value of the workers' labor will be some thousands of dollars per hour, it will be no great cost to have frequent servings of lobster, fine steaks, or even French or Chinese cuisine. For some people, the most lasting memory of construction days may be the filet mignons. There will, of course, be a full-time cafeteria staff to prepare the food and serve it. But for all that, even the best turkey dinner will not taste the same as it would on Earth. In zero-g there are shifts in body fluids which result in sinus and nasal congestion. Without a sense of smell, foods just don't taste the same.
In the dining areas, as well as in the recreation areas, one of the most crew-pleasing features will be plenty of large windows that face Earthward. The Skylab astronauts all acclaimed the large window in their wardroom. For them, Earth-watching from the wardroom was one of their favorite pastimes. They all wished that the window had been bigger and that they had had more clearance as well as hand-holds, so that they might change their positions as Earth passed below. They recommended that there be an observation bubble.
There will be many other recreations besides Earth-watching. There will be daily movies, college extension courses, videotape TV, and saunas. The space crews will want libraries for books and music, telescopes, and regular mail and telephone service. In addition, since some couples will get married and others will want to act as though they were, it will be important to have a few padded rooms (possibly equipped with stereo and other amenities) where lovers can do what they feel is appropriate.
There will also be recreations of a more conventionally athletic nature. The periphery of the recreational area can readily be a jogging track; there was just such a track aboard Skylab. Interestingly, a jogger running at ten miles per hour would generate one-quarter g of artificial gravity. A sprinter at twice that speed would enjoy a welcome reacquaintance with normal gravity. There would be a large open area for such activities as zero-g acrobatics and gymnastics or handball. In addition, the people would welcome the kind of exercise equipment used aboard Skylab. This equipment included a treadmill for running, a stationary bicycle, and arm exercisers.
The five large crew modules will not be the only ones. The most senior people on the project may have a special module all their own, which may also have rooms for visiting VIPs. Just as the first presidential trip overseas came when Teddy Roosevelt visited the Panama Canal construction in 1906, so a century later the power satellite project may give the occasion for the first flight by the president to space.
Another module will be set aside to serve, at least in part, as the hospital and sick bay. It will provide as well for dental offices—and a morgue. There will be a training and simulation module to work with new construction equipment and techniques. Some of the new people who come to the project in midstream will receive training here. A maintenance and rest module will provide the opportunity to work on large items of construction equipment or spacecraft components in a normal atmosphere. The operations module will house the control center and will serve as headquarters for the managers and administrators. Perhaps this will also be where they will live and where the VIP suite will be located. Finally, a transient crew quarters module will provide for crews who are to be transported to and from their work locations.
There will be several different kinds of crew transportation. The space shuttle and its passenger modules will carry crews down as well as up. A few dozen picked crew members will staff a lonely outpost in geosynchronous orbit to make final preparations when a powersat is to be completed. They will be transported and supplied by once-a-week flights of an Orbital Transfer Vehicle, a two-stage reusable rocket craft. This craft will burn hydrogen and oxygen brought from Earth.
For general day-to-day transport, there will be a network of enclosed buses running on tracks—a space rapid transit system, if you will, to carry people from the living centers to the far-flung reaches of the construction base. Each will carry two dozen commuters in shirtsleeve comfort. An airlock at the end of a cherry picker boom will allow the bus to pick up and deliver individuals to their enclosed control cabins on the construction equipment. Also, small free-flying spacecraft will carry inspectors or maintenance people to areas that otherwise would be inaccessible or hard to reach.
How long should people stay at their jobs? To establish the crew schedules, one can draw on expertise from Skylab, from nuclear submarines and undersea habitats, and from such activities as the construction of the Alaskan pipeline. On the Alaska pipeline, the most that people could stand to stay was eight weeks at a time. Similarly, the longest nuclear submarine missions are limited to seventy days; this is also the longest time that large crews have been confined in laboratory tests. However, medical data from the Skylab flights show that stays in space of up to one hundred days can be accepted. At Boeing studies of this problem suggest that the tours of duty should be ninety days. The work week would be six days, with one day off per week (one wonders if schedules would be adjusted to make the day off fall on July 4 or Christmas). There would be two shifts, each working ten hours a day. The work day would be split into two four-hour sessions punctuated by half-hour breaks for lunch and dinner, followed by a two-hour evening session.
As on the Alaska pipeline, between their tours of duty the space workers will have time away from the project. They will be rotated back to Earth, their pockets bulging with their high wages, ready to enjoy ninety days off the job. Much of that time will be available for vacation, though some may get a few days of refresher training before returning to space. Yet there may be limits to how many tours of space duty a crew member can serve.
The crews will be exposed to radiation while in space. To diminish their exposure, most of the construction will take place at low orbital altitudes, where there is a good deal of protection from Earth's magnetic field. But there is danger from heavy atomic nuclei, such as iron, which stream in as high-energy cosmic rays. They can destroy cells, acting like hot needles as they pass through the body. Their effect on the brain is particularly damaging, since nerve tissue does not grow back when damaged. It has been calculated that on a three-year trip to Mars, some 10 percent of the brain's cells would be seriously damaged or destroyed.
There will be solar storm shelters in the living areas to provide radiation protection during the worst cosmic-ray outbursts from the Sun, but still the space workers will accumulate exposure to the point where they will have to be grounded. What may ground them at an even earlier date could be the effects of living in weightlessness. In zero-g, bones tend to lose calcium and to grow weaker. On the basis of results from the longest Skylab flight, eighty-four days, it appears that in one year of weightlessness a person would lose 25 percent of his bone calcium.
The body tends to replace its lost calcium when back in normal gravity, and indeed the Skylab evidence suggests that workers would replace their lost calcium during their ninety-day times on Earth between tours of duty. However, studies of middle-aged men undergoing prolonged bedrest have found that after six weeks in bed there is a tendency for the calcium replacement to go only part way and then level off. So it appears that long-term bedrest, and possibly space flight, causes permanent and cumulative damage to bones. There are drugs which promote the growth of bone tissue, but whether they will help remains to be seen.
In the whaling towns and seaports of the last century, it was not uncommon to find men who were prematurely aged, bent, gnarled, and dreadfully misshapen. They were victims of the bends: deep-sea divers who had come up too fast from under water, thus releasing bubbles of nitrogen from their blood and producing horrible agony. No one wants space workers to end up in similar straits, suffering from the peculiar diseases of their long-term flights. But if spacemen are to avoid such fates, it may become necessary to limit them to at most a year or so of total space flight in the course of their working careers. Should this happen, the resulting frequent turnover of people would drive up costs. There will be an obvious conflict between the two imperatives: "Protect the health of the people" and "get as much work done per worker as possible."
So there will be two trends that will shape developments once the power satellite program is well under way. The first will be to reduce further the number of people needed in space to build each powersat by bringing in more automation, better teleoperators, and robots. The rise of intelligent computers will greatly hasten this development, and may in fact be brought about by the demands of powersat construction. The second trend will be to provide longer staytimes and more allowable space exposure for those who indeed must live in space. This trend will lead to incorporating artificial gravity and better radiation shielding in work and living areas. There may also be attempts to grow food in space and to produce there the needed oxygen and water by closing the cycles of the environmental systems. With this, the habitations will advance in the direction of becoming true space colonies.
For all that robots and automation will grow in importance, an increasing demand for satellite solar power could well lead to a steady growth in the number of people needed in space, and hence to the sophistication of their living areas. It will be immaterial whether they think of themselves as astronauts, as orbiting work crews, or as space colonists. Specialists all, they will share the deep satisfactions of difficult work well done, of challenges faced and mastered. Whether their space careers be long or short, they all will feel the pride of achievement, of having lived through experiences shared by few. Their actual working careers will not be romantic, science-fiction idylls or space odysseys any more than the experience of building the Alaska pipeline was an exercise in communing with the northern wilderness. But they will share the camaraderie and fellowship of bold ventures, as in the days of Shakespeare's King Harry:
[King Henry V]