Ozymandias
I met a traveller from an antique land
Who said: "Two vast and trunkless legs of stone
Stand in the desert. Near them on the sand,
Half sunk, a shattered visage lies, whose frown
And wrinkled lip and sneer of cold command
Tell that its sculptor well those passions read
Which yet survive, stamped on these lifeless things,
The hand that mocked them and the heart that fed.
And on the pedestal these words appear:
`My name is Ozymandias, King of Kings:
Look on my works, ye mighty, and despair!'
Nothing beside remains. Round the decay
Of that colossal wreck, boundless and bare,
The lone and level sands stretch far away.
Percy Shelly
The Project
All that we have of the ancient civilizations that preceded our own are tatters and little pieces. Whatever remains of literature, music, art, or even architecture we have managed to find are but a microscopic sample of what was there when these civilizations flourished. As the civilizations become more ancient, the remains become more fragmentary and faded. What is preserved is done so by processes which work mostly at random, and therefore the best is usually lost. We have only hints of the magnificence of the former great cultures, hints which still enthrall us, but we are looking at shadows.
The artistic riches of King Tut which so shocked the world were those of an extremely minor Pharaoh. We have never seen the undisturbed tomb of a mighty one. We know also that the Greek tragedies that have survived are but a fraction of what was produced by their authors. The ones that have survived were considered by contemporaries to be the lesser ones.It is universally recognized, however, that even the fragments of Greek and Roman civilization that were carried down the time stream played a pivotal part in the development and history of the western world. The Renaissance was a rediscovery and extension of the ideas, art, and science of the classical period.
This brings us to our own place in the time river. What should we preserve? What can we preserve? In our own time a tantalizing possibility has arisen: Our civilization has now in its grasp something denied to all previous cultures. We can, if we wish, save our best works of art and music for thousands, hundreds of thousands, or possibly millions of years. It is now possible to save not just the written word ( the Sumerians were remarkably successful at this with their clay tablets), but also painting, photography, sculpture, music, and dance.Should we attempt this? I think we should. The legacy we ourselves have been able to preserve from the past is of inestimable value. Our own civilization would be no less valuable to those who will follow us. This is true not merely from a standpoint of "progress" which is a problematic concept, but from the view that for future generations to lose all of Beethoven, Van Gogh or Shakespeare would be a loss of irreplaceably unique products of the human spirit.
The technology now exists for saving a painting in a form so accurate that ten or twenty thousand years from now, our descendants could hold a reproduction of Rembrandt's self portrait that was as sharp and as perfectly colored as the best printed photographs of it are today. We could save our musical performances intact, right down to the particular voices of our singers, for 100,000 years, and they would be as clear as our present recordings.We could save not just the flotsam and jetsam and the faded tatters that we have scraped together from previous civilizations, but a clear and accurate sample of the present as lived today, a true window into the past, from our movies to our art, unfragmented and in no less detail then we see it ourselves.We could in fact preserve a representative sample of our greatest literature, art, and music in a very small space at a cost that would not be prohibitive. I think that if many people became aware that this possibility was real and practical, they would see upon reflection that this would not only be worthwhile, but a major accomplishment of our age.I am proposing that an international institute be set up to implement this project, which I have dubbed the Ozymandias Project (from Shelley's poem), for the sake of extracting the maximum irony from the situation.
The Technology
The key to this possibility lies in current digital technology. The central concept is very straightforward: Information can be stored in only two possible manners: analog and digital. Analog information must, in the course of time, be lost, digital need not be.A photograph of a painting is an analog storage method. A piece of music stored on a phonograph record is also analog. In the case of the photograph, small areas of colored dyes correspond to the original distributions on the painting. If the painting is representational, then it could also be considered an analog representation of the original scene, with areas of paint representing colors and shapes of the original subject. In the case of the music recording on an L.P., the width of the grooves represents the loudness at each point, and the density of the undulations represent the pitch of each note.All information that is stored in an analog way must eventually be totally lost. This is an unavoidable and inflexible constraint on this mode of recording. The painting will deteriorate, and a thousand or certainly a million years from now it will be dust. At first one might think one could get around this by taking a photograph of the painting when it was new, and then simply re-photographing the photographs every few years before they faded. Unfortunately, every time you make an analog copy, you lose a small amount of information. In the case of a photograph each successive copy is slightly fuzzier then the one before. The thousandth copy would be very indistinct. By the millionth, it would be unrecognizable.In information theory this is known as adding noise to the signal, and it is theoretically (and practically) impossible to make an analog copy without adding noise. If we take the example of a phonograph record, then the millionth copy of any piece of music would simply sound like a steam kettle hissing (called white noise by audio scientists).These conventional ways of recording music and art would lead at last to their ultimate total loss.The digital mode of information storage is quite different. This method is widely used today in such technology as digital audio disks (CD's), Digital video disks(DVD's), hard disks in a computer, and other electronically oriented information handling. However, digital storage is not new, it is, in fact, a method in use since the invention of writing.The term digital is somewhat misleading, actually, in this context it has no necessary connection with numbers. Digital information storage is just storage where an arbitrary unit( a mark, a stick, an indentation in a clay tablet) represents something else. Any alphabet is in this sense digital.When copying information represented in this form, the mechanics are altogether different. If you take the word "ALBATROSS" and copy it over and over again there is no necessity to lose information. An A is either an A or not.
If you have an A in an old manuscript with a piece missing, and a patient monk is making a copy, when he sees:
He recopies it as " A ". The copier is not copying the appearance of the letter, but only which letter it is. Therefore faint or partial letters are reborn as new ones. The "erosion" is not passed on to the copy.
If you want to make a perfect copy of a manuscript, all you have to do is make it before it is significantly faded, and your copy can be exact, not just approximate. The one millionth copy can be identical to the original. Of course, you also have to make no mistakes in transcribing, and this may seem inevitable, but there is a way of getting around this problem which I will explain in a moment.The central idea here is that for information encoded in digital form, a copy made from a slightly worn original need lose no information, and there is no mathematical necessity to lose information with successive copying. Or, to be more precise, the loss of information can be made as small as you wish.This implies that if you could get a Van Gogh in digital form, then the ten millionth copy in the year 20,000,000 could be as perfectly sharp as the original copy.Well, we can get a Van Gogh in digital form, and the technology is already available, in fact, it's in widespread use.First, I have to explain why we do not have to make mistakes when making copies. Anybody who has typed knows typos are always present. Even a machine which scans print and then produces digital code is expected to make occasional, if rare, mistakes. However rare, these mistakes should accumulate in successive generations of copies, so that by the 27,000,000th copy, for example, we should be left with gibberish. We would expect the same kind of noise we were plagued with in analog copies.We can, by clever footwork though, reduce the number of these necessary mistakes to almost zero, and in fact to as close to zero as we want! The trick lies in making not one copy of the original, but multiple copies, and then comparing them.If for instance we had the word " ALBATROSS " in a manuscript, we could have ten typists copy it. When the copies started to fade and it was time to make the new copies, we could compare each letter in all ten copies.If one of them had the word " ALBAFLOSS ", we could simply disregard that version and copy only the others. This is, of course, terribly tedious and time consuming for humans, but it is exactly what computers are born to do. The computer need have no judgment, artificial intelligence, or anything of the kind, it can simply be programed to go with the majority vote on any discrepancies. To make the chances of copying an error into the next generation as close to zero as we want, we simply make a larger number of copies of whatever we are duplicating, (each copy must be made independently). Then we always compare all the copies, letter by letter, before making the next generation copy. The mathematical chances of, say, six of ten copies having the same random error, and therefore winning the "vote" and being copied mistakenly as the correct letter, are infinitesimal. You could copy a novel through several million generations without expecting a single error to come through. We can get as close to perfection as our budget allows.Considering that several hundred copies of a thousand page novel can be placed on one optical disk, and all the copies compared letter by letter in less than a second by a computer, this is not difficult in modern practice.Literature would of course be the most easily encoded this way ( In fact writing is encoded into a string of ones and zeros every time one uses a word processor. ) Painting, photography and even sculpture in its full three-dimensionality could be encoded this way by scanning it with light sensors. This is the technology employed in digital cameras and video recorders, and also in the higher resolution imagers used in telescopes. These give an output which can be reconstructed as little squares called pixels. The pixels are then put onto a video screen, or printed. This is how NASA transmits pictures across space-- as a string of numbers which encode how bright each square is, and what color it should be. It is how computer monitors and printers work. The pixels can be stored permanently as a digital array, and multiple copies compared before recopying in the same way as letters. You can make the picture as sharp as you want by simply splitting the original image into more and more pixels. Current technology can easily store a color photograph as a digital string, and then reproduce it to create a new photograph that can't be distinguished by eye from the first photograph of a painting. (A Van Gogh would be better stored as a sculpture, because the paint is so thick, but this merely requires more storage space than a flat image.)Music is digitalized and stored as a pattern of dots on a compact disk. Every second of music is chopped into at least forty thousand pieces, each of which is reduced to a number of dots. Your disk player then reads the dots. When it is put together again, most people don't seem to notice that it isn't actually continuous. This information too, could be duplicated and compared "dot by dot" by computer before recopying.Thus could be preserved indefinitely not only Beethoven's music, but Beethoven's music as interpreted by current musicians. Ten thousand years from now people could listen to Charlie Parker, Glenn Gould, or Jascha Heifetz as clearly as we can today. (Better, if their stereos have improved).
What should be saved?
Now that I have indicated how we can actually do this, I think we can lay out the following broad outline for the Ozymandias project:An international committee would be set up to answer the following questions:
1) How far into the future can we feasibly attempt to preserve our works? (I think ten thousand years is a minimum goal.)2) What is the best and most cost effective technology out of the many available?3) What should we choose to preserve?
The answers to the second two questions depend entirely on the answer to the first. In order to set a goal for the longevity of the information, we need to make certain guesses about the far future. The technology could be structured in one of two main ways: On the assumption that civilization will continue in its present form, and on the assumption that it will fall into a regressed state, or change into a kind that would not share our goals.Our guesses on this are crucial in deciding our strategy. If our civilization will continue "as such" for the whole period in which we intend to preserve, then we can just set up a small office building in any location deemed geologically stable, and set aside a fund for a maintenance and secretarial staff. Their duties would be straightforward: First the material that was selected for preservation would have to be "inputted" (typed in, or scanned and then corrected). Then the duplicates would have to be made. The only duty left would be to simply recopy the duplicates at set intervals by the comparison method previously discussed. The recopying would be an automatic, fast and cheap process done entirely by computer. New material chosen for the honor of preservation would be added from time to time.For this institution (call it the Ozymandias Achieve Project) conventional magnetic or optical disk storage would be most cost effective. The above structure would be very inexpensive; once the original work involved in gathering and inputting the material is finished, it would probably cost less to run than a single post office for a middle-size city.In this "stable world" model the information would be saved simply by recopying periodically before it degrades (about once every ten or twenty years). What is more, if advances in technology produce much better storage media in the future, all the stored information can be simply transferred onto the new medium without any loss.
In the Eventuality of the Fall of Civilazation
This scenario, however, is probably unrealistic. The oldest continuous civilization so far has been the Chinese, which is about 5000 years old. Most other civilizations have lasted at most for 3000 years, and there is no particular reason to believe ours will last longer. In the event of wars, or the real collapse of our civilization, the storage site would have to be self-maintaining, very difficult to get into, and not dependent upon trained and willing personnel to insure its continuity. If we are so ambitious as to want to bequeath our works to very distant generations, say 100,000 or 1,000,000 years into the future, we will have to prepare for some pretty rocky times between now and then.Not only would such an installation need to be subterranean and fortified like a missile silo, but it would have to be sealed against entry like an ancient tomb, and yet marked in such a way that far future generations could find it. I would call this version of the project the Ozymandias time capsule. The idea would be to make it barbarian proof, but accessible to civilizations of our technological level or better. In order to make this very subterranean capsule findable by a technological civilization, it could be tailored to yield an unusual radar echo, or have an anomalous magnetic field, for instance, so that any civilization with radar or geomagnetic mapping satellites would notice it.We would also have to use a very different technology when making the records. We could not rely on conventional magnetic storage or optical disks, for the simple reason that they are made of plastic and deteriorate quickly. It would not even help to have a set of machines which made duplicate copies at regular intervals, even if these machines were self maintaining and had a nuclear reactor as their power source. They would still have to use plastic from stored reserves, and on the time scale we are talking about it would all be dust. Only if the installation could manufacture new plastic and other materials by itself could it maintain itself indefinitely. Active, self-fixing and maintaining machines of this nature are beyond current technology.The practical answer to the question of how to preserve a Rembrandt for one million years is a bizarre mix of the very modern and the very ancient. The method owes as much to Sumeria as the computer chip. The key is that the information is digitally encoded, but on such stable materials that no copies need be made.First take the painting and scan it with a camera that turns its image into digital output Then get a printout of the digital code that represents the painting (usually a very long string of ones and zeros). Then take an extremely durable material like ceramic (the Sumerians used clay tablets), or stainless steel rods or granite. Drill, chop or shape the material to encode the digital sequence. Perhaps you could take a granite slab and drill micro holes in it, with a hole representing zero, and any space between the holes representing one:00 000 0 0 0 00 00 0Now you can leave the slab for posterity, along with five or ten copies each of however many other slabs you feel represent your best achievements.This is of course enormously more laborious than normal procedures, because you need materials that will last for millenniums. Moreover, a single painting encoded this way, even with holes the size of thin spaghetti, would take up a room, and weigh tons. In addition to your library of slabs, it is necessary to leave clear instructions for their decoding. You could leave a particularly indestructible rock (say a large slab of quartz) which would be the Rosetta stone for your stored slabs. Since you can't assume anyone will still speak English, or anything like it, you leave a cartoon story depicting what you have done, how it was encoded, how to decode your sequence, and whether to "play" each part of it as music, pictures, or writing.The only real difficulty comes with literature. Music is music, and art is art, they explain themselves, but if all current languages are totally lost to whoever digs up your capsule, then no matter how good the copies are it would still be meaningless. (What would ZYGMORPT mean to you if you found it on an ancient cave?) The easiest way out of this impasse is to start an A,B,C primer with cartoon illustrations. Explaining words like "irascible" or "poignant" with cartoons is slightly tricky. At least we will have the whole art file, and as many illustrative photographs as necessary to explain each word.We must assume that people will still be human, and therefore able to interpret cartoons. We must further assume they have the fairly high level of technology necessary to get into a well-sealed installation, and that they will have the necessary machines ( digital technology) to recreate the image.The idea is that if they are savages they won't be able to get into (or even find) your underground installation. If they have a technology equal to or better then our own, it will be child's play to decode our records, and out will pop the Mona Lisa, or The Marriage of Figaro, in perfect clarity. I wonder if they will like them? Our Science and Math need be saved too. Although any civilization which can enter and decode the capsule will by necessity have a high technology, there may be gaps in their science that we would fill in. (It is possible to imagine a science with a fair degree of physical sophistication but without any knowledge of DNA for example). Even something like General Relativity may not exist in the future, being an example of a unique approach and quirky viewpoint. Our Mathematics too, may have looked into areas that other civilizations will not have. (Group theory for example, may be never invented again). The last question is how the selection of the "forever file" is to be chosen. Interestingly enough, if the first scenario is in effect (the office building, in an uninterrupted civilization), then, surprisingly, the best way to choose which works deserve semi-immortality may be not to choose at all. Current technology has already produced miracles of compactness. It has achieved information densities of 100 billion bits per square inch in magnetic recording. Using micro pointed needles (Atomic Force Microscope) technology, they have been able to achieve ten times that density on a plane surface. Optical researchers have now begun to exploit the potential of using three dimensional rather than flat mediums to store data. Using lasers to create microscopic irregularities in solid crystals and polymers, they have demonstrated densities of about one trillion (1,000,000,000,000) bits per cubic centimeter (16.3 trillion bits per cubic inch). Each letter of a word takes up about one byte or eight bits. A pixel takes from one to six bytes to store. At this density, a library's worth of words would fit into a sugar cube, and all the world's written literature, in all languages, could easily fit into a shoe box. It might be cheaper (and certainly more egalitarian), to simply encode all the books ever written, and all the music now known, a selection of the photographs, and all the well known paintings, into one huge data base than it would be to hire many thousands of professors and experts who would acrimoniously debate the relative merits of each book forever. We could simply save all of it, including ten redundant copies, in a cube the size of a refrigerator and let poor posterity decide which of it is any good.Movies and even television could also be saved, but this would require millions of times more storage space (and television would raise certain embarrassing questions about our taste, or even our sanity with our decedents).Still, if we desire it, there is no technical reason why "I Love Lucy" could not sit magnetically encoded in the lap of Faust.
If, on the other hand we wish to save our civilization even against its own demise, with savagery in place ( at least locally ), then we cannot use a technology based on copying, since nobody will be reporting for work at the archives. This second, more drastic scenario ( The Time Capsule), calls for a different kind of project altogether. We would have to construct an installation with "permanent" copies, although they may be redundant to insure accuracy. Storage becomes much more expensive and space consuming. The situation might be much brighter than depicted in the "stone slab full of spaghetti holes" scenario described above, however. The question of how best to encode digital information in permanent materials has only begun to be investigated. If, for example, we use a laser to drill micro holes in an opaque ceramic, and then fill the holes with a clear glass with the same chemical properties and the same thermal expansion as the ceramic, we might get a very high density medium that would last for many millennia. In this, more circumscribed, project, only a small representative sample could be preserved, along with the necessary decoding Rosetta stone.An international committee would be set up which would take (I assume), many years to resolve what is truly great and worth saving.Although our music and art would, therefore, be somewhat culled, we would nevertheless be able to preserve a vastly more complete and representative sample of our best than we have ever received from any previous civilization, and it would be passed down in pristine exactitude, not in faded, ghostly suggestions.
Why?
We should begin to look seriously into what would be necessary for this effort now. The motivation for doing this boils down to the subtle but deep feeling that the accidents of time, culture and individual spirit that give rise to the creation of a great work can never happen twice. That some of what we have, be it Hamlet, or Beethoven's Ninth Symphony, is simply so inherently wonderful, so beautiful, or so wise, that it constitutes a gift to future human beings worth great effort to bestow.
contact: David Green
greend@newschool.edu