For media to endure, whether we’re talking years or millenia, you need to answer three questions:
What to record it on.
Where to store it.
How to encode it.
The first is, perhaps, the easiest to answer, we have cave paintings from the stone age, velum and parchment that have lasted thousands of years, clay tablets that have similarly endured and various other historical artifacts that could be considered ‘media’. Alternatively, there are more modern techniques, such as etching into plates of non-coroding metal or other substances, or encoding in stable chemical structures such as DNA. Each approach has different benefits, but largely depends on the second question, where to store it?
Storing the media is the next question, and the answer is going to depend on both what media you’ve chosen and how long you want it to last. Somewhere dry, solid and geologically stable probably makes the most sense, but it also needs to be accessible to those you hope will find the information later. The Voyager disks are a good example of long term storage that will probably outlast the human race, but they’re not exactly handy to retrieve. The same goes for the various efforts to send archives to the moon; they’re more for posterity than use. Finding, or digging, a tunnel in a bedrock type mountain, far above current, or expected, sea levels is probably a good bet. Don’t forget to record how to find the repository, and find a way to keep that knowledge available. That could be a record in another repository, or maybe engineered into the landscape such that signs point to it.
The last question is how to encode the information. If you’re aiming to store the information for an extreme duration, you have to assume the entity finding it will have little or no cultural, technological or language connection with you. Pictures may work, we’ve learnt a lot from stone-age cave paintings and Egyptian tomb paintings, but they can only convey so much. For textual information you’re probably going to need multiple layers, the first being illustrated and readable without special equipment or techniques, and showing how to access the latter layers. The next layer should include information about how to read the rest of the layers (do you need magnification? If so, how do you do that? Have you included a lens that could be used? Do you need special illumination, or other techniques? How should the reader do that?), information about the language(s) used (perhaps a Rosetta Stone type artifact to help translate if they know any of the languages) and information about things like the units used (how long actually is a metre? What is a second?). The next layer would explain any envoding used for the rest of the information. If it’s binary stored as pits engraved in stone, or DNA base pairs, or holographically stored in metal plates, how do you extract the data, and how do you convert the naughts and ones to text for example. You must assume the entity processing this has no common base with you to work from, so everythis must be explained in detail. Finally you can store the information you actually wanted to store in the final layer.
These concerns hold whether you’re trying to store information for millenia or just a few decades. For instance, if I handed you an 8" floppy disk containing a Wordstar file, could you read it without more information? Even once you’ve found an appropriate drive (very rare now-a-days) and a machine that can use it (likewise) you need to hope the disk has been stored appropriately.
If you’re trying to store information to be accessible to humans over a timespan of no more than a few lifetimes you’ll probably be better served by arranging for it to be reencoded and tranferred to new, modern, media every few years.
For media to endure, whether we’re talking years or millenia, you need to answer three questions:
The first is, perhaps, the easiest to answer, we have cave paintings from the stone age, velum and parchment that have lasted thousands of years, clay tablets that have similarly endured and various other historical artifacts that could be considered ‘media’. Alternatively, there are more modern techniques, such as etching into plates of non-coroding metal or other substances, or encoding in stable chemical structures such as DNA. Each approach has different benefits, but largely depends on the second question, where to store it?
Storing the media is the next question, and the answer is going to depend on both what media you’ve chosen and how long you want it to last. Somewhere dry, solid and geologically stable probably makes the most sense, but it also needs to be accessible to those you hope will find the information later. The Voyager disks are a good example of long term storage that will probably outlast the human race, but they’re not exactly handy to retrieve. The same goes for the various efforts to send archives to the moon; they’re more for posterity than use. Finding, or digging, a tunnel in a bedrock type mountain, far above current, or expected, sea levels is probably a good bet. Don’t forget to record how to find the repository, and find a way to keep that knowledge available. That could be a record in another repository, or maybe engineered into the landscape such that signs point to it.
The last question is how to encode the information. If you’re aiming to store the information for an extreme duration, you have to assume the entity finding it will have little or no cultural, technological or language connection with you. Pictures may work, we’ve learnt a lot from stone-age cave paintings and Egyptian tomb paintings, but they can only convey so much. For textual information you’re probably going to need multiple layers, the first being illustrated and readable without special equipment or techniques, and showing how to access the latter layers. The next layer should include information about how to read the rest of the layers (do you need magnification? If so, how do you do that? Have you included a lens that could be used? Do you need special illumination, or other techniques? How should the reader do that?), information about the language(s) used (perhaps a Rosetta Stone type artifact to help translate if they know any of the languages) and information about things like the units used (how long actually is a metre? What is a second?). The next layer would explain any envoding used for the rest of the information. If it’s binary stored as pits engraved in stone, or DNA base pairs, or holographically stored in metal plates, how do you extract the data, and how do you convert the naughts and ones to text for example. You must assume the entity processing this has no common base with you to work from, so everythis must be explained in detail. Finally you can store the information you actually wanted to store in the final layer.
These concerns hold whether you’re trying to store information for millenia or just a few decades. For instance, if I handed you an 8" floppy disk containing a Wordstar file, could you read it without more information? Even once you’ve found an appropriate drive (very rare now-a-days) and a machine that can use it (likewise) you need to hope the disk has been stored appropriately.
If you’re trying to store information to be accessible to humans over a timespan of no more than a few lifetimes you’ll probably be better served by arranging for it to be reencoded and tranferred to new, modern, media every few years.