Slashdot Mirror
Ask Slashdot: Permanent Preservation of Human Knowledge?
Wayne2 writes "While there have been many attempts to preserve human knowledge in electronic format, it occurred to me that these attempts all assume that human civilization remains more or less intact. Given humanity's history of growth and collapse with knowledge repeatedly gained then lost, has anyone considered a more permanent solution? I realize that this could be very difficult and/or expensive depending on how long we want to preserve the information and what assumptions we make regarding posterity's ability to access it. Alternatively, are we, as a species, willing to start over if we experience a catastrophe, pandemic, etc. of significant magnitude on a global scale that derails our progress and sends us back to the dark ages or worse?"
It "protects" content right?
http://en.wikipedia.org/wiki/Wikipedia:Terminal_Event_Management_Policy
Rosetta, stone tablets, parchment scrolls and other works which have survived destruction only by obscurity, sleight and secrecy which instructs that the methods are not as important as the means to which you secure knowledge for posterity.
...for Leibowitz
http://en.wikipedia.org/wiki/A_Canticle_for_Leibowitz
... store knowledge within.
One wonders what would could be saved if things like pyramids and tombs are used to save a cubic ass tonne of knowledge.
Check out the Rosetta Project - http://rosettaproject.org/about/
If you're planning for the fall and rise of civilisation, you need to prepare for the possibility of deliberate destruction - it's possible that a future civilisation might be so sickened by the actions of the past they seek to destroy all their works, or a religion might emerge which considers your documents heretical and in need of destruction, or perhaps a king feels that his people are living in the shadow of legendary greatness and only by destroying the legend will their story be honored.
So you're going to need to mass-produce whatever storage media you choose - make them by the millions and put them all over the world. In museums, in caves, burried or sunk offshore (Add a big chunk of iron, ready for when the metal detector is reinvented), as many as you can. So many it'd be impossible to destroy them all.
As for the actual storage medium? Tiny etchings on iridium would work. It's corrosion-resistant, and very, very hard wearing. It's last for millenia with ease, even in burried in moist soil or scoured by desert sand, and with such a high melting point it'd be untouched even if the containing building burned down. The only issue is the price: That stuff is expensive. Really expensive. It's cheaper than gold, but not by much.
I'm not sure that I can really think of good examples of this happening - at least not on a global scale. Sure, there was a regression in Europe after the Greeks and Romans. There were quite a few works lost. And it seems that there was a very early civilization around India that was abandoned (apparently due to crop failures). But the main protection for lost knowledge seems to be to spread knowledge around the world. The world has never simultaneously regressed (the Middle East and China weren't doing so bad during the European dark ages). The works of the Greeks wouldn't have been lost if their writings had larger distribution (instead of being confined to a relatively small area, which makes the fate of those earlier works dependent on the local conditions). As long as people keep writing and reading books, I don't see how much knowledge is going to be lost. There wasn't even much knowledge lost in Europe during the Black Plague - and that killed off 1/3rd of the people in Europe.
Perhaps the concern over "lost knowledge" says a lot about people's perception that some massive apocalyse is going to happen. I think, in general, people tend to grab onto ideas about "apocalypse" (which necessarily results in some massive social rearrangement) because they're not happy with the state of the world. Apocalyptic thinking is a little bit of a fantasy about starting over.
Books. It worked before, it should work again.
The electronic preservation angle was my wife's thesis.
http://explorer.cyberstreet.com/CET4970H-Peterson-Thesis.pdf
Unless the genetic information is being actively selected for in some way, random errors and natural selection pressure will quicky weed it out. The Star Trek episode scenario, while being quite cool, doesn't make sense, biology-wise.
Ezekiel 23:20
Please note that they weren't fired, originally. They didn't have that much fuel in Mesopotamia to fire everything they wanted. Ironically, many of the preserved tablets come from libraries that burned down in random fires. These events stopped being celebrated by archaeologists after Middle Easterners switched to other writing materials around 100 CE.
Ezekiel 23:20
Barring the development of a strong-AI level pedagogical expert system that can be stashed away somewhere, the task of actually preserving the present state of human knowledge in the absence of the background is pretty difficult.
Mere storage is actually the easy part: Even clay tablets have a modest survival rate when you burn the civilization that inscribed them down on top of them, and with modern materials and machine tools, we could mass produce something better(or really, really, really mass produce something cheaper, and distribute it all over the world).
The trouble comes once you start dealing with knowledge that exists largely in the form of continually-refreshed human capital, and with tools that exist largely in the form of a long chain of worse tools building better tools building better tools, etc. The amount of pure written knowledge you would need to restart/rebuild all the supporting industry to refill, say, a totally undistinguished hardware store, would be considerable, quite probably more than actually is written down(rather than learned on the job by the new guy from the old guy, and fabricated on tools that were built with parts fabricated with tools that go back to the early 20th century if not earlier).
You also run into encoding problems. "Graecum est; non legitur", and that was the allegedly educated class in a civilization that probably had some greek speakers available(and it'd hardly been a global thermonuclear holocaust that ended classical civilization). You'd need to choose some human languages, and god help you with the digital file formats...
Periodically send up long distance spacecraft loaded with not just data but the means to view it and the means to rebuild from first principles, assuming a child was viewing it - here is how you find iron deposits, mine and refine them, this is how you forge ploughs, these are the basics of algebra. Have them programmed to circle around somewhere just inside Jupiter's orbit, and have multiple stations here on earth sending out a deadman signal - when they stop broadcasting, the vessels begin to return in waves seperated by ten years or so, with the last waves arriving once a century.
When they make it home, have them attempt to locate likely inhabited areas whether by thermal imaging looking for fires at night or just vegetation profiling for fields, then drop down nearby, broadcasting light and sound, even radio, until someone comes to investigate.
It's relatively easy to permanently preserve all of mankind's knowledge, just pack it in a rocket and send it Oort-cloud bound. Well permanently as in astronomical timescales. The trick is to preserve all of humanity's knowledge in a way that's useful to humanity in the future.
"The message will be accompanied by a short video message by Wikipedia founder Jimmy Wales, and images required for the re-creation of fundraiser banners."
I can tell it's definitely the real deal and in no way an April Fool's joke!
I think that's a good place to start, but more detail is always interesting:
I would start with a Rosetta Stone, specifically designed to explain as much of the language and text as possible. Included in those would be directions, both in terms of location and recovery procedures, to a much larger collection of paper documents stored in sealed casks of an inert gas to prevent degradation. From there it would be possible to at least describe the basic procedures and formats needed to read much, much denser long term storage options, I thought I remember reading about modified DVDs that would be stable across centuries.
You can either assume that the discoverers will reinvent basic electronics, or, if you have the capacity in your paper archive, lay out a plan that would get them there. If you assume say early 1800's level tech for example, would it be possible to bootstrap them to reading data from a DVD using only printed word? Could you describe the materials, designs, manufacturing techniques necessary? Would they even care enough to try to follow the directions? If you can get them that far you could have petabytes of data stored and ready for use... but that's a big if.
While that's true to some extent, it doesn't mean that knowledge shouldn't be preserved in a format that would be accessible by a recovering civilization. Just because they don't have electricity now, doesn't mean they never will, and a handy guidebook telling them how those things work will speed things up later.
It does mean though, that the information should be prioritized - there's a T-shirt/poster floating around the internet full of "things to take credit for discovering if you go back in time". Most of the items it lists are either critical discoveries that led directly to improvements in quality of life, or were the basis for other technologies. Pasteurization, antibiotics, electric generation, radio, flight, and more. (It's here by the way.)
A guide like that is a good start - build things up in stages, add in more (useful) detail, never assuming that the reader will already have a tool unless it has already been explained how to make it. Then if you want to go hog-wild, after you've reached the part explaining how to make a computer and digitize information, put the stuff that would require a heavily industrialized civilization into a digitized code format and explain how it's encoded, so they can read it when they're ready/able to use it.
Random data being used for research though, is likely totally useless. Not only is the DNA/RNA sequence from that rat likely to be useless to a recovering civilization, depending on what sort of cataclysm happened, the DNA/RNA of a rat may not even match what was recorded. Leave stuff like that to DNA/Seed banks, unless it's part of an explanation of "what DNA/RNA is", and even then, the whole set is pointless. (Also probably patented.) A Tokamak reactor may not be useful to a low tech civilization, but with the boost provided by being taught how to make hydro-electric generators, lights, heaters, radios, and internal combustion engines (they can run on cheaply made alcohol, they're just less efficient that way, and wear out faster.), they might be able to make use of that information in only a few generations.
The real problem of course, is format, and ensuring that not only does the information survive, but that these future people are able to understand it when they do see it, rather than thinking "Oh, pretty metal plates with squiggles on them. I bet I could melt those down and make a great set of knives out of them."
about to tease you for saying something doesn't make sense
Why doesn't it make sense?
while your sig refers to a bible verse
These Bible quotes are all the rage! I didn't want to be left behind.
Ezekiel 23:20
I regularly carve pictures and patterns into various rocks around my property. I often wonder what future scientists will think of them. And now I wonder if someone will try to construct something meaningful in the crap I leave around my ranch...
The data may not be useful immediately, but presumably society would begin rebuilding at some point.
It may be a long time before the information is useful, but once that time arrived, it would save a great deal of wheel re-invention.
Most of the books from classical times were passed down by copying them periodically. Very few original texts from that era, save on stone. Generally educational classic or important religious works were worth copying.
Oh sure, that will work for very low densities of information, but what about something the size of the Wikipedia? That article states that the Wikipedia has over 2.4 billion words across over 4 million articles. The article has a nice visual image of what would happen if you took all that information and printed it into 1000 page encyclopedia volumes (each containing 8 million characters). It totals over 1800 print volumes.
Now, where are you going to find that much stone writing surface in one place, and how are you going to economically carve it in a reasonable lifetime, and how are you going to arrange it in a fashion that it's human readable/explorable?
Even reproducing something immensely valuable for a recovering industrial society like Machinery's Handbook in stonework would take an immense amount of space, time, and money to do. Just something as simple as the Georgia Guidestones cost about $225,000 to do.
No, try again when you come up with something practical.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
Boring? Are you insane? Do you realize how much actually *useful* stuff we manage to glean from garbage piles and laundry lists, as opposed to illuminated genealogies of royal fuckers? Who cares if Henry the whateverth had ulcers on his left leg or on his right leg? Did his peasants wear socks? How often did they wash? What were their daily concerns? *That* is useful stuff for any modern scholar.
Ezekiel 23:20
If you're not familiar with The Long Now Foundation you should check them out. They have a project to build a clock that will last 10,000 years (about as long again as there's been civilization on earth), and are making progress constructing it in a cave in a mountain in Nevada.
Of course, the next questions are things like "well, who is going to be around to read it?" and "how will they read it?", and "how do we maintain a level of civilization where people can create replacement parts for it?"
Neal Stephenson consulted with them for his book Anathem, which I highly recommend, which is based around these sorts of questions.
I would suggest using material that will not be used for firestarter.
The optimists:
http://longnow.org/
the pessimists:
http://en.wikipedia.org/wiki/Earth_Abides
the second kind doesn't need any storage of information I would think.
Some might not even call them optimists or pessimists.
Je me souviens.
So has anyone ever looked at cockroach DNA with a crypto algorithm on a supercomputer?