A key idea in cybernetics is the self-reproducing machine: a machine that can make a perfect copy of itself, and importantly, a copy whose copy can itself make a copy.
Sorry I find this rather depressing. I'm gonna go out and appreciate what's left of the natural world we already have which is far more amazing that anything our somewhat limited minds can envision. If there are 13 types of intelligence (did I get the number right?) why do we always seem to end up using just one? Cheers.
What kind of human must exist for that seed to grow?
Because a seed does not grow in every soil.
The same library placed in two societies may produce entirely different civilizations depending on the psychological maturity, values, and relational capacities of the people rebuilding it.
In other words, civilization is reproduced not only through information, but through the people who interpret, embody, and transmit it.
Knowledge tells us how to build.
Our psychological development shapes why we build, for whom, and toward what ends.
Perhaps that is the deeper recursive loop.
A civilization ultimately reproduces itself by cultivating the kinds of people who will, in turn, reproduce the civilization—as much through their way of being as through the information it preserves.
This article seems to treat civilization as fundamentally an information processing problem. But what if civilization has mission-critical parts that are more experiential, like trust, shared norms, systems of authority, rituals, institutions... A chemistry textbook doesn't create chemists. A constitution doesn't create constitutional government. A medical library doesn't create physicians. Those emerge from communities with practices that transmit more than explicit knowledge. So maybe, instead of asking "What is the smallest library that can reboot civilization?", a better question to ask is "What is the smallest self-sustaining social process that can regenerate civilization?"
I suspect you are correct. Just like an acorn can't make a tree without soil and environment, there is very likely social soil needed. What do you think the minimum would be?
Not sure, but at the very least it would have to reliably produce people who can trust one another and who share a sense of a future beyond themselves. You would need both. Trust alone can sustain a village or a business partnership. But building cathedrals, governments, and scientific traditions all require people to orient themselves toward something that outlives them.
Now call me old fashioned, but is not this whole concept somewhat flawed?
If it's reset then we cannot assume an environment as we have. As reset takes time, will there be anyone who can read? Who teaches this knowledge to others & why would the 'old gurad' be let anywhere near those who've managed to survive?
Personally, I'd be wanting find my own way in order not screw up again...
This reminds me of a project to design 50 open source tools that are needed to "build a small, sustainable civilization with modern comforts." I believe their effort has stalled, but it is an interesting goal.
I really appreciate the sentiment behind these ideas and indeed have a post-industrial SF novel about half written that includes a quasi-monastic order that builds and maintains a Great Library "to save what is worth saving". The novel is set 700 years from now on the NW coast of North America (Cascadia). In the world-building I explored many ideas for preserving knowledge AND a population that can make use of that knowledge. It turns out medieval monasteries already optimized post-industrial data storage: vellum, ink and a lot of copying! In my novel, the literate community that grows up around the library turns out to be just as critical as the saved knowledge; for what good is knowledge/information if no one can read or understand it?
The entire thought experiment has convinced me that the people of the future will just have to be clever enough to figure out whatever they need to know. We figured out what we needed - they will too. Or not. And to be quite frank, there is a nearly infinite volume of data existing today that is utterly useless outside of our fossil-energy/industrial/globalized context (and much that is utterly useless anytime!). So the books that must be copied onto vellum (and eventually recopied, ad infinitum) are mostly conceptual rather than technical and more in the field of the humanities than the sciences. I'm actually more intrigued by what copying these important books by hand might have on the scribes and the culture of the local community. One effect might be very interesting conversations at the dining hall tables!
I don't think our technology really exists outside of industrial civilisation. Wouldn't you be better off documenting in detail how to kickstart the industrial revolution, than to document how to make paper?
If you want a seed that doesn't require billions of humans and the industrial revolution etc, you probably need to really strongly reconceive every piece of the tech stack that you're going to include, to be on a more self contained path. Harder than it sounds I think! It'll be extremely tough to scrub out industrial and pre-AI tech assumptions. Like if you are trying to describe how to refine petroleum you might have wandered off the track? etc, lots of etc.
I think at this point LLMs and future AI are the key. An LLM is an incredible compression of all of our knowledge. Wait a bit, and I suspect we're going to get new computing technology that doesn't require multibillion dollar fans and the modern industrial ecosystem, but rather that you can put together (grow?) with solar, batteries, personal robots, all guided by local synthetic intelligence. If you can run LLMs on that, then there's your seed.
Sorry I find this rather depressing. I'm gonna go out and appreciate what's left of the natural world we already have which is far more amazing that anything our somewhat limited minds can envision. If there are 13 types of intelligence (did I get the number right?) why do we always seem to end up using just one? Cheers.
You ask:
What is the smallest seed of civilization?
I find myself asking an earlier question:
What kind of human must exist for that seed to grow?
Because a seed does not grow in every soil.
The same library placed in two societies may produce entirely different civilizations depending on the psychological maturity, values, and relational capacities of the people rebuilding it.
In other words, civilization is reproduced not only through information, but through the people who interpret, embody, and transmit it.
Knowledge tells us how to build.
Our psychological development shapes why we build, for whom, and toward what ends.
Perhaps that is the deeper recursive loop.
A civilization ultimately reproduces itself by cultivating the kinds of people who will, in turn, reproduce the civilization—as much through their way of being as through the information it preserves.
This article seems to treat civilization as fundamentally an information processing problem. But what if civilization has mission-critical parts that are more experiential, like trust, shared norms, systems of authority, rituals, institutions... A chemistry textbook doesn't create chemists. A constitution doesn't create constitutional government. A medical library doesn't create physicians. Those emerge from communities with practices that transmit more than explicit knowledge. So maybe, instead of asking "What is the smallest library that can reboot civilization?", a better question to ask is "What is the smallest self-sustaining social process that can regenerate civilization?"
I suspect you are correct. Just like an acorn can't make a tree without soil and environment, there is very likely social soil needed. What do you think the minimum would be?
Not sure, but at the very least it would have to reliably produce people who can trust one another and who share a sense of a future beyond themselves. You would need both. Trust alone can sustain a village or a business partnership. But building cathedrals, governments, and scientific traditions all require people to orient themselves toward something that outlives them.
Now call me old fashioned, but is not this whole concept somewhat flawed?
If it's reset then we cannot assume an environment as we have. As reset takes time, will there be anyone who can read? Who teaches this knowledge to others & why would the 'old gurad' be let anywhere near those who've managed to survive?
Personally, I'd be wanting find my own way in order not screw up again...
If you wish to make an apple pie from scratch...
....start scratching, right?
This reminds me of a project to design 50 open source tools that are needed to "build a small, sustainable civilization with modern comforts." I believe their effort has stalled, but it is an interesting goal.
https://www.opensourceecology.org/gvcs/
Kevin, I have to ask, though it blatantly exposes my idea lust, will you share that list of 200 books?
What language(s) would that Forever Book be written in?
You're not sharing the list?
I really appreciate the sentiment behind these ideas and indeed have a post-industrial SF novel about half written that includes a quasi-monastic order that builds and maintains a Great Library "to save what is worth saving". The novel is set 700 years from now on the NW coast of North America (Cascadia). In the world-building I explored many ideas for preserving knowledge AND a population that can make use of that knowledge. It turns out medieval monasteries already optimized post-industrial data storage: vellum, ink and a lot of copying! In my novel, the literate community that grows up around the library turns out to be just as critical as the saved knowledge; for what good is knowledge/information if no one can read or understand it?
The entire thought experiment has convinced me that the people of the future will just have to be clever enough to figure out whatever they need to know. We figured out what we needed - they will too. Or not. And to be quite frank, there is a nearly infinite volume of data existing today that is utterly useless outside of our fossil-energy/industrial/globalized context (and much that is utterly useless anytime!). So the books that must be copied onto vellum (and eventually recopied, ad infinitum) are mostly conceptual rather than technical and more in the field of the humanities than the sciences. I'm actually more intrigued by what copying these important books by hand might have on the scribes and the culture of the local community. One effect might be very interesting conversations at the dining hall tables!
I don't think our technology really exists outside of industrial civilisation. Wouldn't you be better off documenting in detail how to kickstart the industrial revolution, than to document how to make paper?
If you want a seed that doesn't require billions of humans and the industrial revolution etc, you probably need to really strongly reconceive every piece of the tech stack that you're going to include, to be on a more self contained path. Harder than it sounds I think! It'll be extremely tough to scrub out industrial and pre-AI tech assumptions. Like if you are trying to describe how to refine petroleum you might have wandered off the track? etc, lots of etc.
I think at this point LLMs and future AI are the key. An LLM is an incredible compression of all of our knowledge. Wait a bit, and I suspect we're going to get new computing technology that doesn't require multibillion dollar fans and the modern industrial ecosystem, but rather that you can put together (grow?) with solar, batteries, personal robots, all guided by local synthetic intelligence. If you can run LLMs on that, then there's your seed.
(multibillion dollar fabs, that should say)
Interesting reboot and expansion of the idea. Nice illustration, too. (Though, as an atheist, the buildings seem a bit too church-like for my taste.)
I hear the idea of a forever book in my head. I will hear it every time I open a good exploration with my AI tools.
I like the idea. It offers a reason for human optimism, despite the present turmoil. It makes civilization resilient.
But, Kevin Kelly, how do we keep the rest of civilization from killing the idea aborning?
If it wants to kill ideas, it's not civilization.