5 ms·

> ...and then the population realizes they can't build those anymore because the know-how is dispersed, and many key people aren't around anymore. Yes, this is

> ...and then the population realizes they can't build those anymore because the know-how is dispersed, and many key people aren't around anymore. Yes, this is the "civilization has so far been binary-only, not open source" problem I observed when I started looking into reproducing civilization on a space habitat, for an eventual transition into a generation ship. It turns out a heck of a lot of civilizational artifacts are "write many times, read once" across generations: lots of iterative tinkering, used until the next "git commit"-like widespread adoption, then quickly receding from memory. It was once known that longbowmen could shoot arrows curving around an obstacle. Scantly recorded, then the actual skill was forgotten until an archer geek who knew about the obscure written references worked out how to do it again from scratch. So we have that skill in our civilization now, yay. Except: he hasn't published his experiences, fleshed out into lessons for newcomers, etc. If he passes away before doing that, then we're likely to lose that skill again. Piles and piles of human experiential information like this is simply not efficiently recorded, much less transmitted even in our "advanced" civilization today. I'm keen on learning machining and tool making. The more I look into it, the more I realize what is written down barely scratches the surface. Lots of master machinists talk about a feel for doing the job right the first time such that only about 10% of machinists are really good, 20% can follow explicit directions, and the rest should find another line of work. This is I believe the foundation of why civilizations are so brittle and fragile, and so damnably difficult and time-consuming to build back up aspects that are lost. What worries me these days: the elites do not grasp this; if they did, they'd be a fuck ton more scared of war than they are. They'd also be way less sanguine about globalization and losing vast swathes of this binary-only technological know-how. The Shenzhen ecosystem is starting to show that even the "we'll out-innovate them, and they'll 'just' always manufacture our ideas" canard is wearing mighty thin over time. If this feels overwhelming, then it gets worse: just tackle a scaled-down problem space, "prepping" or "self sufficiency", and you quickly realize that at an individual level, the challenges are magnified. You have to wind back the technological level (and quality of life) quite a few centuries before you can reasonably do everything yourself...nearly.

Retric 9y ago
Because this stuff is discoverable it's less important than you might think. When you have 10's of millions of archers doing that as a job then curving an arrow is not going to stay secret for that long. The really major discoveries of our history are things like a positional number system aka zero. That's the kind of thing that's likely to survive because it's both really useful, easy to teach, and really common. Germ theory might not stick around, but it's really widespread and also both useful and the kind of thing we can pick up again. The jump from an 8080 including an understanding of Quantum Mechanics getting to a modern multi billion transistor i7 is 100's of billions in research. But, going from an 8080's took us less than 100 years and we could do it again assuming a large relatively stable global economy.
- yourapostasy 9y ago
  I thought the same, until I looked back at when experiential knowledge was lost and then reacquired. If it is within a generation or two of loss and recovery, the process seems to proceed along the lines of what you mention. Beyond that, and recovery success rates looks like a cliff function. Metallurgy has been independently worked upon around the world, and specific aspects lost many times over. Someone demonstrated they were on the right path to working out precision hand-built gearing with the Antikythera mechanism. The discovered mechanism wasn't very accurate, but we lost that tech for about 13 centuries. Another way of considering that lost tech: it took about three centuries to get the first mechanical computers from the 14th century onwards (about when we recovered the level of the Antikythera mechanism's tech); we could potentially have had software on mechanical if not electronic computers say, 7 centuries ago, if the Antikythera tech was not lost. Fast, bi-directional mass communications seems to be key to discoverability. Lose that, and all sorts of losses apparently accumulate quickly. Of more immediate concern to me is the economic effects of classifying certain economic verticals as "low margin" and then deprecating them, as if some kind of central planning committee decreed that the industry shall forevermore never cross-pollinate or value-add any other industries. I'm not convinced it is that straightforward nor predictable. Somewhere in China at this moment, some college kid studying materials science has parents who work at a "low margin" garment factory, who through osmosis over the decades taught her about the ins and outs of industrial-scale, high-precision thread-manufacturing and tight weaving, and she's noodling about adapting that tech to manufacture say, glass foam insulation at 1/1000th the current cost, and revolutionize the global insulation industry forever. It certainly is economically efficient in our local maxima to arrange our "developed-world's" industrial policy in this manner, but I think the jury is still out on whether or not it is sustainable; final arbitration will be history in a hundred years, but my bet is on China, who seems to have no qualms absorbing the industries "developed nations" no longer want, and doesn't seem as eager to divest as much or as fast to "lower tier" economies. > ...and we could do it again assuming a large relatively stable global economy. Here is where I have concerns "there be dragons". I've yet to read of many losses of industries (much less civilizational collapse) where there was sufficient stability to quickly recover; usually the recovery process, even when successful, was arduous and time-consuming. As rapidly as the global economy changes now, and with the global economy's new normal turbulence, I'm not convinced that recovery is any easier today than in earlier eras. I am thankful for people who publish information like "From NAND to Tetris".

Retric 9y ago

Because this stuff is discoverable it's less important than you might think. When you have 10's of millions of archers doing that as a job then curving an arrow is not going to stay secret for that long. The really major discoveries of our history are things like a positional number system aka zero. That's the kind of thing that's likely to survive because it's both really useful, easy to teach, and really common. Germ theory might not stick around, but it's really widespread and also both useful and the kind of thing we can pick up again. The jump from an 8080 including an understanding of Quantum Mechanics getting to a modern multi billion transistor i7 is 100's of billions in research. But, going from an 8080's took us less than 100 years and we could do it again assuming a large relatively stable global economy.

yourapostasy 9y ago
I thought the same, until I looked back at when experiential knowledge was lost and then reacquired. If it is within a generation or two of loss and recovery, the process seems to proceed along the lines of what you mention. Beyond that, and recovery success rates looks like a cliff function. Metallurgy has been independently worked upon around the world, and specific aspects lost many times over. Someone demonstrated they were on the right path to working out precision hand-built gearing with the Antikythera mechanism. The discovered mechanism wasn't very accurate, but we lost that tech for about 13 centuries. Another way of considering that lost tech: it took about three centuries to get the first mechanical computers from the 14th century onwards (about when we recovered the level of the Antikythera mechanism's tech); we could potentially have had software on mechanical if not electronic computers say, 7 centuries ago, if the Antikythera tech was not lost. Fast, bi-directional mass communications seems to be key to discoverability. Lose that, and all sorts of losses apparently accumulate quickly. Of more immediate concern to me is the economic effects of classifying certain economic verticals as "low margin" and then deprecating them, as if some kind of central planning committee decreed that the industry shall forevermore never cross-pollinate or value-add any other industries. I'm not convinced it is that straightforward nor predictable. Somewhere in China at this moment, some college kid studying materials science has parents who work at a "low margin" garment factory, who through osmosis over the decades taught her about the ins and outs of industrial-scale, high-precision thread-manufacturing and tight weaving, and she's noodling about adapting that tech to manufacture say, glass foam insulation at 1/1000th the current cost, and revolutionize the global insulation industry forever. It certainly is economically efficient in our local maxima to arrange our "developed-world's" industrial policy in this manner, but I think the jury is still out on whether or not it is sustainable; final arbitration will be history in a hundred years, but my bet is on China, who seems to have no qualms absorbing the industries "developed nations" no longer want, and doesn't seem as eager to divest as much or as fast to "lower tier" economies. > ...and we could do it again assuming a large relatively stable global economy. Here is where I have concerns "there be dragons". I've yet to read of many losses of industries (much less civilizational collapse) where there was sufficient stability to quickly recover; usually the recovery process, even when successful, was arduous and time-consuming. As rapidly as the global economy changes now, and with the global economy's new normal turbulence, I'm not convinced that recovery is any easier today than in earlier eras. I am thankful for people who publish information like "From NAND to Tetris".