That longbow story seems similar to one I had heard about the martial art of two-handed sword combat being largely lost. The argument was essentially that a late medieval European knight with two-handed sword and pike was actually a more effective soldier than the Japanese samurai with daisho and bow, but the latter had been better remembered and afforded better retroactive status, since their military tradition has been preserved via traditions of pseudo-religious martial arts societies.
In Europe, soldiering moved on to muskets, and fencing became an aristocratic pastime, preserving only epee and foil--weapons that could still be used effectively by people that were not professional soldiers. The techniques of two-handed sword fighting were recorded on a handful of pages, which are now the sole historical basis for academic reconstructions of typical combats. The remainder was reconstructed by history academics and medieval combat geeks.
Nobody thought it was important to preserve the state of the art of two-handed sword combat once you could stand back and punch holes in a big, armored man with a thicket of pikes, or a swarm of bullets, darts, arrows, or bolts. War evolved, and the formerly effective techniques were forgotten. The last surviving military blades were variations on the cavalry saber and naval cutlass, and were rarely used outside of ceremonial occasions. They just don't work very well against a guy with a machine gun. But the (dueling) saber was shoehorned back into the sport of fencing as its third blade, and other sabers have been fixed in various military traditions, being the last swords considered to be militarily useful.
That same disregard for historical preservation seems very common, especially considering technological obsolescence.
But in that case, do we really need it? That knowledge is mostly dead-end knowledge, useful only at the apex of a local maximum that has since been bypassed. The 70% that the master machinists believe should leave the field are the perfect candidates to improve CNC machines and 3D printers. Certainly, they would be useless if someone bombed all the chip fabs and murdered the lithographic mask designers, but then so would the bomb-droppers, who also rely on electronics packages to hit their targets.
It might actually be more effective to record all the things that became obsolete, and only keep details if they can be used as a stepping stone to the next thing. Then you only need to build vacuum tubes until you can build discrete transistors again, and you only need to build discrete transistors until recovering integrated circuits. In a civilization-preserving bunker or space habitat, nobody needs the expert-level knowledge on those stepping stones. The basics will do, until next month when the 3000 nm 8088-like chip fab comes online, and then that can leapfrog to an 800 nm MIPS chip fab, and those are only needed to control the machines making the 100 nm chips, and those run the machines that manufacture the universal hardware controller systems-on-a-chip that everyone needs to run the water purifiers, aeroponics tube farms, elevators, smart power inverters, and everything else. You don't need to remember the ox-drawn plow if you can jump straight to a small tractor, and you don't need to remember the tractor if you already know how to farm with PVC pipe, PEX plumbing, and a chem lab, instead of dirt.
Though it took civilization thousands of years to get where we are now, rebuilding from stone age to modern age could be done in a century or less, provided that someone is still around to remember the critical path to it. The scary part is that some of those critical sections may only be remembered by a few people, and they might be concentrated enough to all be killed by the same bomb. In that case, you're back to waiting through those thousands of years again.
In Europe, soldiering moved on to muskets, and fencing became an aristocratic pastime, preserving only epee and foil--weapons that could still be used effectively by people that were not professional soldiers. The techniques of two-handed sword fighting were recorded on a handful of pages, which are now the sole historical basis for academic reconstructions of typical combats. The remainder was reconstructed by history academics and medieval combat geeks.
Nobody thought it was important to preserve the state of the art of two-handed sword combat once you could stand back and punch holes in a big, armored man with a thicket of pikes, or a swarm of bullets, darts, arrows, or bolts. War evolved, and the formerly effective techniques were forgotten. The last surviving military blades were variations on the cavalry saber and naval cutlass, and were rarely used outside of ceremonial occasions. They just don't work very well against a guy with a machine gun. But the (dueling) saber was shoehorned back into the sport of fencing as its third blade, and other sabers have been fixed in various military traditions, being the last swords considered to be militarily useful.
That same disregard for historical preservation seems very common, especially considering technological obsolescence.
But in that case, do we really need it? That knowledge is mostly dead-end knowledge, useful only at the apex of a local maximum that has since been bypassed. The 70% that the master machinists believe should leave the field are the perfect candidates to improve CNC machines and 3D printers. Certainly, they would be useless if someone bombed all the chip fabs and murdered the lithographic mask designers, but then so would the bomb-droppers, who also rely on electronics packages to hit their targets.
It might actually be more effective to record all the things that became obsolete, and only keep details if they can be used as a stepping stone to the next thing. Then you only need to build vacuum tubes until you can build discrete transistors again, and you only need to build discrete transistors until recovering integrated circuits. In a civilization-preserving bunker or space habitat, nobody needs the expert-level knowledge on those stepping stones. The basics will do, until next month when the 3000 nm 8088-like chip fab comes online, and then that can leapfrog to an 800 nm MIPS chip fab, and those are only needed to control the machines making the 100 nm chips, and those run the machines that manufacture the universal hardware controller systems-on-a-chip that everyone needs to run the water purifiers, aeroponics tube farms, elevators, smart power inverters, and everything else. You don't need to remember the ox-drawn plow if you can jump straight to a small tractor, and you don't need to remember the tractor if you already know how to farm with PVC pipe, PEX plumbing, and a chem lab, instead of dirt.
Though it took civilization thousands of years to get where we are now, rebuilding from stone age to modern age could be done in a century or less, provided that someone is still around to remember the critical path to it. The scary part is that some of those critical sections may only be remembered by a few people, and they might be concentrated enough to all be killed by the same bomb. In that case, you're back to waiting through those thousands of years again.