3 ms·

Agreed. If there were something catastrophic that were to happen, there would be a whole lot of levels between the skills needed to camp out and the skills nee

Agreed. If there were something catastrophic that were to happen, there would be a whole lot of levels between the skills needed to camp out and the skills needed to stand up a large server farm or wafer fab that would be challenging to traverse. Some of the intermediate levels are still in living memory or are well-documented, but some of them would need to be reverse engineered through trial and error and may not even be apparent at this point.

vidarh 13y ago
A lot of it would be tremendously difficult, but at the same time even piecemeal knowledge and resourcefulness would place people at a massive advantage when "re-inventing" a lot of the things we already know is possible. I lack the knowledge to build a high performance modern internal combustion engine, for example. But the idea is enough to start someone on figuring out the basics much faster than the initial invention even if all other knowledge of it was lost. And in the meantime, a steam engine is "trivial" and easy to fuel. Heck, you'd be able to find even school children with a sufficient if basic understanding of old iron making processes to kickstart metal-making processes if need be. And understanding even something as basic as how a bicycle dynamo works allows you to bootstrap an electricity supply. I wouldn't be able to build an advanced chip fab, but I vaguely remember enough beginning electronics that I'd know roughly what to experiment with to figure out how to get to working transistors, diodes and resistors. And I know enough to start looking at the right things to be able to etch basic circuit boards. And just knowing the idea of photo-litography and scaling down a mask gets you far towards figuring out how to create basic circuits. It'd still take a lot of work and a long time, but nowhere near as long as it took to get there initially. And so on. Even if we did not just lose all the technology, but also "lost" all the experts who could otherwise fill in vastly more details than people with the pop-sci version of the knowledge of the fields, there are so many areas like this where understanding the basics and knowing the principles if not the implementation of what has been proven to work can massively short-circuit the process. We would certainly face massive setbacks, but there is also another issue here: The resources required. Even if we had the knowledge to continue to make high end CPUs, if we lose the fabs, we don't want to go straight back to trying to build a high end chip fab for the simple reason that it would take an immense amount of other technology and resources, while we could stand up the pre-requisites for producing simple, 1970's level boards, components and even chips in a tiny fraction of the time and with far more manageable resources and already start to reap a massive amount of benefits, including tools to help substantially speed up the effort of bringing up the next iteration.
- VLM 13y ago
  You've also got formally proved knowledge vs gut level guess. There exists formally proved knowledge that pretty much the ideal liquid coolant for your internal combustion engine is 50:50 ethylene glycol and water. If all that formal knowledge is lost it'll make future progress difficult, but at least gut knowledge that it works would save a lot of time bootstrapping. This is the origin during warfare of stealing the oppositions machinery and reverse engineering it. If it was faster and easier to work from first principles, the winners of wars would tend to do that instead of reverse engineering.

vidarh 13y ago

A lot of it would be tremendously difficult, but at the same time even piecemeal knowledge and resourcefulness would place people at a massive advantage when "re-inventing" a lot of the things we already know is possible. I lack the knowledge to build a high performance modern internal combustion engine, for example. But the idea is enough to start someone on figuring out the basics much faster than the initial invention even if all other knowledge of it was lost. And in the meantime, a steam engine is "trivial" and easy to fuel. Heck, you'd be able to find even school children with a sufficient if basic understanding of old iron making processes to kickstart metal-making processes if need be. And understanding even something as basic as how a bicycle dynamo works allows you to bootstrap an electricity supply. I wouldn't be able to build an advanced chip fab, but I vaguely remember enough beginning electronics that I'd know roughly what to experiment with to figure out how to get to working transistors, diodes and resistors. And I know enough to start looking at the right things to be able to etch basic circuit boards. And just knowing the idea of photo-litography and scaling down a mask gets you far towards figuring out how to create basic circuits. It'd still take a lot of work and a long time, but nowhere near as long as it took to get there initially. And so on. Even if we did not just lose all the technology, but also "lost" all the experts who could otherwise fill in vastly more details than people with the pop-sci version of the knowledge of the fields, there are so many areas like this where understanding the basics and knowing the principles if not the implementation of what has been proven to work can massively short-circuit the process. We would certainly face massive setbacks, but there is also another issue here: The resources required. Even if we had the knowledge to continue to make high end CPUs, if we lose the fabs, we don't want to go straight back to trying to build a high end chip fab for the simple reason that it would take an immense amount of other technology and resources, while we could stand up the pre-requisites for producing simple, 1970's level boards, components and even chips in a tiny fraction of the time and with far more manageable resources and already start to reap a massive amount of benefits, including tools to help substantially speed up the effort of bringing up the next iteration.

VLM 13y ago
You've also got formally proved knowledge vs gut level guess. There exists formally proved knowledge that pretty much the ideal liquid coolant for your internal combustion engine is 50:50 ethylene glycol and water. If all that formal knowledge is lost it'll make future progress difficult, but at least gut knowledge that it works would save a lot of time bootstrapping. This is the origin during warfare of stealing the oppositions machinery and reverse engineering it. If it was faster and easier to work from first principles, the winners of wars would tend to do that instead of reverse engineering.