29 ms·

What I worry a lot more instead is how knowledge of manufacturing and engineering could be lost due to our greed. Typical scenario: Industry I is not doing fin

What I worry a lot more instead is how knowledge of manufacturing and engineering could be lost due to our greed. Typical scenario: Industry I is not doing fine in country C (i.e. the fund managers are not happy about lack of growth of the public companies in this sector) due to reasons R1, R2, ..., Rn. Then management decided to outsource and eventually dismantle the factories to "globalize" it. Knowledge retained by the older generation of engineers, technicians and workers were completely lost when they passed away.

ge96 2y ago
Didn't that already happen with US to China from the 70s
- tempodox 2y ago
  Those who cannot learn from history are doomed to repeat it.
  - makeitdouble 2y ago
    Shareholders in the 70s made a lot of money. They'll be OK to rinse and repeat that history.
    - tempodox 2y ago
      True. And the people selling shovels for the “AI” gold rush are making boatloads of money now. But it's still up to us whether we buy their customers' products.
    - __MatrixMan__ 2y ago
      That'll only work if somebody is willing to do the job for a lower price. The US may be in a different position the next time around.
      - trashtester 2y ago
        Somebody, or SOMETHING. There will not be much work that cannot be done by Figure, Optimus, Atlas, Claude, Grok or GPT by 2035.
        ozfive 2y ago
        We've heard this for many decades. I'm sure we will be hearing it for decades more...
        __MatrixMan__ 2y ago
        I think we'll reach "most work done by unattended robots" around 2070. But we'd be better off if started working on post-labor-scarcity economics ASAP--might as well start learning to swim before the ship sinks. It might even be fun.
        ozfive 2y ago
        Someone once told me be a cook, a hairdresser, or a repairman.
        gaze 2y ago
        It doesn't really matter if they can or can't do the work. If enough execs are convinced AI can replace the workers, they will replace the workers, and management will have moved onto other things before they suffer consequences.
- dudefeliciano 2y ago
  german automakers are experiencing this right now
  - aleph_minus_one 2y ago
    The problem with the German automotive industry is in my opinion different: they developed cars which are not different from the kind of cars that many people in their market want.
mschuster91 2y ago
> Knowledge retained by the older generation of engineers, technicians and workers were completely lost when they passed away. That has happened quite often in the past [1]. [1] https://en.wikipedia.org/wiki/List_of_lost_inventions https://en.wikipedia.org/wiki/List_of_lost_inventions
- aaronax 2y ago
  Only three real examples on that page. Petrification: no real loss to society but I will concede that this appears to be an example of actual loss of "progress". Very few people are saying "oh diety, if only I could preserve this animal with perfect color and texture (but discarding all other characteristics). Greek fire: the history of when this knowledge was lost is unclear. Also it was intentionally kept secret. Certainly we have functionally similar tech now. I'll give 50% credit. Panjagan: We don't even know if this was a weapon or a technique? Everything about this is incredibly vague. In summary, this has happened approximately 1.5 times in the past. Not "quite often".
  - mschuster91 2y ago
    There's also the still open question on how the pyramids of Egypt were built, and AFAIK NASA lost the knowledge how to build some stuff as well. And the recipe for Roman concrete was also long lost, with researchers being able to reverse-engineer what it likely was only a few years ago [1]. [1] https://news.mit.edu/2023/roman-concrete-durability-lime-casts-0106 https://news.mit.edu/2023/roman-concrete-durability-lime-cas...
  - __turbobrew__ 2y ago
    Fogbank was a famous case where we forgot how to make it: https://en.m.wikipedia.org/wiki/Fogbank https://en.m.wikipedia.org/wiki/Fogbank Similarly expertise in plutonium manufacturing has been lost, and now that countries have run out they are re-learning how they did it 40 years ago.
  - dghlsakjg 2y ago
    That list doesn't seem exhaustive in the least. A modern example is Fogbank, a classified nuclear weapons material that we recently had to reverse engineer to successfully rebuild certain warheads. https://en.wikipedia.org/wiki/Fogbank https://en.wikipedia.org/wiki/Fogbank Edit: I see someone beat me to the punch with Fogbank. For another example, look to the F1 rocket engine that took us to the moon. Despite having the actual engineering documents, we just don't have the manufacturing skill to rebuild one: https://apollo11space.com/why-cant-we-remake-the-rocketdyne-f1-engine/ https://apollo11space.com/why-cant-we-remake-the-rocketdyne-...
DAGdug 2y ago
Some about being pedantic, but what’s the value of I,C, R1-Rn here? Seems like a distraction!
kjkjadksj 2y ago
This is a huge deal in some industries even today. Such as film photography. No one makes 35mm cameras although it would be all too easy to rehash a good 90s point and shoot model and have it sell like hotcakes (see prices/hype on olympus mjuII). No one even makes or services lab scanners anymore; all those noritsus and fronteirs for pro lab scans are approaching 30 years old and nothing is even close on the market or could even catch up to what went on in terms of R and D building these machines for a seriously profitable industry at the time. Film has been getting increasingly popular. Local film labs are busy and new ones are opening. It is the vinyl record of imaging. Yet despite this, kodak and fujifilm have responded as most lack of forsight businesses do and cut film stocks, raised prices, constantly trying to squeeze more blood from the stone which goes on to kill growth in the industry vs supporting it and fostering actual growth and more profits more than 1 quarter out.
- whizzter 2y ago
  There is always retro enhusiasts of "obsolete" tech/methods, people knit despite there being cheap cotton clothes, people tinker with 50s cars despite everything. While there is tons of money to be made for niche-enthusiasts, these niche's aren't always large enough to properly re-industrialize (chemical regulations, expensive machinery,etc).
  - kjkjadksj 2y ago
    Retro enthusiast solution in this case is a 3d printed can you attach to your lens and its on you to revert and grade the color in lightroom. Better hope the camera is perfectly aligned to the negative that is perfectly flat and you have a lens that can focus a flat field free of distortions. It is a far cry to the engineering that went into these pro lab scanners. Just the color science alone people struggle to replicate, resolution and sharpness throughout the frame as well, speed of handling for lab setting is second to none. People still pay a premium for these scans on these particular machines. Kodachrome homebrew efforts don’t get close either. The process is too bespoke and reliant on instruments that no longer exist. Even proper c41 chemistry is down to just when the full kodak kit is in stock and not backordered by desperate developers: those instagram brands kit contains blix and are inferior as a result to a separate bleach and fix kit. Fuji press kit the old alternative is hard or impossible to find. And no one is making new cameras. Only fixing old ones and only popular models able to be fixed and worth the time creating a secondary market of parts for (so basically Leica and pay for that or be relatively SOL to varying degrees for most else). The biggest issue is this stuff was created when film was a global industry. tens of thousands of engineers were working on every step of the process for decades. That is gone now. All that old knowledge and learnings mostly gone because these companies kept poor records and destroyed old obsolete notes and material. Not just how the camera works but how to manufacture it and everything else at scale or at least in a way that it is actually profitable. And perhaps it could never become profitable without efficiencies brought on by scale and massive investment in manufacturing. And it would actually be forever be lost once all this equipment we still have falls apart.
al_borland 2y ago
I think about this often and it bothers me a lot. I think the same can happen for knowledge work. Country A ends up turning into a bunch of managers outsourcing to Country B, and then at some point Country B realizes they can manage themselves. Companies are quite literally training their future competition. Once it reaches a tipping point, I don’t think it’s easy to reverse.
- pca006132 2y ago
  I feel like top management and share holders don't really care about things that may happen ~30 years later. And even if they care, there are a lot more other issues that may ruin a company, e.g. the entire industry is replaced due to technological advancement.
  - ferguess_k 2y ago
    They definitely not, and there is no safe valve for it. Can't blame them though, this is what Capitalism teaches all of us. I'm short sighted too, but the thing is I won't be able to go wherever I want when SHTF while fund managers can probably do so.
  - al_borland 2y ago
    It's not necessarily the fault of the company. The government created the environment to make it the only way to stay competitive for a lot of companies. Some bucked the trend, and I respect them for it. There is a push to bring things back now, but it's going to be ugly. It's much better to avoid the issue in the first place like most countries, but it's too late for that.
- cultofmetatron 2y ago
  > Country A ends up turning into a bunch of managers outsourcing to Country B, why the need to be hypothetical? this is more or less how it played out between china and america. America outsourced all manufacturing to China while doing all the R&D and "innovation." one look at companies like DJI, BYD and Bambulab and its clear that china can innovate just fine on their own now. Their products are becoming objectively better than the US designed ones.
jayd16 2y ago
What is the scenario exactly? The knowledge is proprietary and thrown away? And also the worker never wrote or used the innovation elsewhere despite it not living in a patent?
- amiga-workbench 2y ago
  There is such a thing as institutional knowledge, you build it up by retaining staff properly, training replacements and handing this knowledge down, you need continuity. Another important part of this is doing a thing often. For example, if you halt manufacturing of a widget and then want to restart manufacturing it two decades later, you are going to run into every single snag and problem found during the first iteration of the process, you're starting from scratch and you might not even be able to actually replicate the original process properly. The effect even applies on a larger national scale, where if a country stops investing in infrastructure projects for a long period, they will find themselves incapable of executing these projects properly in the future.

ge96 2y ago

Didn't that already happen with US to China from the 70s

tempodox 2y ago
Those who cannot learn from history are doomed to repeat it.
- makeitdouble 2y ago
  Shareholders in the 70s made a lot of money. They'll be OK to rinse and repeat that history.
  - tempodox 2y ago
    True. And the people selling shovels for the “AI” gold rush are making boatloads of money now. But it's still up to us whether we buy their customers' products.
  - __MatrixMan__ 2y ago
    That'll only work if somebody is willing to do the job for a lower price. The US may be in a different position the next time around.
    - trashtester 2y ago
      Somebody, or SOMETHING. There will not be much work that cannot be done by Figure, Optimus, Atlas, Claude, Grok or GPT by 2035.
      - ozfive 2y ago
        We've heard this for many decades. I'm sure we will be hearing it for decades more...
        __MatrixMan__ 2y ago
        I think we'll reach "most work done by unattended robots" around 2070. But we'd be better off if started working on post-labor-scarcity economics ASAP--might as well start learning to swim before the ship sinks. It might even be fun.
        ozfive 2y ago
        Someone once told me be a cook, a hairdresser, or a repairman.
        gaze 2y ago
        It doesn't really matter if they can or can't do the work. If enough execs are convinced AI can replace the workers, they will replace the workers, and management will have moved onto other things before they suffer consequences.
dudefeliciano 2y ago
german automakers are experiencing this right now
- aleph_minus_one 2y ago
  The problem with the German automotive industry is in my opinion different: they developed cars which are not different from the kind of cars that many people in their market want.

mschuster91 2y ago

> Knowledge retained by the older generation of engineers, technicians and workers were completely lost when they passed away. That has happened quite often in the past [1]. [1] https://en.wikipedia.org/wiki/List_of_lost_inventions https://en.wikipedia.org/wiki/List_of_lost_inventions

aaronax 2y ago
Only three real examples on that page. Petrification: no real loss to society but I will concede that this appears to be an example of actual loss of "progress". Very few people are saying "oh diety, if only I could preserve this animal with perfect color and texture (but discarding all other characteristics). Greek fire: the history of when this knowledge was lost is unclear. Also it was intentionally kept secret. Certainly we have functionally similar tech now. I'll give 50% credit. Panjagan: We don't even know if this was a weapon or a technique? Everything about this is incredibly vague. In summary, this has happened approximately 1.5 times in the past. Not "quite often".
- mschuster91 2y ago
  There's also the still open question on how the pyramids of Egypt were built, and AFAIK NASA lost the knowledge how to build some stuff as well. And the recipe for Roman concrete was also long lost, with researchers being able to reverse-engineer what it likely was only a few years ago [1]. [1] https://news.mit.edu/2023/roman-concrete-durability-lime-casts-0106 https://news.mit.edu/2023/roman-concrete-durability-lime-cas...
- __turbobrew__ 2y ago
  Fogbank was a famous case where we forgot how to make it: https://en.m.wikipedia.org/wiki/Fogbank https://en.m.wikipedia.org/wiki/Fogbank Similarly expertise in plutonium manufacturing has been lost, and now that countries have run out they are re-learning how they did it 40 years ago.
- dghlsakjg 2y ago
  That list doesn't seem exhaustive in the least. A modern example is Fogbank, a classified nuclear weapons material that we recently had to reverse engineer to successfully rebuild certain warheads. https://en.wikipedia.org/wiki/Fogbank https://en.wikipedia.org/wiki/Fogbank Edit: I see someone beat me to the punch with Fogbank. For another example, look to the F1 rocket engine that took us to the moon. Despite having the actual engineering documents, we just don't have the manufacturing skill to rebuild one: https://apollo11space.com/why-cant-we-remake-the-rocketdyne-f1-engine/ https://apollo11space.com/why-cant-we-remake-the-rocketdyne-...

DAGdug 2y ago

Some about being pedantic, but what’s the value of I,C, R1-Rn here? Seems like a distraction!

kjkjadksj 2y ago

This is a huge deal in some industries even today. Such as film photography. No one makes 35mm cameras although it would be all too easy to rehash a good 90s point and shoot model and have it sell like hotcakes (see prices/hype on olympus mjuII). No one even makes or services lab scanners anymore; all those noritsus and fronteirs for pro lab scans are approaching 30 years old and nothing is even close on the market or could even catch up to what went on in terms of R and D building these machines for a seriously profitable industry at the time. Film has been getting increasingly popular. Local film labs are busy and new ones are opening. It is the vinyl record of imaging. Yet despite this, kodak and fujifilm have responded as most lack of forsight businesses do and cut film stocks, raised prices, constantly trying to squeeze more blood from the stone which goes on to kill growth in the industry vs supporting it and fostering actual growth and more profits more than 1 quarter out.

whizzter 2y ago
There is always retro enhusiasts of "obsolete" tech/methods, people knit despite there being cheap cotton clothes, people tinker with 50s cars despite everything. While there is tons of money to be made for niche-enthusiasts, these niche's aren't always large enough to properly re-industrialize (chemical regulations, expensive machinery,etc).
- kjkjadksj 2y ago
  Retro enthusiast solution in this case is a 3d printed can you attach to your lens and its on you to revert and grade the color in lightroom. Better hope the camera is perfectly aligned to the negative that is perfectly flat and you have a lens that can focus a flat field free of distortions. It is a far cry to the engineering that went into these pro lab scanners. Just the color science alone people struggle to replicate, resolution and sharpness throughout the frame as well, speed of handling for lab setting is second to none. People still pay a premium for these scans on these particular machines. Kodachrome homebrew efforts don’t get close either. The process is too bespoke and reliant on instruments that no longer exist. Even proper c41 chemistry is down to just when the full kodak kit is in stock and not backordered by desperate developers: those instagram brands kit contains blix and are inferior as a result to a separate bleach and fix kit. Fuji press kit the old alternative is hard or impossible to find. And no one is making new cameras. Only fixing old ones and only popular models able to be fixed and worth the time creating a secondary market of parts for (so basically Leica and pay for that or be relatively SOL to varying degrees for most else). The biggest issue is this stuff was created when film was a global industry. tens of thousands of engineers were working on every step of the process for decades. That is gone now. All that old knowledge and learnings mostly gone because these companies kept poor records and destroyed old obsolete notes and material. Not just how the camera works but how to manufacture it and everything else at scale or at least in a way that it is actually profitable. And perhaps it could never become profitable without efficiencies brought on by scale and massive investment in manufacturing. And it would actually be forever be lost once all this equipment we still have falls apart.

al_borland 2y ago

I think about this often and it bothers me a lot. I think the same can happen for knowledge work. Country A ends up turning into a bunch of managers outsourcing to Country B, and then at some point Country B realizes they can manage themselves. Companies are quite literally training their future competition. Once it reaches a tipping point, I don’t think it’s easy to reverse.

pca006132 2y ago
I feel like top management and share holders don't really care about things that may happen ~30 years later. And even if they care, there are a lot more other issues that may ruin a company, e.g. the entire industry is replaced due to technological advancement.
- ferguess_k 2y ago
  They definitely not, and there is no safe valve for it. Can't blame them though, this is what Capitalism teaches all of us. I'm short sighted too, but the thing is I won't be able to go wherever I want when SHTF while fund managers can probably do so.
- al_borland 2y ago
  It's not necessarily the fault of the company. The government created the environment to make it the only way to stay competitive for a lot of companies. Some bucked the trend, and I respect them for it. There is a push to bring things back now, but it's going to be ugly. It's much better to avoid the issue in the first place like most countries, but it's too late for that.
cultofmetatron 2y ago
> Country A ends up turning into a bunch of managers outsourcing to Country B, why the need to be hypothetical? this is more or less how it played out between china and america. America outsourced all manufacturing to China while doing all the R&D and "innovation." one look at companies like DJI, BYD and Bambulab and its clear that china can innovate just fine on their own now. Their products are becoming objectively better than the US designed ones.

jayd16 2y ago

What is the scenario exactly? The knowledge is proprietary and thrown away? And also the worker never wrote or used the innovation elsewhere despite it not living in a patent?

amiga-workbench 2y ago
There is such a thing as institutional knowledge, you build it up by retaining staff properly, training replacements and handing this knowledge down, you need continuity. Another important part of this is doing a thing often. For example, if you halt manufacturing of a widget and then want to restart manufacturing it two decades later, you are going to run into every single snag and problem found during the first iteration of the process, you're starting from scratch and you might not even be able to actually replicate the original process properly. The effect even applies on a larger national scale, where if a country stops investing in infrastructure projects for a long period, they will find themselves incapable of executing these projects properly in the future.