We Rarely Lose Technology (2023)
hopefulmons.com
hopefulmons.com
For example, if you make something like a magnetic field sensor, your customers are a handful of university labs, and your supplier of some niche ferromagnetic material stops their production line because they sold primarily to the spinning-rust hard drive industry and that market no longer pays the bills, your product becomes impossible to make. Your use case will never supply the quantities needed to run the crucibles.
For a few decades it's perhaps theoretically possible to reproduce if, somehow the survival of the world depended on it and the old material production line can be restarted, but after that, people with the knowledge have all passed away and any documentation is scattered and incomplete.
Is there a particular reason you think it's irrelevant to have documentation on how to build something that might become uneconomical to build to the point of losing the technology? Or are you just being an asshole?
Written documentation almost never captures the complete details needed to produce a complex technology, and outside of extremely vertically-integrated industries, such documentation would rarely exist in one place anyway. You're lucky enough to find detailed documentation from the source company that makes the sensor/widget, but even so a document that opens with "step 1: purchase alloy #755-D from Westchester Special Metals, hydrogen annealed" is not much help if Westchester Special Metals went out of business 60 years ago, and had the usual number of trade secrets.
There are many different types of ultra-accurate magnetometer. They are made in many different ways. If one relies on some byproduct of some other industry that goes under and becomes nonviable, then the many competitors will fill the niche with a different type of magnetometer that will be just as good.
Even if we consider a different hypothetical industry that truly is single point of failure, well, there are academic papers explaining how these things work and how to make them. Maybe the specific model of metal is no longer made, but the instructions there and an intelligent, motivated person will figure it out. And then they'll figure out how to improve it.
I don't think the failure mechanism you describe will in practice lead to anything more than brief gaps in the availability of a technology. We can remake things to spec, like FOGBANK.
A result of this is that, in some rare cases, there are certain machines and instruments where the best ones in existence were manufactured in the '80s and are jealousy hoarded in temperature-controlled vaults, because nobody today can make a better one and the industry that produced them is completely gone.
FOGBANK is a great example of a lost technology that had to be rediscovered at great expense, and that only happened because the military could write a blank cheque. But that's exactly a lost technology; they had to do the R&D all over again. And even then, they didn't have to make up for a bigger market that vanished; they were always the only customer for FOGBANK, so the economics hadn't changed. Imagine if the economically producing that material required a customer base the size of Kodak's peak market share to amortize the capital cost of a production run.
Most technologies don't have infinite money available to recreate from scratch; they depend on economies of scale and die out with their markets.
I also think that some parts are really weakly-reasoned:
> The Ancient Egyptians cut stone with an impressive level of precision. The Incas in South America did too. So much so that people sometimes claim that the Egyptians and the Incas used some kind of now-lost technology. But they most likely didn’t: they were just really good at cutting stone.
Yeah, but we're not that good at cutting stone anymore. So what gives? The explanation here is very lacking. They either had a technology that let them cut stone so well, or some special know-how (itself a form of technology, in my opinion) that enabled cutting stone so precisely.
It's the ability to precisely measure and mark huge pieces of rock what looks like a miracle to me. Producing the exact shape is easy when you can always check how close to the desired shape you are, and it's the hard part.
Current methods of cutting pretty much anything including stone are absurdly more precise than what Incas and Egyptians had. We can cut stuff like diamond lenses down to 10-100nm roughness.
https://youtu.be/uR-hY7hUsaY?t=79
I have nothing but respect for the skill and professionalism of the ancients, but I find it extremely distasteful when someone tries to express this by putting down their modern counterparts.
We also don't build carriages as well or have an army of craftsman doing it, but it's lost/regressed because there's no economic incentives.
There are a few that know how to do it by hand, but mass production has evolved.
Here's a rabbit hole if this sounds like fun to you - https://natrontheory.com/index.html
The focus and color in those graphics makes them look like AI-generated graphics to me.
Normally you purify the quartz first, but here we're discussing what molten washing soda would or wouldn't do to granite. The quartz in granite normally forms a continuous phase, so as long as the quartz remains solid, the granite will remain solid.
Granite melts at a lower temperature than pure quartz, and I'm not entirely sure it wouldn't just melt before the washing soda had an effect.
I think the question is also really hard to answer in a non-answerable way because if you don't know you've forgotten who do something you simply just don't know. If you sub-divide humanity in to factions certain factions have lost the economic ability of doing some things for example have been lost to America. While humanity can still do it certain places would be totally lost without the existing tribal knowledge. It would take us a long time to reinvigorate a tool and die industry in America its not a lost thing but if you can't do it commercial currently can you do it at all?
Hopefully that's nota weird direction to take the discussion in
For example, Roman concrete, that strengthens in sea water: https://news.mit.edu/2023/roman-concrete-durability-lime-cas...
No technology was really lost, we just can't agree on how they stacked the stones into pyramids. The stonecutting technology they used is very well understood and we even have real actual tools used by the actual ancient Egyptian stonecutters and some of the tools used to manipulate and place stones.
There's also several incredibly solid theories on pyramid construction, but to verify them would require investigating the actual pyramids, which Hawass will not allow.
https://www.youtube.com/watch?v=mysYT260dqU&list=PLuROoe7EZ3...
We are pretty good at cutting stone, but not very good at cutting stone with the resources available to the ancient egyptians. So in some sense the technology has been lost: but that's more because there aren't really people try to perfect that craft, as opposed to because of some magical trick or technique that we don't know anything about (Though it's murky exactly what approaches they used, that's because there's multiple plausible ones and it's hard to pin them down, not because there's no plausible ones we can think of).
The first was Airadar. (Not "AI", "Air".)[1] I wrote about this on HN in 2016. This was a small phased-array radar for light aircraft, developed in 1973. It was suppressed by a patent secrecy order, because it was better than what the USAF had at the time. The inventor was a really good RF designer. Phased array radars existed back then, but they were huge ground-based installations. Mini phased array radars are available now, but it took decades for them to be available for light aircraft.[2]
The second was the electronic fluorescent lamp ballast. This was a replacement for those bulky magnetic ballasts found inside fluorescent light fixtures. The inventor licensed it to MagneTek, the biggest maker for magnetic ballasts, which didn't make it and didn't pay any royalties. So the inventor went to Townsend, Townsend, and Crew, the IP law firm in Palo Alto, and, after much litigation, came out with a hundred million or so. The law firm put this in their reception room brag book. Today, electronic fluorescent lamp ballasts are a commodity.
The problem with FOGBANK, the aerogel used in fusion weapons, turned out to be that the original process only worked because of some impurity in the raw materials. Attempts to replicate the process used a source for a raw material which was now better purified, and the process failed. It required tens of millions of dollars and a special appropriation to figure out the underlying problem. There was a period of over a decade during which the US could not make new H-bombs.
[1] https://books.google.com/books?id=NWzlTqj0gQ4C&pg=PA64#v=one...
It also seems it wasn’t fully understood initially.
We knew what it was and roughly how it worked, but it was very much a process of re-inventing from scratch.
Consider fire. Or fluorescent lamps. Or radio. There's a trick, we don't really understand what's going on, yet there is clearly something here worth more investigation and gradually we get a better understanding.
A give away is often that there's perceived to be an "art" to making it work, and then once we understand properly the art dissolves, replaced by science. Making fudge without modern technology involves hard to learn judgement to get the right temperatures, but, today you can buy an accurate digital probe and use the recipe and that'll work reliably because it's just sugar chemistry, no art needed.
Cost does factor into it. They're never going to make more of those last gasp Walkmans, where the device is barely bigger than the tape and it has excellent sound quality and long battery life. Not because it's somehow lost technology but because it doesn't make economic sense any more. It's worse along so many dimensions at once than what we have today, and yet it'd be incredibly expensive, so it won't happen.
More significantly, steel. Making steel was hit-and-miss until the capability to analyze ores was developed. Even then, it took about 10,000 tries to debug the Bessemer process. Steel has been made in small quantities for thousands of years, but quality steel only dates from the 1880s. The first steel I-beams were made in the late 1880s.
(Despite what Microsoft Copilot says, steel I-beams were not made in 1849. Those were wrought-iron I-beams.)
I agree with “let’s hope it remains that way.” But you may be underestimating the fragility of modern human civilization and all the systems that are so effective at preserving knowledge. At the scale of history, most of our technology has only existed for the blink of an eye. It would not take much, really, for all this to go away.
Absent the internet, power grid, and other systems of global infrastructure, absent anyone to take care of the libraries (which are falling out of favor), absent an authority to gather together all the experts with their tightly focused specializations, absent the chip fabs and the global supply network, how exactly do you build a computer? Or even something as “simple” as a pencil?
You develop that first. If there is a nuclear war and somehow I survive with 5000 people near me survive (if we are scattered around the world that doesn't help) that is enough for genetic diversity and we can start rebuilding society. There are plenty of books in libraries. I can figure out how to make a primitive printing press (likely from clay - fragile but just the ability to print say 10 books from a plate is huge and seems possible). Books in libraries will last for a while, and we can figure out better ways to store them. In many be 1000 years, but what of physics and chemistry we are able to preserve will help whoever. They will develop different supply chains. They will start knowing a spinning wheel is possible (I don't know enough about the spinning jenny to automate it) - I know where I live warm clothing is critical and so I will be sure to take the time to develop a spinning wheel thus allowing women to not have to spend 12 hours a day with a drop spindle. That is time they can then spending on investing something else while the men farm. (there is good reason for this sexual division of labor so I'd be forced to bring it back in the early days)
You might want to check that. There are definitely a few libraries with plenty of books but, as I was shocked to discover a year or so ago, public libraries don't have nearly as many books as they used to. They can "get them for you" via inter-library loan, if you are willing to wait a few weeks.
I see people in this thread talking past each other and I think the reason why is because the author is conflating 3 very different topics (inventions, technology and knowledge).
E.g. we absolutely have forgotten the knowledge in how Egyptians built the pyramids.
That doesn't mean per se that Egyptians had better/worse technology - but from a knowledge of how they built the pyramids - that absolutely is lost.
These are of the top of my head mostly covered by HN posts. Especially the antikuthera mechanism the loss was that entire that even the memory of this mechanical calculator or even the existence of other such calculator was lost. Also lost apparently was the process of creating an H bomb that"s for good but probably it has been secretly reinvented by now.
This is a clear case of a discussion appearing as it has a tech content and in fact its historical. Statistically and given tge infinite creativity of people infinite things will have disappeared, through a process of randomness, usefulness and politics a few prevail. The wheel was forgotten because it was useless. It required working economies, trade and usable roads, even dirt roads. As for printing letters if that prevailed when it was invented 2-3k years back we could have reached the stars by now.
In the times its even easier to lose technologies, they are not intuitive anymore, they are stored in miniture formats easily forgotten and stored easily corrupted and fail media, guarded behind laws, encryptions and interests. Medical technology is our lifetime's major culprit.
Even in the past 20 years the Iraqi Archives & Mosul Library were destroyed -- not to mention dozens-to-hundreds destroyed since WW1 , as well as many destroyed in ancient conflicts.
"do you think we had anything important in there?" -- "not possible, we've got everything we need already"
In other words, we don't know what we don't know because it's been destroyed and so have our records of it.
You can get even m,roe interesting if you look too China, India, or the Americas for civilizations - but most of us don't really know much about them and don't think about them when we think of lost history.
Lost tech in China is pretty fascinating. There's archeological evidence of very advanced clocks that predate similar European ones, but they seem to have lost the knowledge on how to build it with the builder (City was conquered, clock was dismantled, his son was unable to reconstruct it). Meanwhile euros were able to push their knowledge forward more or less uninterrupted from the middle ages on.
Seems pretty clear to me that we could easily lose a bunch of knowledge again. Could even argue we're very close.
I was thinking of the Antikythera mechanism, an analog computer/calculator from the 2nd century BC. There's a long gap in history from that time until Europe reached that level of technical sophistication in science and machinery.
A simplistic explanation might be: the fall of Rome led to the Dark Ages through political instability, loss of educational infrastructure, reduced trade, shift in cultural emphasis from classical science and philosophy to religion..
That reminds me of the Lost Libraries of Timbuktu, about the preservation of knowledge against time, wars, fires, and thieves. Not only technology but entire civilizations can get lost in time, like those pyramids in Natchez, Mississippi. Or the Lost City of Z, buried in the jungle.
In that sense technology - and the knowledge to understand and produce it - is always being lost to entropy unless we make an effort to keep it alive. Even then, shifts in economics or cultural context can make it impractical, unaffordable, or otherwise leaving no one to maintain it.
Even in my lifetime I feel like certain ways of thinking and living have been lost culturally. It's in the memories of those who experienced it, but after they're gone, there will only be footprints left in books, photographs, audio and video recordings, online archives and blogs.
But it's a good principle to be aware of: economic incentives often make it impossible to intentionally lose a technology once it's developed.
I would love to go back to a time in which I could safely walk my small town's streets without 3-ton steel boxes noisily charging through at speeds far faster than a horse could gallop. But we've developed the technology to build 300 horsepower mobile living rooms that the average person can buy (for a staggering quantity of debt) and we've decided to allow them everywhere. In addition to the greenhouse emissions and costs and noise pollution, they're the leading cause of death for people from ages 5 to 22, and the second most common cause from ages 23 to 67. But the staggering utility means we're not putting those back in Pandora's box.
Air conditioning, likewise, is incredibly comfortable and a massive boon for health and productivity in the hotter regions of the world - not to mention the incredible nutrition benefits of freezing or refrigerating food - but in a vicious cycle, you have to use energy that makes the world hotter to make your tiny part of the world cooler. Personally, I'll go without until June, leave the thermostat high during the summer, and turn it off come September. But a shop can generate traffic and an employer can generate productivity by spending a little more on energy costs. The genie offered us the vapor compression refrigeration cycle, and that will never be put back in the bottle by a selfish society.
New construction with moderate insulation and an efficient heat pump cooling system will have relatively small cooling costs. Include some rooftop solar in the build and it’s negligible or even zero marginal cost to run the cooling unit.
https://www.dutchreach.org/car-child-murder-protests-safer-n...
Cars have some benefits, but they're not such an impossibly high utility that they can't be deprioritized. It's a policy choice.
A nitpick but people weren't dying at 50; they were dying as infants. The average life expectancy was around 50 because so many children died, but the life expectancy of people who made it past age 5 was around 60-70: https://pmc.ncbi.nlm.nih.gov/articles/PMC2625386/ .
We have already experienced local loss of the details of technology. I think this is what the article is missing. He's looking at big things that are easy to state, while missing the tools-that-build-tools-that-build-tools foundation. I think the also "That's not tech, that's high skill" distinction is a technicality.
It is also viewing history through a filtered lens. A bit like labeling an age as a "stone" or "bronze" age, because that's the durable material that survived. Or anything involving the fossil record.
Decade perhaps, but probably not longer. We already build a lot of stuff around the world. The US makes things that we did in the 1950s when the US was the world manufacturing powerhouse - but we do so on about 1/10th as many people in manufacturing while population as doubled. We could switch many people to manufacturing - which to a large part would be automation of things China does by hand and in the end be better off - but that decade of switch we are all worse off because those jobs people are doing mostly have value.
This is a more interesting example- the theory isn't that we lost the tech, but that the new tech was too good, too pure.
The incidentals of the old, 'dirty' way of manufacturing it (that we just spent billions and billions to destroy and clean up) apparently (speculation, since classified) added some unknown impurities that affected its performance.
And either it needed to be redesigned with the new manufacturing process, or go back to the old process to fit the specs of existing weapons or the weapons would need to be redesigned.
We didn't lose the tech, but other 'advances' in both tech and society (not having workers manually handle dangerous stuff) caused an overall regression, not advancement.
It's the EPA DC chlorine case all over again, of 'progress and safety' actually increasing danger and causing overall regression.
As an immediate example, my wife's business needs p-channel small signal JFETs. These apparently are no longer fabricated, and with the way the semiconductor industry moves, they are likely never coming back in any appreciable quantity. So once the world's supply of obsoleted semiconductors dries up, the technology will basically be lost.
People used to design scientific calculators out of a relatively modest number of discrete transistors and diodes. Telephone central offices used to run on a hardwired relay computer called the "marker". How many people still walk this earth who can design stuff like that? More immediately, with your phone casually using integrated circuits with transistor counts in the billions to watch a Youtube video... people used to watch off-the-air colour TV where the TV in question was made out of less than two dozen vacuum tubes, equivalent to 1-2 transistors each. That stuff is still easily found but in another few decades will also be essentially forgotten.
Oh, and what would it take to manufacture, say, Cibachrome paper and Kodachrome film? I know the Polaroid instant film technology has been laboriously resurrected but lots of obsolete chemical processes are essentially lost.
But other historians believe they reflect a lost technological sophistication. The pyramids are incredibly precise: perfectly aligned to true north (within ~0.05°), made with millions of 2-70 ton blocks and precise internal engineering for passageways and chambers. Someone recently advocated that there's substantial subterranean infrastructure under the pyramids. The technology to move 70 ton blocks didn't exist again until the 1800's.
One of modern science's ideological straitjackets is Oliver Heaviside's restatement of Maxwell's 20 equations with 20 unknowns into four vector calculus equations. Heaviside's restatements made the math accessible to regular engineers who wanted to build things. But the restatements are arguably a simplification that neutered electromagnetism: https://x.com/TaxiCabJesus/status/1964345590604845487 (Grok had a nice answer for "what phenomenon are inadequately explained by Heaviside's four equations?").
The Coral Castle in Florida was a labor of love, built by a single man: https://en.wikipedia.org/wiki/Coral_Castle
The stones are fastened together without mortar.
They are set on top of each other using their
weight to keep them together. The craftsmanship
detail is so fine and the stones are connected
with such precision that no light passes through
the joints.
Ah yes, those giant subterran pillars beneath the pyramids. That actually interested me, so I decisively delved deeply into it, to do my due dilligence. Only to discover an early case of advanced AI-sloppery with no proven source at all. Completely made up.
And everybody copy pasting the shit! GRRR! :(
> Not really. We know how to make durable concrete; it’s just more expensive than the alternatives that last 30-40 years.
... For a span of almost 2000 years this was not the case, there was nobody you could pay to make a concrete dome like the Pantheon's, not at any price. And we only properly reverse-engineered the Roman recipe in the past decade.
I think a technology still counts as "lost" even if it gets rediscovered in the distant future!
Example, ancient natural asphalt, which literally surfaced at Red sea.
So, probably natural material, not always exactly known how appear, but fortunately existed and used by ancient people.
I only cannot agree that all lost technologies are not that we might truly care about. As example, Roman concrete could be really valuable in modern world, as current concrete just is being destroyed with time, and this is very serious problem, because could be dangerous for constructions like large dams and bridges; also, need some solution for long term burial of radioactive trash.
How to save such info, very interest question and not easy. Possible example could be some sort of long term independent shelter, under Moon surface, where could store digital archives with information.
I was surprised to learn that most cobblestone was actually quarried rectangular blocks[1], used as ships ballast, discarded after cross-Atlantic voyages.
[1] https://en.wikipedia.org/wiki/Sett_(paving)
[2] https://indianaalbum.pastperfectonline.com/photo/5FA11838-DE...
I think this could be how modern technology will be lost in the future. We have been seeing a reversal of globalization. We have seen countries guard their raw materials more tightly than before. Trade routes could disappear for geopolitical reasons. The United States was happy using Russian RD-180 rocket engines and Russian uranium after the Cold War; and look at how quickly this trade has become undesirable. Now consider materials and technologies lesser known than uranium or rocket engines, or simply less critical to a country's military might. A country might not even know it has a single sourced component until trade has stopped.
Knowledge might not be lost but rebuilding tech might take centuries depending on how worse it is. And that is probably very similar to the past. Middle age people probably understood how Romans build their water supply but where unable to reproduce it because how the bad the time was.
Roman concrete, greek fire, Indian musical sound pillars, Inca Masonry, Stradivarius Violins, Damascus Steel (the ancient technique), Polynesian Wayfinding - there are dozens of such lost arts.