It’s funny when I read the Foundation novels, it seemed silly to me that people could lose track of technology and fall backwards.

Then you read something like this and realize Asimov had a point!

its a bit of an over used trope in fiction to me that if you kill the one person inventing a thing you stop it. think t2 and the chip. most major discoveries i have heard of in important areas had multiple people running neck amd neck.

when it comes to rebuilding a thing, i think whats important to wonder is would you even build it the same way? I remember seeing an article on ars about the engines for the Saturn 5. They took over a thousand hours of weld time at a level of skill and quantity that was built up during that time period. And they got them working via experimentation. Building the now was done differently. They 3d scanned the parts and used modern additive and cnc machining and used way less welds (one can argue additive metal is just cnc welding but I digress). And then simulated it in the computer. I remember a discussion about the baffles in the nozzle to stop pulsing in the combustion, that was figured out in destructive trials of this incredibly difficult machine. You'd do that in simulation now.

I likely got details wrong.

But it's to say that part of "they don't build em like they used to" also means "they learned". Look at bridges, they don't look like Brooklyn or Eads anymore. They used to build them with waaay too much steel.

That said roman concrete is a thing they've been chasing for a while

> That said roman concrete is a thing they've been chasing for a while

Roman concrete isn't some magical substance that we're hoping to replicate - we already know how it works - it just would really suck for the type of building projects we have now.

Aah, I thought we were trying to replicate it or one of it's properties still, maybe in the context of modern concrete. Would love to know more.

Unfortunately a lot of the discussion online regarding Roman concrete veers uncomfortably into a rather, shall we say, "supremacist" angle. For a level discussion about Roman concrete, I found this to a nice place to start.

http://engineeringrome.org/understanding-roman-concrete/#Anc...

In my (probably flawed) understanding, I think the drawbacks of Roman concrete in a modern setting is that it takes a really long time to cure (like a year or so) and it uses salt water, which would not interact well with regular steel rebar.

If you ever watch the documentary "Connections" by James Burke, you get the impression basically all inventions are caused by the circumstances where two realms of science meet up in a new way to solve a problem, and that the people involved are a detail of passing importance.

> It’s funny when I read the Foundation novels, it seemed silly to me that people could lose track of technology and fall backwards.

> Then you read something like this and realize Asimov had a point!

And also wildly over-optimistic (as is typical for SF). In the books, I think it took something like 10,000 years to loose the technical knowledge. I bet we could do it in 100.

20 years.

A complicated technology like present aircraft, nuclear energy, petrochemical refining, pharmaceuticals, etc. is resting on the shoulders of a few thousand 40-something-year-olds that have the book learning of the 25-year-olds coupled with the experience and judgement that makes the technology practical, and particularly the knowledge of how to go from zero to one multiplied across many many sub-problems.

If something puts a particular technology out of practice for 20 years, many of those now-60-somethings, while still alive, will have forgotten much of what they knew, and not be suitable to put the hard hats back on and get in the field.

And this is not accounting for the effects of whatever catastrophe put the technology out of practice to begin with.

There are two forms of loss, loss of the theory, and loss of the practice. We will lose the practice - all the tribal knowledge that comes from doing it quickly. However so long as we can still read the physics, engineering and math text books we can recreate something similar. It will take a few years and we will make mistakes, but take a bunch of "smart people" and give them time and they could re-create anything from first principals.

It only took a few years for the Manhattan project to create nuclear bombs once physics advanced to the point where we realized such a bomb wouldn't need to be the size of an aircraft carrier. Only a few years after that (and a much smaller budget) to make a non-exploding power reactor. There is no reason society couldn't duplicate that if we wanted.

Now the tribal knowledge does help a lot, it you still have it you can take a lot of time off. However it isn't needed.

Wasnt it rather that it would take 10,000 years to recover to restore the galactic civilization after the crash ?

The key insight is that data is not knowledge.

We probably still have all the papers and journals and schematics and whatnot. What we don't have, however, is people with the appropriate training to make sense of those things. Without those people - without actual working knowledge of the technology - all that data is useless.

It's like reading other people's code.

You can figure out what it does rather easily. Figuring out why it was done that way requires a lesson from the school of hard knocks in a lot of cases.

In the case of the nuclear industry those knocks can be rather significant.

On this topic, I recommend watching this video:

https://www.youtube.com/watch?v=ZSRHeXYDLko

Thanks for posting that link; I had not seen that lecture before and found it highly enjoyable. For others that might gloss over a random link to youtube, this is a lecture from 2019 delivered by Jonathan Blow titled "Preventing the Collapse of Civilization." In this talk, he discusses the problem of generational transfer of knowledge within the historical context of civilizations who failed to do so successfully. At about the halfway point, he transitions to the related topic of declining software quality, which I suspect will resonate strongly with many programmers on this forum.

Even if you didn't have loss of knowledge or increased security measures, unless future French nuclear power can lower its costs, it will inevitably have a lower share in the French energy mix.

The CO2 cost is low though. You don’t need to fire coal and gas power plants when it’s dark and not windy.

The long term vision is for those gas plants to be hydrogen-powered, with the hydrogen generated by the surplus of wind and solar during peak times. Pumped hydro for power storage is also something that will become viable. You lose a lot in conversion, but excess wind power is very cheap.

Honest question, but do people actually believe that? It sounds like a tactic by oil companies to sucker in gullible people in order to punt a problem down the road. It also acts as a great excuse by politicians to avoid having to deal with a problem.

IMO, the reality is that these plants will never convert to anything other than natural gas and will eventually be shut down, still running on natural gas.

Hydrogen is not a fuel, it's an energy storage mechanism, and not a very good one at that. If you're going to store energy there's much more efficient and/or cheaper ways of doing it.

Natural gas is still a good fuel but I don't think anyone in the US at least is under any illusion that these will ever run on anything other than natural gas.

Well, any plan for green energy has major holes in it. If it was easy someone would have already done it. The EU is taking this very seriously however as part of their climate plans.

The big problem with the hydrogen economy is not the conversion of the gas plants, it is having enough wind to generate the hydrogen in the first place. You need excess wind capacity to make hydrogen, and no private venture will invest into wind turbines that aren’t needed. Without a strong government hand shaping this market it will not happen.

If you are looking for a skeptical take, here’s a good one: https://cleantechnica.com/2021/12/06/new-eu-hydrogen-policy-...

In Des Moines more than 80% of all power comes from renewable sources. Despite that my power costs are good. Iowa has a lot of land relative to people, but still there are a lot of power that other states are throwing away. (Texas as someone else mentioned is also doing well)

Texas of all places are building tremendous numbers of wind turbines primarily driven by market forces.

Having their own grid has some serious downsides as they found out last February but the upside is that they get to run things so that their installed base of renewable power is larger than the backlog of projects waiting for approval which is the case most of the country is in.

Moving quickly to renewable energy with only a few downsides sounds like a great tradeoff. Especially if they're downsides that consumers can work around.

I seem to remember that you can’t even pump hydrogen through existing natural gas infrastructure because the smaller hydrogen molecules are more prone to leaking. I would imagine the equipment at a plant has a similar problem and combine that with lower efficiency from burning a fuel that the plant wasn’t designed for, and the abysmal efficiency from converting electric energy to hydrogen and it seems like pure snake oil.

Edit: the problem I am remembering is that hydrogen actually degrades the metals used in pipelines (https://www.hollandsentinel.com/story/opinion/columns/2021/0...) so I don’t see how any of this infrastructure is useful at all for hydrogen.

Not to mention ignited hydrogen leaks being invisible in daylight!

You can walk right into one if you are not careful - which is why you might see people walking with brooms in front of them around hydrogen equipment. :)

Coal gas contained significant amounts of molecular hydrogen (~50% or so) and it had been used for decades. What magic did those people know back then?

Pumped hydro has been viable for energy storage for decades and decades, running 70-80% efficiency. It's more efficient than most alternatives, to include compressed air, but it can't be dispatched quite as quickly.

The real problem is the fact that environmentalists tend to hate dams, especially large hydro projects, so there's a major political battle on top of finding funding for these kinds of capital-intensive projects.

Even if we could build dams with no worry about the environment, it wouldn't help. There are not many places left to build them, and a lot more would be needed. Sure use them where we have them, but they are not a something we can grow.

> There are not many places left to build them

...unless you go radical? https://www.anu.edu.au/news/all-news/anu-finds-530000-potent...

Pumped hydro can be built off rivers. The opportunities for that are vast.

Not at the scale needed.

Incorrect. There are some geographic restrictions, but on a global scale the potential off river pumped hydro resource is enormous.

http://re100.eng.anu.edu.au/global/

One would hope that by 2050 energy storage and/or long-range transmission will be a non-issue for renewables, rendering this argument moot.

Given the latency for big projects we’d likely have at least a somewhat clear idea of what that storage would be approximately now. Do we?

wind/solar/battery projects are very fast to spin up. That's another thing which makes them attractive over nuclear.

I hope too.

In unrelated news, lowering nuclear to 50% of electricity is set as a political goal.

In other unrelated news, safely maintaining the current nuclear electric power would have required a massive buildup starting 10 years ago (to replace the reactors that are currently arriving at the end of their life).

I thought the political goal was maintaining nuclear as high as 50%.

Political promises are rarely so detailed.

Though that sarcasm of mine is slightly outdated, a new promise (exploiting the recent high energy prices) is indeed to build some new reactors, postponing the previous promise of reaching 50% (from the current 75%) from 2025 to 2035 :

https://www.world-nuclear-news.org/Articles/Macron-says-Fran...

(Because it's not like the politicians aren't aware of these issues, it's just that it's rarely politically convenient to raise them.)

That's 28 years away. People not yet born will be trained nuclear engineers by then. On what basis does RTE regard the loss of knowledge as already locked-in?

Areva has a lot of trouble hiring new engineers. Nuclear power is not as popular in France as it once was. Also, you don't only need engineers: you need lots of industrial workers (such as welders), which, because of deindustrialization, are not that common in France anymore.

It's not only the loss of knowledge, it's also the higher security standards. The last reactor being built, Flamanville 3, was supposed to be ready by 2012, is still being built and not estimated ready before the end of 2022.

Probably on the basis of not expecting a rapid increase in people wanting to become nuclear engineers...

That would contradict "even if she wanted to, France couldn't". If she wanted to, France could make nuclear engineering a highly-paid profession with strong job security. It would just cost more money.

It's impossible to make binding long-term promises in a democracy. Voters are allowed to change their minds, and the risk that they do that and completely destroy nuclear engineering as a career option is always present.

Something like software engineering is a much safer career option, because the risk that your country will categorically ban it is pretty low. And because it doesn't tie you to a small number of possible employers or force you to live in a specific location.

And thus we arrive at today's world, where the best and brightest of our generation are working on optimizing ad clicks instead of solving grand challenges. le sigh

Austria comes to mind - they built the fully functional nuclear power plant, and never had it started - because the public referendum was narrowly (50.47%) against it. Billions of dollars down the drain, hundreds of careers derailed.

https://en.wikipedia.org/wiki/Zwentendorf_Nuclear_Power_Plan...

> It's impossible to make binding long-term promises in a democracy. Voters are allowed to change their minds, and the risk that they do that and completely destroy nuclear engineering as a career option is always present.

That's true of any political system, the problem with democracy is more that it's invariably fickle. A dictator may also be fickle, but there's a least a chance he has a long term vision with a late payoff that he pursues for a his lifetime.

But somebody has to maintain all those nuclear missiles and nuclear subs.

What's the percentage split for civilian vs military employment for nuclear industry jobs?

> It's impossible to make binding long-term promises in a democracy.

We can make binding financial promises at least. It's entirely possible to grant nuclear workers each a personal 60 year annuity managed by a consortium of foreign banks, and requiring the worker to pick between working as an engineer, returning the annuity, or sitting in prison.

This, it really is just a question of political will and cost.

It's difficult to ascribe a single will to a nation. It's more of a split-brain issue.

And making certain professions high-paying isn't as viable a strategy as you might think. Multiple factors go into that. Nuclear engineering isn't a popular thing, just like animal testing.

I have serious doubts that would ever happen. Everything is gridlocked for programs like that in France.