Hacker News Reader

Comment by pg | Hacker News Reader

Despite the generally frivolous answers, this is an interesting question and one I often wonder about. What turns out to take up all the time if we want to reproduce where we are now? Presumably the optimal plan is to spend practically all your effort on machine tools.

It would be interesting to be able to figure out what would be the best benchmark of progress. Would it be the precision with which you could machine metal? That might do up to about 1900.

It might turn out that most of the time was spent on something nontechnical, like moving stuff from place to place before you'd developed fast ways of doing that. So maybe in practice the most important benchmark would be how fast you could move stuff.

Reproducing where we are now would in some ways be harder than getting here was. E.g. the most accessible coal and mineral deposits used to be sitting right on the surface, but now those are gone.

View on HN

Minerals are definitely the hardest part.

You wouldn't need coal. If you can make a fire, charcoal is easily produced from wood.

Making forges and blast furnaces isn't too difficult. To build one, pile up dirt or clay walls to form a bathtub like structure. Make sure you have holes at the base so air from your bellows can make it though. Construct a fire in the tub and pile on alternating layers of charcoal and iron ore. Keep pumping on the bellows for the next few days until the the charcoal burns up. Your iron ore should now have turned into pieces of high carbon steel, perfect for making any tool from ploughs to blades.

You might be able to find iron ore as dark colored sand in stream beds. I don't know where you would find copper or zinc.

Once you have steel, everything follows. If society knows about existence of a technology, reproducing it shouldn't be difficult. The hardest part about progress is inventing brand new things.

If you are interested in learning how to make machine tools from very simple materials, look up the Gingery series on making a complete metalshop from scrap:

http://www.lindsaybks.com/dgjp/djgbk/series/index.html

I've built the gingery furnace, an poured aluminum twice. It's hot and scary. I've made arrowheads out of old glass bottles using a broken antler.

I think you're underestimating how spectacularly hard it is to make anything without tools. Go into the woods naked, and show me how to make a bellows.

I'm pretty sure i could do it, if i could keep my glasses, and my shoes and a knife and a bunch of food. I think i could run down a deer, if all the other people i'd be in competition with didn't kill me after i was 10 miles into the run.

I don't think steel matters very much. Sanitation, clean water, lots of food.

"If society knows about existence of a technology, reproducing it shouldn't be difficult. The hardest part about progress is inventing brand new things."

I disagree. Once upon a time, people drove cars on the moon. Should be a n easy thing to reproduce eh? heck, we've got 40 years of technology on them. The hardest part about progress is convincing other people to do what you want them to do.

The gingery books are FANTASTIC. i'd highly recommend them to ... everyone.

I guess I meant that coming up with new ideas is hard. Invention is hard.

In the hypothetical case that everything is gone tomorrow, the ideas still exist. All we need to do is implement them.

I'll explain it in CS terms. What would happen if all copies of the quicksort algorithm were destroyed overnight? Someone would spend a few hours and write another one because he/she knows about the algorithm. It would take about an hour.

Consider the further case that no one on earth knew that the quicksort algorithm even existed. How long would it take to be duplicated? Months? Years?

How did someone get the idea to smelt metal? It took thousands of years. Now that we know that smelting exists, all that is left is finding a way to do it. Many people in modern society will have the knowledge to rebuild technology. I would argue that expanding technology is much harder than rebuilding technology.

As to pouring molten metal, sure it is scary at first, just like driving a car. After much practice, like most things, it ceases being scary.

> I'll explain it in CS terms. What would happen if all copies of the quicksort algorithm were destroyed overnight? Someone would spend a few hours and write another one because he/she knows about the algorithm. It would take about an hour.

That's like saying you just woke up in a Blacksmith's shop, with the anvil and fire ready. That's the last mile, which isn't the hard part. Compare it to destroying all traces of CPUs and computer hardware/software. Okay, now go fabricate a processor.

Okay, now go fabricate a processor.

Step 1: Go and build a glassblowing shop.

I don't think I am explaining myself well.

What I mean to say is that once you know about an invention, reproducing it is much less difficult than actually inventing it. Do you disagree?

In high school, I was obsessed with metalworking and built several forges. I played around with melting metal and forging blades. I created charcoal. I read everything I could get my hands on about blacksmithing. I read a lot of fiction books about rebuilding post-apocalyptic societies.

I guess I have a different perspective.

I understand where you are going. If we lost it all tomorrow, it would be quite hard to reproduce tools, machinery, high precision equipment, etc. to get back to where we are. For new inventions, you need both to do all the building of the machinery that builds that machine that will build the new invention, but you also need the novel new idea that is the invention to be made. 2 hard (and quite different) problems to solve, instead of just the first (which, as mentioned, is ridiculously hard in and of itself).

thx for the gingery series link

i'm also interested in how-to-make-stuff-from-chemistry series (candle, soap, pulp, paper, ink, oil, nylon, plastic, etc) preferrably from the same author(s)

not necessarily from scratch (burn tree to get ash etc)

if you have links please post, thx

Minerals wouldn't be hard. The existing artifacts lying around are better than the richest old ore deposits.

The most urgent work would be to make books or scrolls to write down the knowledge of the survivors before they get senile.

One of the hardest intermediate steps would be developing a sufficiently large cooperative social unit. Right now I'm typing this on a computer that was made by people I've never met in a number of different countries that I've never visited; structures like international trade agreements and currency exchange mechanisms make it possible for us to have a mutually beneficial economic relationship. (Size of cooperative unit makes a pretty good benchmark of progress, extending past 1900 to the present.)

If technology disappeared, our social and economic organizations would suddenly extend no further than the tribal level, i.e. the number of people we could have cooperative relationships with would be limited to the people we saw in person on a daily basis.

Even assuming we still had the necessary schematics to understand what we were trying to build, it would take a long time for us to get sufficiently well-organized to manufacture it in quantity. Of course, it's possible that in this starting-over scenario, technological progress would outpace the rebuilding of social organization, to the point where we'd have all the same electronic goods, for instance, but each region would make its own. Or maybe we'd have to duplicate the transportation-technology revolution of the early and mid-20th century before we could get anywhere with the communications-technology revolution of the last 20+ years.

>One of the hardest intermediate steps would be developing a sufficiently large cooperative social unit.

Churches would probably be the nucleus for these. There are already numerous churches that are beyond tribal size, and they are of course geographically localized.

It's funny to see this discussion popping up here, I've been thinking about this quite a lot. I've never really made any progress: coming up with good answers requires a lot of cross-discipline knowledge that I just can't bother to track down in my spare time. Geology, chemistry, materials engineering, mechanics/industrial design etc, etc. It stands to reason that all the science and engineering disciplines would be involved.

Machine tools are the foundation of just about everything, but there's a lot more to do once you can produce and shape metal. Chemical production is a big task once you have the handling of basic materials out of the way, and energy is important right from the start (electricity!). You'd have to have easy access to clean, and hot water to maintain good hygiene and a decent standard of living, which is important for morale. Which questions do we consider - just the technology part, or the kinds of social precautions you would have to take? A struggle for resources causes all sorts of difficulties. How would the economy work? There's no way to maintain governments like those we have today when there are no means for fast communication.

What is the easiest way to bootstrap technology from scratch? It would probably be a lot easier in temperate climates where food is abundant; in many parts of the world too much effort would have to be spent on just staying alive. The focus would have to be on tools right from the start: gather resources to build stuff and then use your new gear to build more stuff. Even if everybody knew exactly where we were going, it would take decades at the minimum. There is just so much to do, and so much of our technology depend on things we have developed already.

We could write books about this stuff. You could probably dedicate whole academic careers to the question, especially if you really want to try out the practical aspect. I'm disappointed about most of the posts on this story, there is so much interesting stuff to explore here.

Speaking of writing books about stuff, communication equipment would be probably of an equal or even higher priority to sanitation since efficient information transfer will greatly speed up the re-engineering effort. Furthermore it wouldn't be terribly difficult to do early on. You don't need too much metallurgy to cast type, and the rest of a basic letterpress could be made of wood. Paper isn't too hard to make. Neither are batteries, so as soon as you can produce wire efficiently you can set up a telegraph. Telephone, radio, typewriters, Linotypes, and lithography would all require more mechanical sophistication, so they can wait a bit.

>What is the easiest way to bootstrap technology from scratch? It would probably be a lot easier in temperate climates where food is abundant; in many parts of the world too much effort would have to be spent on just staying alive.

If technology ceased to exist today, there would certainly be a massive die-off. Farmers would not have the ability, and possibly not the motivation to feed all of the newly-useless urban dwellers. Even if it would be possible for low-tech farming to feed everybody, it would be impossible to retool and retrain millions of people in time to plant this year. California would get it the worst. How would they even get across the desert and mountains to places where food is grown without irrigation?

If we can head towards magnets and wire, we could be on for an early boost from electricity and radio as an alterative to the entire postal service (much much easier than rebuilding all the roads and railway).

How much useful electricity can you get from a donkey walking round in a circle driving a generator instead of milling flour?

Cut ahead a few years, we can avoid the effort of roads entirely if we brush the ground flat and pick hovercraft and hot air balloons for longer distance matter movement.

If the population would remain the same, then all the effort will have to go into feeding it. And most likely will fail.

The next hurdle is organizational. You have to get the engineers together and working.

Once you cleared both, I'd guess not so long. Furnaces aren't really high tech, if you don't go for volume. Simple machine tools could probably be made out of wood, with only the essential parts of metal and using alternative power (animal, water, wind...).

In the end it'd still be a matter of administration (market?). If you get a group of knowledgeable people do one thing only (a furnace, a machine shop, an assembly line) it wouldn't take more then 1-2 years per step and many can be done in parallel.

"Reproducing where we are now would in some ways be harder than getting here was.."

Opposing factor to speed up these sorts of setbacks is that much of the experimentation/testing would already be done. We'd know the fuel efficiency of coal for example. We'd know not to bother with methods that have failed.

I'm more curious what technologies we'd realize never needed to exist. Would electric cars be the route we choose? Would open source operating systems be 100% adopted...etc

What you say is interesting, but it would also be interesting if you'd actually answer the question, i.e., give a guess in years.

Reply on news.ycombinator.com