Suppose it happens some 250-300 years in the future.

There are some (medium-sized) cities on Mars, settlements throughout Solar System, an interplanetary communications network, many orbital space stations, etc. The population outside Earth is some 25-30 million people.

Yet the life on Earth unexpectedly dies of some disaster.

The most of industry is still located on Earth: microelectronics, most of the metallurgy capacities, especially advanced, chemical plants, etc.

The most of scientific knowledge is stored on Earth.

Mars (and other settlements) can download data from Earth's internet for some time before Earth's death, but their storage capacity, while big, is still limited, so they have to be picky on what data to store.

There is no microchip industry outside Earth. There is some iron metallurgy on Mars. They also can produce food. They have some 3Dprinters, power plants...

How can they rebuild the civilization? What's the best strategy?

For the purposes of this question, what happened with Earth is a kind of quick runaway greenhouse effect that made it look like Venus, with remaining on other celestial bodies people lacking equipment to launch rockets from Earth's surface or deploying robots there.

Essential constraints

  • they have limited data storage (at the scale of exabytes - 10^18)
  • cannot download anything after the disaster
  • the Earth's infrastructure is destroyed, except some selected valuable bits could be evacuated in the last hours or days.
  • In the span of a few months before the disaster they can download data and evacuate some equipment.
  • They have 3D printers, for construction using Mars soil
  • 3D printers for plastic, limited supply as that plastic was exported from Earth.
  • Some limited stock/stash of advanced industrial alloys and metals.
  • no technology for producing chips.
  • a few data centers are in existence, on a plus XXL side, as it was used for cached internet access.

(ed. note.: some constraints a bit inconsistent, as a result of evolving of the idea of the premise, taken from the chat/comments)

  • 1
    $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$
    – L.Dutch
    Commented Sep 26, 2021 at 12:23
  • 1
    $\begingroup$ Are there any terraformed bodies that can support human life without specialised equipment? Also, how developed is robotics? 25-30 million people may be enough to maintain and rebuild an advanced civilisation, but only if they already have advanced robots that can replace humans. $\endgroup$
    – Otkin
    Commented Sep 26, 2021 at 17:50
  • $\begingroup$ @Otkin no bodies terraformed to that extent. Robots are not comparable with humans in intelligence. There are drones and self-driving cars. $\endgroup$
    – Anixx
    Commented Sep 29, 2021 at 14:33

5 Answers 5


Well, for reference -- the surface temperature of Venus is 460 degrees Celsius, which is higher than the melting point of lead. So if Earth is in that boat, game over for Earth. In the short term, the colonies are going to have to work together to survive and grow back into industrial, inter-planetary civilisations.

I would think chips are important but pretty low down the survival pole. The thing is as long as you can get enough silicon, set up a clean-room and get some high-precision lasers / tooling you can build your own chips again. All they need is the designs of the latest generation of chip architecture and some electrical engineers to take up the discipline / try to reverse-engineer one in a laboratory. From those first few reproductions, confidence and skill grows then you scale up, I think you could build a chip industry from nothing in 10-15 years. Humanity isn't going to get far without that. Or without enough space ships to trade with each other. So they've all got to build up some industry

I would be more worried about shelter, sufficient electrical power, heat, greenhouses, water reclamation, radiation shielding... these are hostile environments and living in them is going to be difficult.

In the long term, they might work together to find a way to terraform Earth back into something habitable again, but they'd have to reintroduce life to it from scratch and given how much biodiversity was lost that is a very tall order. Without plant and animal life just having humans in isolation may well be a lost cause.

  • 3
    $\begingroup$ We have a chip shortage on Earth currently. With massive incentive from major governments and corporations to do something about it, even their huge resources haven't created new chip fabs. Building chips of the type in use 30 years ago wasn't that hard, building latest generation is almost impossible except for the best on the planet. Which is a long winded way of saying that a few million people scattered around the solar system are not going to be reverse engineering the latest chips, they can make transistors soldered to circuit boards or - if really lucky - 1990s era chips. $\endgroup$ Commented Sep 26, 2021 at 14:56
  • 6
    $\begingroup$ You are grossly underestimating the difficulty of making those ultra-modern integrated circuits. For example, there is only one company in the world, the Dutch ASML making advanced immersion photolithography machines capable of resolving features to a few tens of nanometers. They rely of a network of thousands of suppliers. Their machines are sold several years in advance. Their closest competitors are the Japanese Canon and Nikon and the American Ultratech. If it was reasonably easy to make such machines, other companies would have joined in, no? $\endgroup$
    – AlexP
    Commented Sep 26, 2021 at 15:03
  • $\begingroup$ it can't be venus temp cuz water, which makes clouds and reflects stuff. @KerrAvon2055 temporarily shortage is just not sufficient incentive, and it is not that big, sharp, suffocating - not painful enough, it more like delays and not a strict cut. Also, it is less about chips but more about packing those chips in encasing(?)(not sure about correct English word here), in a sense worse than you expect, probably, cuz it is not the most high tech part of the process. I mostly hard because it is expensive and you need to know what you do. Otherwise, it is just another factory - no magic there. $\endgroup$
    – MolbOrg
    Commented Sep 26, 2021 at 17:15
  • $\begingroup$ @AlexP 'other companies would have joined in, no?' - where do they get their expertise and know-hows, eh? spying targeting technologies isn't just for movies - it a real deal. But yeah, op may be too optimistic. okay, needed to read it first, quite opposite of ops concerns. electricity is simple stuff, shelter also not such a biggie. food water may be a problem, definitely mission critical but probably not urgent and was solved before. $\endgroup$
    – MolbOrg
    Commented Sep 26, 2021 at 17:28
  • 2
    $\begingroup$ Chips are pretty important when you're living in an entirely artificial environment where everything needs to be carefully controlled. Without the ability to maintain complex automated systems, everyone will die, because every other environment in the solar system is extremely hostile. $\endgroup$ Commented Sep 26, 2021 at 17:57

This is a Frame Challenge

Of the 20-25 million people outside Earth, the vast, vast majority of them will be on planetary or lunar bases, not manufactured bases. And the vast, vast majority of those will be on Mars. This means you have a community (if not a single city) on Mars with all the complexities of Osaka, Japan (Or, any other major city on Earth).

If you review what it takes for big cities like Osaka, London, New York, Tokyo, etc. to operate and survive, you'll realize that your premise about fundamental manufacturing remaining exclusively on Earth is mandatorily false.

The cost of creating, for example, integrated circuit manufacturing on Mars (which would actually be easier than on Earth thanks to the thinner atmosphere), is much less than the cost of transporting those goods. Never underestimate the transforming power of economics.

Here's your practical reality: while an artificial space station will be created with a dependency on some planet, no permanent outpost on another planet with the expectation of population growth that would include children would ever be planned without the capacity to operate entirely autonomously. If you think about it, it's a requirement. Otherwise a good meteor impact on Earth would kill everyone on Mars, too.

And as a small business owner who just experienced the COVID-19 pandemic, let me assure you that all it would take to create the transportation disruption I'm talking about is just that — a pandemic. If a Mars colony was as dependent as you're suggesting and the disruptions we're experiencing right now occurred, they'd be dead.

But, if you insist on removing primary industry and autonomy from Mars...

Mars would still have access to the future's equivalent of the Internet, right? That means that every piece of fundamental information needed to rebuild exists. In fact, a great deal more than is required to rebuild exists. That means you are missing only one thing:


Because the power plants, hydroponics, basic manufacturing, etc. already exist. Frankly, the only thing that could possibly hold anyone back is a lack of raw materials...

But the more I think through this, the more unrealistic the idea that such a large colony could possibly be dependent on Earth. The community would have its own university and colleges, its own corporations and public utilities. You need to process water and food, building materials (and recycling!)... I'm back to my Frame Challenge. It's unbelievable that a community the size of Osaka, Japan wouldn't actively work to guarantee that any stoppage of trade with Earth wouldn't threaten the community.

It would be a primary mandate for survival.

  • 2
    $\begingroup$ "Frankly, the only thing that could possibly hold anyone back is a lack of raw materials..." and enough energy to grow fast enough. With enough controllable energy, you can get almost anything you need from a planetary body. $\endgroup$ Commented Sep 27, 2021 at 6:19
  • $\begingroup$ Yep, the fact that there are colonies on other planets shows that Earth wants a "backup plan" for civilization. This includes all the data we've gathered and learned over the years. This can be relatively easy to do, since the Library of Congress only has about 74 TB of data. blog.storagecraft.com/big-data-storage-library-congress You can buy 18 TB drives for your home computer, so that storage capacity isn't a problem. And the Asimov book "The Martian Way" solves the problem of water and minerals by mining asteroids, etc., so industry will definitely be distributed to colonies. $\endgroup$ Commented Oct 28, 2021 at 20:44

Without the capability of producing chips, they are basically stuck somewhere in an Iron Age, meaning anywhere between Ancient Egypt and late Victorian era.

All their equipment will continue to work as long as their electronic survives, once it start failing they will need to downgrade the production to what is possible with manual control.

Food production and iron production can likely proceed, if some smart mind can re-figure out how we have done until electronic took over. For 3D printing you can simply forget about it: even assuming the 3D printers are the last to die, they will run out of the raw material in the necessary purity grade for their operation.

Their priority, once they realize their standpoint, should be taking care of medicine and food production, followed by metallurgy processing. With some luck they might be able to stand somewhere equivalent to the onset of industrial revolution, and from there try to climb the ladder again, with the advantage that they know what can be done (how to do it is another story).

  • 1
    $\begingroup$ I'm not sure why you'd think that getting things of "sufficient purity" is problematic or even really necessary. Two centuries into the future with a mature spacefaring civilisation seems like that problem would be solved. But then, the background for the question is basically an idiot plot so I guess nothing really needs to make sense. $\endgroup$ Commented Sep 26, 2021 at 18:00
  • 1
    $\begingroup$ You can't colonize a planet without a certain degree of self-reliance, one that Iron Age will simply not provide. Besides, there's no economic incentive for Earth to establish colonies entirely reliant on Earth without providing something in return - it'd be just a waste of resources. It follows that the colonies have to be good at something valuable for Earth, a thing I fail to see in a provider at an Iron Age, between ancient Egypt and Victorian times. AI drones and teleoperation would be cheaper than sending humans who require 10MJ/day just to survive $\endgroup$ Commented Sep 27, 2021 at 6:15

25-30 million people living throughout Solar System, yet not on earth is HUUUUGE number. One simply CANNOT sustain that population with industry grounded on Earth. Of course, electronics to be used in space cannot be so sophisticated as one designed to use planet-side (i.e. smartphone would be nearly-unusable in the asteroid mining environment), so they must be much more rugged, which means less capable, but by the same token easier to produce. Adding that it may be produced in microgravity, there are pretty awesome options to explore. With that presence in Solar System, most of the industry, supporting not only them, but also Earth, would be located in Earth's orbit (said microgravity offers very unique advantages for all kinds of industry) and would consume off-planed produced raw materials, and they would be already extensive and would come with their own datacenters (when watching Friends to relax after work, nothing annoys more than lag)...

But if those would be lost too, then not a big problem, too. Off-planet habitations have three major advantages going for them:

  1. Easy access to a lot of raw, high quality materials (rare earth elements are anything but rare in well-chosen asteroids.
  2. Lot of empty space. For example, moon settlements could rapidly build multiple datacenters with materials sourced from mining outposts, but basically any settlement could do that, and those datacenters would be very resistant to vast majority of external dangers.
  3. Access to workforce not only trained, but used to work in space. There could be a lot of redundant datacenters, all backing-up into each other, up and running very, very quickly.

I think you need to revise your assumptions somewhat.

For example, it would be dead-easy to produce a lot of magnetic tape in space. Currently there are projects aiming at producing tapes which can store up to 580 TB of data in the size of the human palm. make it 4 times bigger for same capacity. Best part is that it's not contingent on the very precise machines, it just needs a lot of pure elements blended the right way. In other words: no high-tech industry needed.

As of 2020 internet is estimated to take about 40 zettabytes. If you want to back up every one megabyte of it, one needs about a hundred milion tapes to back up everything. Since a lot of that are duplicates (pretty sure Netflix still has Friends on it's servers, they didn't hand their copies to HBO Max when they lost the rights, for example), so right away half of that 40 ZB goes away. Lower quality and another 40% goes away... Lower further if you want content over form... So on and so forth, and in the end you end up with much more realistic number of maybe 500 EB. This is much more manageable 500k tapes.

Even so, downloading data usually requires prioritization. If one leaves newest popculture for last (bonus content, so to speak), one can focus on literature, sciences and so on, which actually not only is not heavy on visuals, but can compress very well, backing up Earth's data rapidly should not be a problem (one tape backs up all the books ever written 4 times over)...

  • $\begingroup$ I think, lack of space is not a challenge for building datacenters, even on Earth. $\endgroup$
    – Anixx
    Commented Sep 29, 2021 at 14:41
  • $\begingroup$ @Anixx Normally, no. But since we're talking about different tech level. Or not even that. Cruder tech level. Imagine one tape I envision is the size of a building brick. I need half a million of them in optimal scenario, but double that if my calc is off... That's a lot of space. On earth one would have to consider geologically stable location, needed and available power levels, cooling, maintenance. Moon makes that easier: no quakes, can build extra nuclear reactor for power and cooling is super easy. $\endgroup$
    – AcePL
    Commented Sep 29, 2021 at 15:20

The situation isn't that bad if there aren't some special factors.

Their grand grand grand grand×8 parent's technologies, aka 21 century, aka today - are good enough for them not to die out and restore Earth to pristine conditions.

30 million people and exabytes of data - quite a disproportion in data volume and abilities to process it and work with it considering that only a fraction of them will do that. So some assistive information processing technologies may be important for the getting, at least it will make a huge difference in outcomes and how that situation develops over time. Finding relevant data in the cat videos will be not that easy, and then build up some structure out of it.

I can't get over myself and imagine such technological solutions and situations as 200-300 years in the future. I agree with Starfish Prime, with what he said in the comment section which is moved in chat now, I cite

  • "And if colonists have no ability to travel between planets anymore, and no native industry, then they've been deliberately sabotaged (or planned by complete idiots, it would be hard to tell the difference) and they're all going to die." - Starfish Prime, 2021 BSE (before space era)

Not necessarily agree on the die part, it depends, definitely on the list of possibilities, but it depends and depends in the first place on organization and cooperation between people, and then on the tools they actually have, and then on creative use of those. (Did they saved Apollo mission with duct tape back in the days, eh?)

But the described situation is not a future, so will consider it as more or less near future and today reality, and course of actions - what can be done in that situation from a position of today's reality. So, do not consider AI overlords and such good stuff, which could help them fly over the problems with ease.

  • I mean, not endorsing the view on the future and future technologies, ave nothing to do with it, just answering a q.

In general, guys are set up very well to solve their problems, we only can dream about such possibilities.

Those Venus colonies have energy advantage and potential for rapid development - and example this answer Landing humans on Venus - MacGuffinite - colonists on the Venus or around it already are in a position to start things, they already have their survival means, etc.

Here are some points, on how fast some situations can grow, how fast they could develop their energy and production capacities on the level of technologies available to them

Chips problem is indeed is an interesting one, but also it needs to point out that there are ppl who DIY that stuff in a garage, more than one, in that sense this q/a is interesting, links to that DIY guy there

This one also somewhat relevant:

This one can be your initial goal, to save the data and replace high-tech data centers, which works well with that K1 Dyson swarm goal. Considering that DIY chips guy and that low-tech CPUs are more than enough for storing and work with those exabytes of data. So no need to underestimate primitive things in space.

The main advantage of those guys is - they are already in space and there are plenty of them, and energy is cheap in space, at mars-earth-venus orbits (less so for Mars, but it is fine if we talk about the vacuum of space, not the surface of Mars)

Plan, sketch

  • To compensate a little bit for the lack of individual approach o your situation and it being mostly links, oh, btw here is a good link https://arxiv.org/abs/1612.03238 (Affordable, Rapid Bootstrapping of the Space Industry and Solar System Civilization) - read it as well.

Initial survival and race against wear and tear is sure an important problem. However, no need to overestimate the importance of it for things they are using. They do not have to replicate those. Getting in space may require more advanced technologies than living in space - live, and thrive(terms and conditions apply, past performance is no guarantee of future results).

Initial survival is important, but it should not be an issue, as delivering food to the guys is nonsense, meaning they have all those basic things covered. (if not they may die, but it not certain)

So their initial goal is the automation of the production of simple things - make sure you get the latest software from the earth and preferably opensource so you can patch and port it on your new not-so-shiny hardware later on. Automation, some basic one, smart enough, but it does not even look in direction of ai overlords - it is the staple food for the situation. It is a lever for those 30kk people to be able to start the development which can bring recovery of many if not the most of their original technological capacities, limited by that data they have.

And to begin with it, they do not need anything super-advanced, our current level of automation and software is sufficiently good for the purposes.

And staple food of staple food is the energy, the base for any activity, and here low-tech solutions can be quite effective, especially if scaled in the microgravity of space.

Focal points for guys are moons of Mars, the Moon of Earth, an orbital station in orbit of Venus(if they have some).

Moons can provide plenty of materials for implementing first second and 3rd necessity things, for bootstrapping industry in space. Venus is a bit a harder case in that regard, but air scooping can be a great help for them and that carbon they can have, but it depends on technologies.

  • creative application of tools is quite important, one guiding computer of a single ship can be used as brains in an operation of transporting asteroid things of materials between planets. Sure engines of ships are used for fine correction, and the main force has to be provided but maybe more primitive devices which have not or almost no guidance or electronics. As in terms of reactive masses - each place has plenty of it not fuel, but mass - which can be used in all kinds of electric engines.

So there are ways, and increasing production, industrial and technological capacities, step by step is possible in this situation and is a main initial goal for them, and it may take few years to get out of an initial crisis and get to the point of being able to trow as much resources at the problems as they need, and never had according to your setting description.

Their goal is just not to die in the next few years, and it should not be a problem for most of them.

P.S. read that arXiv:1612.03238 I linked


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .