We keep getting better at curing disease and preventing death, but this makes little difference in our fight against aging, due to its exponential nature.
Almost no age-related disease or condition can currently be prevented or cured. They can be somewhat slowed, but that’s about it. A rare exception is cataracts; it can be cured by replacing the eye’s lens.
This concept is inspired by established systems like Nordic civilian defense against nuclear threats or lifeboats on ships.
But those systems weren’t designed with the survival of humanity in mind, and so, they’re obviously going to be much less robust.
I might not have emphasized this sufficiently in the post, but the aim is not to achieve near 100% robustness. Instead, the goal is to provide people with a fair chance of survival in a subset of crisis scenarios.
My initial intuition is that even if 70% of the units function effectively in a crisis, this would be a success.
You need to think about how much time these shelters could buy. 70% survival for how long? A few months is probably doable, but shelters and their associated infrastructure will not last forever.
If shelters buy a few months of survival, the crisis will need to be solved in a few months. That also means the shelters will need to be targeted to experts that might be able to provide a solution or allow enough time for a solution that already existed to disperse and kill off the mirror bacteria. If a solution will need to be developed, a lot of time will need to be spent in unprotected labs which will increase risk. Think about this: you’re stuck in a suit, you can’t eat, drink, peep, poop, or even type fast (because you have thick gloves), while at the same time you’re trying to do complicated experiments to save the world. These scenarios aren’t impossible to survive, but I expect they’ll have a high likelihood of failure. So you’d probably want to aim for a least a few years rather than months.
While rigorous testing will enhance confidence and could refine the design, the significant likelihood that the shelters will work as-is—supported by Los Alamos results and cleanroom precedent—suggests that they could prudently be deployed even without exhaustive testing if a crisis emerges and the above testing is not completed.
To stretch survival to years, you’d need to do a hell of a lot more real-world testing and design work. There’s no close-enough precedent for what you’re trying to do; I highly doubt that you can only rely on lessons from cleanrooms, labs, or nuclear bunkers. Has any cleanroom or lab demonstrated perfect containment for years? How about the mobile kind? Nuclear bunkers aren’t designed to be livable for years or be sterile. At best, lab testing and case studies can indicate that hardware may work, not that it will work in the real world.
And there’s a lot more to consider besides maintaining the mechanical and electrical system that supports the suit and shelter filtering system. You’d also need climate control systems; that’s one heat pump for the suit and one for the shelter. You’d need cooking devices and indoor air cleaners or an air recirculation system. And don’t forget about the VHP system. A comms system for the suit would also be nice. But things get complicated pretty fast. I suppose you can have two or three of each suit and shelter and alternate between them to add redundancy. But things get costly pretty fast.
The more you think about it, the more impractical (and less appealing to stakeholders) it seems to get. So, to convince anyone that this is anything other than a hail mary, extensive real-world testing must be done. And maybe you can mitigate the testing showstopper I mentioned earlier by periodically sterilizing and retesting used shelters and suits. Of course, ease of sterilization will need to be incorporated in the initial design.
Unless you seal most of industry inside shelters or risk being outdoors for long periods of time, decades of survival is probably close to impossible.
Your suggestion of using permanent bonds could indeed be a practical solution in such cases.
But you’d still have a gasket where the ductwork meets the membrane (and where it would be more exposed to temperature fluctuations), and a one-piece assembly would increase costs substantially and introduce space constraints due to the need to stockpile many assembly units.
I just thought of another showstopper that makes the other issues now seem insignificant: how could you ever determine whether or not the suits and shelters work to prevent bacterial contamination? The problem here is that humans are already “contaminated” and another problem is that the world isn’t contaminated with a unique kind of bacteria or bacteria-sized particle that you could test for. So, there’s actually nothing to test for. Even if you could test for something, how could you even detect one or a few bacteria that got through? I don’t see any way around this.
Air Supply Leaks
The below diagram illustrates the airflow dynamics. The air system is designed with a series of pressure gradients (P1 > P5 > P4 > P3 > P2), ensuring that any leak results in airflow from clean to dirty areas, not the reverse. This mechanism minimizes contamination risks, even in the event of small leaks. This principle is widely used in cleanroom and laboratory settings to maintain sterile environments.
This still doesn’t address gasket leaks (leaks between the filter’s gasket material and the filter tunnel). The potential for such leaks could be eliminated be permanently bonding the filter to the filter tunnel but that would mean that the filters couldn’t be replaced.
Cleanrooms and labs aren’t failure-proof, and failure would happen a lot more often in the messiness of the real world.
Membrane Integrity and Large Holes
You’re correct that larger holes or tears could compromise the shelter. To mitigate this, the material used for the shelter will be selected for its tear resistance and self-limiting properties. Existing materials for bubble hotels, for example, do not propagate tears. For DIY or lower-cost implementations, layering materials (e.g., plastic sheets reinforced with fabric) could provide additional durability. There is already extensive research on tear resistant fabrics, as well as substantial data from people actually living in such structures, such as bubble hotels. For mass production, it would be useful to carry out research on how to achieve tear resistance across a variety of materials and fabrication methods.
Even if tiny holes or material defects wouldn’t grow into large tears due to air pressure alone, what if something else impinged on the membrane? Couldn’t a large enough stressor conceivably cause a small hole to grow? After all, suits and shelters would often get banged up by normal use and the occasional red truck.
It’s probably safe to assume that small leaks couldn’t deflate these bubble hotels, but I doubt anyone has been motivated to look at whether some of these leaks could grow large enough to let in small amounts of particulates. Suit durability probably suffers from the same lack of research.
Component Failures
While no system is failure-proof, redundancy and robustness are central to the shelter’s design. Key measures include:
Longevity Testing: Components will undergo extensive real-world and simulated stress testing. Suppliers’ lifetime analyses will be leveraged to ensure reliability.
If you’re lucky, you might get away with testing thousands of shelters and suits, but if you want something really robust, you probably need to test hundreds of thousands and potentially millions. How will you get hundreds of thousands of people to isolate themselves for years at the minimum? Mars simulation theme parks? I’m only half joking; perhaps some sort of rotation system might work, but on the other hand, that might defeat (or at least minimize) the purpose of testing the practicality of continuously (without any breaks due to personnel changes) sealing out external contamination.
Redundant Systems: Critical systems like air supply will have manual overrides and backup power (e.g., a UPS to sustain operation during power transitions). Simple mechanical solutions will be emphasized to reduce dependence on complex electronics—in a crisis it can probably be assumed that one could rely on shelter inhabitants for at least some operation and maintenance.
How long could (and should) these redundant systems last? Years? Decades? What would be their failure rate? Spare batteries can fail if they’re not used, gaskets can become brittle or warped, metal can oxidize, and so on.
Redundancy might increase durability in the short term, but it also increases complexity, and complexity can create its own problems. Complexity might not be an issue when you can usually get all of the spare parts you need, but if industry no longer exists (because you want to minimize the time you spend outside), you’d need to stockpile a lot of parts and/or entire shelters and suits. That would increase costs. And how long would that stockpile actually last? How long would membrane material remain folded without degrading along the folds in a garage or warehouse that’s not climate controlled? There are likely to be many issues like this with long-term storage.
User Training: Shelters are designed for inhabitants to manage minor troubleshooting (e.g., switching power sources).
How will you train millions of people about how to live and survive in suits and shelters before a catastrophe happens? This goes way beyond simple maintenance procedures and troubleshooting.
It would be risky to wait for a catastrophe to happen due to the possible social disorder that might occur and logistical issues with distribution (e.g., trying to outrun simultaneous releases of mirror bacteria in all major population centers).
Mass Production Challenges
Scaling production to millions of units is indeed ambitious, but starting with smaller-scale production allows us to address these challenges iteratively. The simplicity of the design—based on off-the-shelf components—makes rapid scaling more feasible compared to more complex systems. Even producing tens of thousands of units could substantially reduce existential risk in high-priority scenarios.
Where will the incentive for mass producing millions of units come from? Or even tens of thousands?
Suit Usage and External Transfers
For outside missions, the focus is on minimizing exposure. Techniques used in gnotobiotic (germ-free) animal research, such as sterilized transfer tunnels filled with vaporized hydrogen peroxide (VHP), could be adapted for human use. Vehicles retrofitted with small shelters can serve as transfer units, reducing reliance on suits for complete protection.
What happens when the suit inevitably gets dirty? There’ll be a lot more mud and dirt in a world in which infrastructure isn’t maintained, and I doubt VHP will be adequate. So, there’ll probably need to be another elaborate decontamination procedure. More complexity, more points of failure, more cost.
Will those retrofitted vehicles be self-driving? If not, the cabin would need to be shelterified. Yeah, good luck with that. If it’s a self-driving truck with a shelter bolted on, you might also need a datacenter to go alone with that. But that means you’d need to maintain the datacenter and have more spare hardware and spend more time outside and maintain a power source for the datacenter, and so on. On the other hand, if self-driving will depend only on a local system, you’ll probably need an AGI for that. But if you have an AGI, you’d also probably have an ASI which should be able to make something way better like almost fail-proof suits and shelters, self-sufficient, impenetrable underground cities, or quickly eliminate the mirror bateria threat (e.g., by drexlerian nanobots, assuming they’re physically possible to construct).
-air supply leaks: the whole air supply is inside the shelter with a fan at the inside end. Thus, any leak goes from clean to dirty and is not an issue
I’m not sure what you’re describing here. Unless you’re talking about some sort of closed-loop system (like on a submarine or spacecraft), leaks are always a possibility. Can you share an illustration of what you’re trying to describe?
-leaks through membrane (including airlock doors): not a major issue, the positive pressure will not let anything from the outside come inside
It might not be a major issue for a tiny pinhole but what about a larger hole or tear? What if that pinhole suddenly creates a larger rupture (helped out by a red truck perhaps?) in the membrane?
-shutdown due to failure of critical components is not foreseen to be an issue
Famous last words. Battery BMS fails → positive pressure is lost → bacteria gets in via tiny membrane hole(s) → everyone in the shelter dies
- all components should be possible to engineer for long continuous operation
These components will need to be mass-produced by the millions and continuously used under real world conditions to have any decent chance of being reliable. Even if certain components are already mass-produced for other uses, integrating them into a reliable system would still require integrating them into millions of shelters. But as I mentioned before, that’s not likely to happen.
The suits are indeed only 50k protection factor but it should be possible to use proven methods used to transfer germ free mice between facilities.
The leak problems that plague shelters would also apply to suits. And we are talking about using suits in the outside world, right? All facilities except shelters and perhaps food warehouses would not be protected and suits would be needed to access them.
I am happy to address this in more detail as we have spent quite a bit of time turning many stones. That said, a team of people can still make mistakes so I appreciate that you are helping me looking into this and this is part of the reason I posted—I would love to take a call to if that would be easier to hash this out.
If solutions to at least some of these issues are documented elsewhere, perhaps you can provide some links?
At least for now, public discussion seems more appropriate.
This shelter idea has many points of potential failure, possible showstoppers, and assuming a small population of shelters (hundreds or a few thousand), seems extremely unlikely to maintain an MVP for more than a few months.
Points of failure:
Leaks from the air and water filtration system (e.g., gasket leaks)
Leaks from the airlock
Leaks from the biohazard suits
Leaks from the shelter membrane
Shutdown of the filtration system due to mechanical or electrical failure
Showstoppers:
Food production or storage will require massive warehouses using the same extreme filtering as the suits and shelters. An alternative is to use some sort of disinfection tech like gamma ray sterilization, but I don’t know how practical that would be.
Producing all food indoors is currently not possible and seems unlikely be achieved anytime soon.
To mitigate the risk of these points of failure, millions of suits and shelters (along will massive amounts of supplies such as food and spare parts) will have to be manufactured and distributed, and millions of people will need to be trained in how to use them before any catastrophe occurred. Obviously, this is extremely unlikely to happen anytime soon, and I strongly suspect it won’t happen before mirror bacteria is created (due to the acceleration of biotech and AI progress) and released into the wild.
There would still be term limits: violent death, revolutions, invasions, and so on.
You might want to consider adding additional protection measures (like a respirator), as the effectiveness of some vaccines can be moderate to non-existent. The effectiveness of the flu vaccine in years when its well-matched to the circulating strains is between 40% and 60%, and when the vaccine is not well-matched, it’s protection against illness plummets, although it may still offer some protection against complications such as pneumonia. Vaccines don’t exist for bad colds and the stomach flu.
Reusable respirators will work well against any fast-spreading pandemic (assuming no ridiculously-long, asymptomatic incubation periods).
There seems to have been plenty of papers on airborne aerosol transmission of the flu and experiments with human subjects strongly suggested that the common cold is transmitted via aerosols. So, this makes it even more surprising that the experts got transmission so wrong and took forever to correct their mistake.
Yes, but your post seemed to focus on the individual, and that’s why I didn’t mention future humanity.
For humanity, it did go from no doom to maybe doom which is worse. And perhaps it’s worse for the individual in the long run too, but that’s a lot more speculative.
In any case, there’s still some hope left that our luck will last long enough to avoid doom, even if it will be by the skin of our teeth.
Until very recently, it was doom for every individual. Maybe-doom is a vast improvement.
And whatever happens, we’ll have the privilege of knowing how human history will have turned out.
The virus most likely leaked from the gain-of-function experiments that they were doing under BSL-2 and not from the BSL-3 or BSL-4 labs.
Third scenario: bat-to-researcher transmission during field work at bat caves or from the bat repository/colony or unaltered bat viruses at the labs in Wuhan.
https://www.nytimes.com/2021/06/25/opinion/coronavirus-lab.html
It’s a tricky situation. As soon as Hong Kong relaxed its pandemic strategy, excess deaths exploded. Since China followed a similar (and even stricter) pandemic strategy, it seemed inevitable that the same thing would happen (all things being equal) and millions would die with many more millions becoming hospitalized. But all things might not be equal; the circulating strains of covid in China might be less lethal than when Hong Kong relaxed its own pandemic strategy. So, it could go either way.
The real problem here is that China is playing Russian roulette; rather than using more effective vaccines and respirators, its using less effective vaccines and poorly-performing masks instead. The expert consensus seems to correctly identity the vaccine problem but still mostly ignores the mask/respirator issue, as they’ve done throughout the pandemic.
I was referring to how docs do brain surgery (e.g., infection prevention procedures, what instruments are used, where incisions are made, etcetera) rather than error rates or second opinions. I highly doubt that many non-experts (even a very motivated brain cancer patient) could successfully determine the appropriateness of specific surgical techniques for brain surgery. And since brain cancer is rare, it’s low stakes from a societal or even a personal survival point of view (although, it will become high stakes if you’ll live a lot longer than the current lifespan).
Nah, bridges (see other reply) and rockets aren’t high stakes enough to be worth worrying about.
What kind of demise are you referring to?
Bridge building is nowhere near as important as cryonics (or more appropriately, “brain preservation” technology which may not involved cryonics at all), because brain preservation tech has the potential to save hundreds of millions and possibly billions of people from certain death. Even if you disagree, it is still potentially important for personal survival way more than bridge building.
My general heuristic is that the higher the stakes (especially for personal and societal survival), the more you need to check the expert consensus (especially for softer sciences such as medicine, sociology, and economics). Examples where expert opinion should be checked (and is or was probably wrong or misguided): cryonics, certain pandemic mitigation strategies, aging research, geoengineering. Examples where expert opinion probably shouldn’t be checked very often by non-experts: brain surgery, bridge building, rocket engineering, archeological excavation.
In most situations (with some exceptions like going to the dentist) and for nearly everyone (with some exceptions like people living in a nursing home), the level of risk remaining after taking reasonable efforts to protect oneself seems miniscule.
I suspect we mostly agree about this, and the apparent disagreement was caused by a misunderstanding.
So, let me clarify: what I tried to say is that as long as individuals can protect themselves, there is no compelling reason for society to force others to protect individuals or for others to voluntarily protect individuals in those situations in which individuals can protect themselves (I probably should have been more explicit about this to avoid any confusion). For instance, if you need a root canal, you obviously can’t protect yourself by wearing a respirator (and assuming that vaccines weren’t effective), and dental staff should wear respirators and perhaps also increase ventilation. In the case of flying, individuals can protect themselves by using a respirator, and there would be no point in having anyone else mask up. Earlier in the pandemic, having everyone mask in most situations was a good policy at the societal and individual level, but now it’s not for the reasons I’ve already mentioned.
How could undecidability, unprovability, self-referential modeling, incompleteness, or any sort of logic generate the redness of red?
Incomplete, self-referential modeling → ? → red
The feeling part remains unexplained.
We still seem to have experience. How can this “seeming” feel like something? If you boil everything down to math, how can math feel like anything?
This seems contradictory.
What about torturing animals?