Log in

No account? Create an account

Previous Entry | Next Entry

With a bang, not a whimper

Huh. I had a somewhat disturbing thought train while driving home from work, about unintended consequences.

  • That computing power will continue to increase exponentially for at least another three decades. (Probability very high)

  • That computers can be run un-emcumbered by hidden programs that allow a more Orwellian style of surveillance on what their users run on them. (Probability... moderate to high. There's always Linux.)

  • That someone will write a free or commercially available program to emulate physics accurately at a nuclear or quantum level. (Probability high. Chemistry would likely benefit from this. And geeks would love to make a reality simulator.)

  • That at some point in the future, asteroid-mining becomes possible for non-goverments. (Probability moderate.)

  • That computer assisted designing manufacturing continues to become more capable, simpler and cheaper. (Probability certain.)

  • That apocalyptic cults will continue to exist. (Probability high. Moderate, if ubiquitous surveillance starts taking them out.)
  • or
  • Sociopathic individuals with access to sufficient monetary resources will continue to exist. (Probability high. Things get cheaper as time goes on.)

It follows that it will eventually be possible for sociopathic cults or individuals to:
  • Design and virtually test nuclear weapons on computers

  • Mine the necessary elements from asteroids

  • Manufacture the necessary components [Edit: for engines and guidance systems.], as well as centrifuges (or other technology) to refine the materials.

  • Assemble working nuclear weapons.

  • Assemble working delivery systems.

  • [Edit: Make appropriate engines and guidance systems to drop asteroids from orbit.]

Thus, it seems likely that the Earth will eventually be cinderized by some cult or sociopath's nuclear [Edit: celestial] fire.

How cheerful. :)


( 15 comments — Leave a comment )
Sep. 15th, 2006 03:45 am (UTC)
Also add
More educated people will be more content (proven false)
The bread and circuses will continue. (ok)
Energy will become a major concern, as will water (very true)
Population will continue to grow in backward areas (very true)
Surveillance will become cheap and ubiquitous (high probability)

I'd worry more about the flood of child soldiers with amoral lives and AK47
clones. If I wasn't too tired to worry.

I don't think we really know enough to accurately model untested chemistry (that is, reactions that haven't been proven in the lab) much less untested physics. But that's my bias from teaching chem, not from knowing the state of the art.

And...unless they're using orbital delivery (from the asteroid belt) it's sort of easy to see it coming.

Fear more in infrastructure taggers/wreckers.

No, wait, fear the lack of caffiene. Good night!
Sep. 15th, 2006 05:42 am (UTC)
Re: Also add
Regarding modelling of physics, etc, we already use supercomputers to simulate nuclear explosions to get around nuclear test bans. I'm just extrapolating this thirty years hence when you could do that on your home computer.

Yeah, I was assuming orbital delivery.

I was totally overcomplicating this, though. If you can mine asteroids, you can drop them on people with much better booms than nukes. No need for extreme physics modelling or manufacturing or refining. Though nukes are smaller and easier to sneakily deliver than planetkiller asteroids.
Sep. 20th, 2006 12:37 am (UTC)
Re: Also add
The real danger of nukes is not their kinetic energy, as you've stated.
You get better results just throwing really big rocks, really fast, from very far up.

The danger is that a 30 lbs device (conservative estimate) can end all life in a precisely determined area, while leaving the rest of the planet realitively unharmed.

Yes, despite the screams of outrage that statement will cause amoung anti-nuke adherents... A nuke will -not- destory the planet, or even cause much long term damage on a wide scale.

Japan is not uninhabited right now. Nor is New Mexico and Nevada and the center of North America[1]. Nor is Southern Ukraine[2]. Nor are parts central Asia[3]. All these areas have were nuked over the last century. Some by design, others by accident.

Add the intention to cause pain, suffering, fear, and terror... the creation of a dirty bomb that kills millions by -NOT- causing a nuclear blast is a very real likely application in the near future.

The single scariest prospect is that the "Neutron Bomb" concept will actually be used. Minimal or no blast, but maximum particle radiation was shown to be techincally possible -and- computer modelling would put it into the reach of the everyman (with the right garage full of percision machining equipment). Hmmmm....

- Krin

[1] http://en.wikipedia.org/wiki/Image:US_fallout_exposure.png
[2] Chernobal does glow in the dark, but the grass still grows
[3] http://en.wikipedia.org/wiki/Image:Nuclear_use_locations_world_map.PNG
Oct. 22nd, 2006 02:39 pm (UTC)
Re: Also add

Hmmm, this map explains a few things...
(Deleted comment)
Sep. 15th, 2006 04:42 am (UTC)
...Seconded. Though you'd have to drop a lot of them, as it's pretty straightforward to deflect one or two if you're willing to spend the money. The old "crowbars from orbit" idea, involving many smaller projectiles that are still damaging and fairly accurately targetted, would work well too. They're hitting with energy equivalent to about 10 times their mass in TNT, so as long as they're big enough that you can neglect atmospheric resistance, medium-sized rocks are still damaging.

A few hundred carefully-aimed thousand-tonne rocks (10 kT yield) would be a lot harder to shoot down than one big asteroid, as well as being easier to drop and probably more damaging.

Sep. 15th, 2006 06:06 am (UTC)
I suspect dropping merely a dozen or so 10 kilometer asteroids at the same time would be effectively just as hard to defend against. At around 5 million Megatons each, that should cause a little stir.
Sep. 15th, 2006 07:03 am (UTC)
It would, but it's also very much more expensive, requiring an extremely large nuclear-powered craft to perturb even a near-Earth asteroid into a collision trajectory. Someone with the resources required to do this could just carpet-nuke the planet's surface instead.

Moving a thousand-tonne rock, on the other hand, takes something closer to 1e10 to 3e10 N*s of impulse. That could be done in under a year by applying about the amount of force you'd use to bench-press. Power requirements: about 10 MW (small nuclear or big solar, and about a tenth the plant I specced for the "Colonial Viper" spacecraft thought-experiment). Depending on how thick you spec aluminized mylar concentrators, you could do the solar version in 10 tonnes or less (and it would be able to move more than one rock if you set them up more than a year in advance). Launch cost: $100M or less. Big, but corporation-level rather than nation-level.

Among other things, this is one reason why I think regulation of what goes into space will get progressively tighter over the years. A much smaller craft than the boulder-dropper could easily do very expensive amounts of damage to satellites in Earth orbit, or to any hypothetical space elevator.
Sep. 20th, 2006 12:13 am (UTC)
Actually, changing the trajectory could be very cheap.

A few cans of carbon-black (art supply store $1.99) would be enough in some cases.


Take your run of the mill frozen slushball in a trojan or eros orbit and apply liberally to one precalculated spot and apply at depth. That spot will outgas and provide enough momentum to perturb the orbit enough to drop the asteroid within the gravity well of earth. Nothing more than changing the albedo asymetrically would be enough.

There are quite a few asteroids in the near-Earth orbit that occupy stable orbits, just barely. Upset their equilibrium would cause some nasty results.

Only issue, it would take decades if not more for them to arrive. But once underway, they'd be almost undetectable because they would be slow approach and are already mapped, so the slight change in their orbits would probably not be noticed until the cumulative change was so great that deflecting them would be difficult.

- Krin ^ _ ^

Link of the Day: http://www.orbitsimulator.com/gravity/articles/cruithne.html

Sep. 15th, 2006 05:27 am (UTC)
That's a really good point. Much more efficient.
Sep. 15th, 2006 04:04 am (UTC)
See the Fermi Paradox article on Wikipedia, specifically:

It's a concern, all right.

It's worse than you suggest, though. The raw materials to extinguish life on Earth are available right here on Earth; there's no need to go anywhere else.

But if we can get to the asteroid belts, the race may survive even if we destroy Earth.

There are plenty of paradoxes in this idea to go around.

. png
Sep. 15th, 2006 05:08 am (UTC)
Re: See the Fermi Paradox article on Wikipedia, specifically:
I'm not convinced that resolving the Fermi paradox requires our destruction (as that's actually pretty hard to achieve). However, it does indeed seem to be something that will bite us in the tail within about 200 years (my estimate for how long it'll be until Von Neumann replicators capable of things like colonizing the galaxy and building Dyson spheres are buildable). We're either the first in our galaxy, or for some reason absolutely everyone who came before us was rendered unable or unwilling to unleash toys like that.

My best guess: They moved somewhere else that was appealing enough that _everyone_ moved. However, this does stretch credibility (there would always be people who chose not to). It's extra-fun to try to figure out where they would have gone (I have about three candidate types of location so far), and how to look for them.

Sep. 15th, 2006 05:03 am (UTC)
Asteroid mining is an unnecessary step. Uranium is present at somewhere around 2 ppm or so in soil if memory serves. As the amount needed for a bomb is relatively small, the amount of soil you'd need to process to extract it is not unreasonable (about 100,000 tonnes). You'd get this by excavating a football field to a depth of around 30 feet, with power being the main bottleneck.

Regarding using powerful computers to simulate nuclear explosives, I doubt a terrorist would need to do this. A gun-style U235 bomb is extremely easy to build.

Regarding simulating quantum mechanics or chemistry, this turns out to be an extremely hard problem to do with classical computers, because the wavefunction of each electron and each nucleus gives added degrees of freedom, and you're trying to solve for all of them. I'd worry more about simulating biological systems to allow rapid design of tailored viruses (you can get away with making more approximations simulating things like protein folding than you can with pure computational chemistry).

You also don't have to wait for the future for this to be a problem. The equipment to let you build a custom virus from scratch is for sale now in the $100k range, and the genomes of several infectious bacteria and viruses are available for download. I even recall reading an article that specifically discussed how to hack up influenza to make it more deadly, based on examination of samples from the great flu pandemic of the early 20th century.

I've thought about this and similar problems, and my conclusion is that humanity needs to spread out greatly within the next 100 years, to form colonies that are very expensive to travel between, forming a natural quarantine. Interstellar colonization would be ideal, but interplanetary would probably be adequate for the time required to build the infrastructure for interstellar colonization.

The flip side, though, is that it would require the application of very great resources - beyond the means of terrorists - to wipe out humanity completely. A future society would be forced to make itself more damage-resistant, and to tolerate routine loss of life from terrorism, but I strongly suspect that drunk driving will continue to kill far more people than terrorism even in such a scenario.

We're living in interesting times indeed.

Sep. 15th, 2006 06:09 am (UTC)
Re: See the Fermi Paradox article on Wikipedia, specifically:
Mining enough Uranium is actually much harder than that, as you have to refine it to get the useful isotopes, which means you need a lot more. On Earth, most all the reasonable sources of Uranium are state controlled by various governments. Sadly, one said government is North Korea.
Sep. 15th, 2006 07:07 am (UTC)
Re: See the Fermi Paradox article on Wikipedia, specifically:
Already taken into account. You need less than 1 kg for a bomb. At 1 ppm, that's 1000 tonnes. U235 is at 0.7%, so 100,000 tonnes is in the right ballpark.

The ore deposits that are state-controlled are the 1000 ppm ores. Common dirt will give you 1 ppm; it's just not considered economical to extract.
Sep. 15th, 2006 07:28 am (UTC)
Cost and power.
Just for giggles, I worked through a worst-case cost estimate for sifting through dirt for ppm elements. Per kg of _dirt_ (not ppm element), you spend about $1 worth of electricity (17 kW*h for 10 eV per atom and assuming mostly oxygen). This amounts to about $100 M per kg of U235 extracted in this manner. Ammonium nitrate is probably cheaper (even if you have to make it out of air and water).

I'm really tempted to try building a ppm-scraper just to see what I can find in a wheelbarrow load of dirt, now. Probably more lead/mercury/cadmium than I care to think about.
( 15 comments — Leave a comment )