Is hard scifi space war possible?

[lPeregrine]

I've seen plenty of speculation on the technological question of "can two spaceships kill each other, and how would they do so?", but much less on the question of whether the battle would even plausibly happen in the first place. In this question, I'm making three fundamental assumptions about

1) It's hard science fiction. While it doesn't have to be strictly limited to what we have in 2009, the laws of physics should be firmly in place, and technology should be limited to things that we can reasonably expect to see given enough time, not just "this book on theoretical physics says it might be possible". So that means no FTL, no magical unlimited-fuel engines, no shields, etc.

2) There is at least rough technological parity between the various sides. I'm thinking of the traditional epic fleet battle, not the first planet to develop FTL travel and shields killing everything in its path. The technology doesn't have to be exactly identical, but it shouldn't be a case where one side has engines that are 100x as efficient (naturally revealed at a critical point in the plot). Advantages and disadvantages are fine, but they should be subtle: for example, Earth might have better focus on their lasers (and therefore better range), while Mars might have more fuel-efficient engines (and therefore better delta-v for the same hull mass), but the differences are not decisive.

3) All sides behave fairly rationally. I know it's not strictly necessary, but, suicidal religious fanatics, absurdly overconfident evil empires, and similar things all feel like kind of cheating a bit. Of course if it works for rational sides, you can always bring in the irrational ones if you want to have them in your story.


It seems to me that there are a couple consequences to this:


First, the extreme speeds and distances will remove much of the strategy from space battles and reduce it to a question of numbers. Sure, there will be strategic decisions in setting up the battle (do I invade Mars, or do I keep my fleet in orbit to defend myself?), but at some point the outcome is inevitable. Instead of the hero's smaller ship beating the villain's battleship through bravery and skill, all that will remain is to add up probabilities: subtract the defender's interceptor missiles from the attacker's offensive missiles, multiply by the probability of each missile scoring a kill, and wait for the explosion. All of this will be done by computers, the human crew will be little more than passive observers.

Secondly, stealth in space will be limited to nonexistent (see here for a detailed explanation). Well before you get to a range where you can shoot your opponent, you will probably have most of the knowledge you need to predict the outcome (see above). From passive observation and a little physics, you'll know the other ship's velocity and acceleration, and therefore its mass. From the mass and assumption of rough parity, you will have a fairly good estimate of its capabilities. From knowledge of its approximate delta-v capability and starting location and velocity, you will have a pretty good idea if an intercept is possible, and if so, when and where it will happen.



So you know how the battle will be set up long before any shots are fired, and you know that once it happens the outcome will be fairly predictable. So why fight at all? If a month before the battle the Mars fleet realizes that the Earth fleet outguns them by 50% and has enough of a delta-v advantage to make escape impossible, why not surrender? Since we are assuming rational behavior, what would be gained by going into a fight that will be suicide? I suppose we could assume that Mars has a second fleet, and the first fleet's goal is to do enough damage so that when the second fleet arrives they will win, but that just broadens the question a bit. In this situation, knowing the capabilities and location of the two Mars fleets, why would Earth attempt to invade in the first place?


In fact, why not abstract it even more? Why even attempt an invasion when you can just send a radio message to your opponent saying "accept our authority and send your taxes to us instead of Jupiter, or we will destroy you"? I could see battleships becoming little more than votes in the political game: one battleship = one vote. Instead of a war, you would have politicians attempting to assemble a bigger coalition of the various armed factions, followed by a list of demands to the opposition. The ships themselves would rarely ever even leave their parking orbits except on training missions, and an actual shooting battle between two small ships might be the kind of legendary event that is talked about for centuries.





So, I ask the experts: plausible or not? Can you think of any ways in which actual combat would happen?

[Sarevok]

The arguement also applies to any Earth battles, On paper the US would have utterly crushed Japan. Yet the Japaneese still initiated war. Why ? War is not so simple that you can just count ships and missiles and never fight for real.

[lPeregrine]

The arguement also applies to any Earth battles, On paper the US would have utterly crushed Japan. Yet the Japanese still initiated war. Why ? War is not so simple that you can just count ships and missiles and never fight for real.

This is true to some degree, but there is also a lot more bad information in historical examples. In the 1930s, you had the horizon limiting visibility in setting up a battle, armies safely hidden away in their bases, etc. In a space war, there is no hiding your forces, everyone knows roughly what you have, and exactly where it is. It simply isn't possible to have a case of "oops, we didn't know they had so many battleships".


Not only that, but the duration of the war is a major factor. The US only utterly crushes Japan on paper once you include years of building up their military in the middle of the war. In the case of a space war, where parking a fleet in orbit around the other planet will end the war, there is much less time to build and train extra ships, and there are no unprotected factories to do it with. By the rules of space war, Pearl Harbor would have meant the end of US battleship construction, and probably a very different outcome to the war.



But regardless, that's the reason why I included the third point. If you want to have space battles, it's very easy to write a story in which you justify a war with "the leaders of Mars are idiots/suicidal religious fanatics/etc", but that's really taking the easy way out. What I'm looking for is scenarios in which both sides could legitimately want to go all the way to an actual shooting war.

[Fire Fly]

I just read a recent article on the The Physics of Space Battle (which is reprinted from here) and it sounded quite interesting.

The Physics of Space Battles

Joseph Shoer is a Ph.D. candidate in aerospace engineering, studying how modular spacecraft could be assembled, and hoping that they will be the telescopes and human exploration vehicles of the future, and not for crushing the dreams of Martian colonists.

I had a discussion recently with friends about the various depictions of space combat in science fiction movies, TV shows, and books. We have the fighter-plane engagements of Star Wars, the subdued, two-dimensional naval combat in Star Trek, the Newtonian planes of Battlestar Galactica, the staggeringly furious energy exchanges of the combat wasps in Peter Hamilton’s books, and the use of antimatter rocket engines themselves as weapons in other sci-fi. But suppose we get out there, go terraform Mars, and the Martian colonists actually revolt. Or suppose we encounter hostile aliens. How would space combat actually go?

First, let me point out something that Ender’s Game got right and something it got wrong. What it got right is the essentially three-dimensional nature of space combat, and how that would be fundamentally different from land, sea, and air combat. In principle, yes, your enemy could come at you from any direction at all. In practice, though, the Buggers are going to do no such thing. At least, not until someone invents an FTL drive, and we can actually pop our battle fleets into existence anywhere near our enemies. The marauding space fleets are going to be governed by orbit dynamics – not just of their own ships in orbit around planets and suns, but those planets’ orbits. For the same reason that we have Space Shuttle launch delays, we’ll be able to tell exactly what trajectories our enemies could take between planets: the launch window. At any given point in time, there are only so many routes from here to Mars that will leave our imperialist forces enough fuel and energy to put down the colonists’ revolt. So, it would actually make sense to build space defense platforms in certain orbits, to point high-power radar-reflection surveillance satellites at certain empty reaches of space, or even to mine parts of the void. It also means that strategy is not as hopeless when we finally get to the Bugger homeworld: the enemy ships will be concentrated into certain orbits, leaving some avenues of attack guarded and some open. (Of course, once our ships maneuver towards those unguarded orbits, they will be easily observed – and potentially countered.)

Now, let’s talk technology.

First, pending a major development in propulsion technology, combat spacecraft would likely get around the same way the Apollo spacecraft went to the Moon and back: with orbit changes effected by discrete main-engine burns. The only other major option is a propulsion system like ion engines or solar sails, which produce a very low amount of thrust over a very long time. However, the greater speed from burning a chemical, nuclear, or antimatter rocket in a single maneuver is likely a better tactical option. One implication of rocket propulsion is that there will be relatively long periods during which Newtonian physics govern the motions of dogfighting spacecraft, punctuated by relatively short periods of maneuvering. Another is that combat in orbit would be very different from combat in “deep space,” which is what you probably think of as how space combat should be – where a spacecraft thrusts one way, and then keeps going that way forever. No, around a planet, the tactical advantage in a battle would be determined by orbit dynamics: which ship is in a lower (and faster) orbit than which; who has a circular orbit and who has gone for an ellipse; relative rendezvous trajectories that look like winding spirals rather than straight lines.

Second, there are only a few ways to maneuver the attitude of a spacecraft around – to point it in a new direction. The fast ways to do that are to fire an off-center thruster or to tilt a gyroscope around to generate a torque. Attitude maneuvers would be critical to point the main engine of a space fighter to set up for a burn, or to point the weapons systems at an enemy. Either way, concealing the attitude maneuvers of the space fighter would be important to gain a tactical advantage. So I think gyroscopes (”CMGs,” in the spacecraft lingo) would be a better way to go – they could invisibly live entirely within the space fighter hull, and wouldn’t need to be mounted on any long booms (which would increase the radar, visible, and physical cross-section of the fighter) to get the most torque on the craft. With some big CMGs, a spacecraft could flip end-for-end in a matter of seconds or less. If you come upon a starfighter with some big, spherical bulbs near the midsection, they are probably whopping big CMGs and the thing will be able to point its guns at you wherever you go. To mitigate some of the directionality of things like weapons fire and thruster burns, space fighters would probably have weapons and engines mounted at various points around their hull; but a culture interested in efficiently mass-producing space warships would probably be concerned about manufacturing so many precision parts for a relatively fragile vessel, and the craft would likely only have one main engine rather than, say, four equal tetrahedral engines.

How about weapons? Well, we have to consider just how you might damage a spacecraft to put it out of action.

Explosions are basically a waste of energy in space. On the ground, these are devastating because of the shock wave that goes along with them. But in the vacuum of space, an explosion just creates some tenuous, expanding gases that would be easily dissipated by a hull. No, to damage spacecraft systems, you can’t hit them with gas unless it’s really, really concentrated and energetic. So unless you want to just wait till your enemy is close enough that you can point your engines at him, the best bets for ranged weapons are kinetic impactors and radiation.

A kinetic impactor is basically just a slug that goes really fast and hits the enemy fighter, tearing through the hull, damaging delicate systems with vibrations, throwing gyroscopes out of alignment so that they spin into their enclosures and explode into shards, puncturing tanks of fuel and other consumables, or directly killing the pilot and crew. You know…bullets. But it sounds much more technical and science-fictiony to say “mass driver” or “kinetic lance” or something of the sort. Of course, the simplest way to implement this sort of weapon in space is just as some kind of machine gun or cannon. Those will work in space (ask the Soviets, they tested a cannon on their first Salyut space station), and the shells will do plenty of damage if they hit anything. However, space is filled mostly with empty space, and hitting the enemy ships might be a challenge. Furthermore, if the impactors are too large, the enemy could counter them by firing their own point-defense slugs and knocking the shells out of line. Therefore, I contend that the most effective kinetic space weapons would be either flak shells or actively thrusting, guided missiles. The flak shells would explode into a hail of fragmented shards, able to tear through un-armored systems of many craft at once without the shell directly hitting its target, or able to strike a target even after it tries to evade with a last-minute engine burn. The missiles would be a bit different from the missiles we are used to on Earth, which must continuously thrust to sustain flight. In space, such a weapon would rapidly exhaust its fuel and simply become a dummy shell. No, a space missile would either be fired as an unguided projectile and power up its engine after drifting most of the way to its target, or it would fire its engine in sporadic, short bursts. A definite downside to kinetic weapons on a starfighter is that they would impart momentum to the fighter or change its mass properties. Very large cannons or missiles might therefore be impractical, unless the fighter can quickly compensate for what is essentially a large rocket firing. Even that compensation might give the enemy just the window he needs…

Radiation-based weapons that burn out the electronics of a spacecraft sound exotic, but are still potentially achievable. This would be the attraction of nuclear weapons in space: not the explosion, which would affect just about nothing, but the burst of energetic particles and the ensuing electromagnetic storm. Still, such a burst would have to be either pretty close to the target vessel to scramble its systems, or it would have to be made directional in some way, to focus the gamma-ray and zinging-proton blast. But while we’re talking about focused energy weapons, lets just go with a tool that we already use to cut sheet metal on Earth: lasers. In space, laser light will travel almost forever without dissipating from diffraction. Given a large enough power supply, lasers could be used at range to slice up enemy warships. The key phrase there, though, is “given a large enough power supply.” Power is hard to come by in the space business. So, expect space laser weapons to take one of three forms: small lasers designed not to destroy, but to blind and confuse enemy sensors; medium-sized lasers that would be fired infrequently and aimed to melt specific vulnerable points on enemy space fighters, like antennae, gimbals, and maneuvering thrusters; and large lasers pumped by the discharge from a large capacitor or similar energy storage device to cut a physical slice into the enemy craft wherever they hit. Such a large weapon would likely only be fired at the very beginning of a battle, because the commander of a ship with such a weapon would not want to keep his capacitor charged when it might unexpectedly blow its energy all at once once he’s in the thick of things.

Deflector shields like those in fiction are not possible at present, but it would still make sense to armor combat spacecraft to a limited extent. The spaceframes of the fighters would likely be designed solely for the space environment; the actual ships would be launched within the payload fairings of a rocket or assembled in space. If launched from the ground, armor must be minimized to reduce the launch weight of the spacecraft. But if built and launched in space, it would make sense to plate over vital systems of the vehicle. Thick armor would prevent flak or small lasers from piercing delicate components, and might mitigate a direct strike from a kinetic impactor or heavy cutting laser. However, the more heavily armored and massive a space fighter is, the more thrust it will take to maneuver in orbit and the more energy it will take to spin in place. (Here’s where computer games get space combat all wrong: the mass of a huge space cruiser would not place an upper limit on the speed of a vehicle, but it would reduce the acceleration a given engine could produce compared to the same engine on a less massive vehicle.)

I’m assuming that we’d have some intrepid members of the United Earth Space Force crewing these combat vessels. Or, at least, crewing some of them – robotic drone fighters would be a tremendous boon to space soldiers, but the communication lag between planets and vessels in orbit would make the split-second judgments of humans necessary at times. (Until we perfect AIs… but if we’re giving them the space fighters from the beginning, we deserve the robot uprising we’ll get.) The crews will hardly be sitting around nice conference-room command bridges with no seat belts; nor will they be standing upright in slate-gray console pits with glowing glass displays all over. It’s not even a good idea for them to have windows, which would be vulnerable to flak and could give the crew an intense sense of disorientation as the spacecraft maneuvers, and could give them tremendous trouble adapting to rapid changes in light levels as the ship rotates near a planet or star. No, they should be strapped into secure couches and centrally located in the most protected part of the spacecraft. They should also be in full pressure suits, and the interior cabin of the spacecraft should already be evacuated – to prevent fires, or any secondary damage if all the atmosphere rushes out a hull breach. This also reduces the need for escape pods. Camera views from the exterior of the ship and graphical representations of the tactical situation would then be projected directly onto helmet faceplates.

Now, for the final word, let’s say the United Earth Space Force defeats the Martian rebels in orbit. What do we do to hit them on the ground? Well, strategic weapons from space are easy: kinetic impactors again. You chuck big ol’ spears, aerodynamically shaped so they stay on target and don’t burn up in the atmosphere, onto ground targets and watch gravitational potential energy turn into kinetic energy and excavate you a brand-new crater. At some point, though, the imperialist Earthlings probably want to take over the existing infrastructure on Mars. Time to get out the Space Marines!

It’s not terribly expensive or difficult, comparatively speaking, to get people from orbit down to a planet surface. You fall. This is the purpose of a space capsule. What’s really, really, prohibitively difficult is getting them back up again. So, the victorious orbital forces would have to bring in a transport ship chock full of Space Marines and drop them all at once in little capsules (little because they can only be so big for the atmosphere to effectively brake them, and because you don’t want all your Marines perishing in some unfortunate incident). Some orbital forces would remain in place to threaten the ground with bombardment and give the Marines a bit more muscle, but really, the ground-pounders are going to have to be pretty self-sufficient. If they ever want to come back up, they would have to build and/or fuel their own ascent vehicle. (This is the problem facing any NASA Mars efforts, too: getting back up through the Martian atmosphere is much harder than any of the lunar ascents were.)

What would combat spacecraft end up looking like?

Well, there are good arguments to have both large and small spacecraft in the Earth forces. A big spacecraft could have a lot more armor to keep its systems and crew safe, more room for large fuel tanks and electrical power supplies, and larger mass to resist impulses from cannon recoil. However, a smaller craft would be less visible to radar, more maneuverable, and could achieve higher accelerations for constant engine thrust. As with just about any military force, the role of the craft would be tailored to the tactical operations required, so the Space Force would probably include several sizes of craft.

Enemies could come at your ship from any direction in space, which means that you would want to react, strike, and counterattack in any direction. So, you would either have to mount weaponry all around your starfighter, put the weapons on gimbals so that they could rapidly point in any direction, or make the fighter maneuverable enough that it could rapidly point in any direction. Gimbals would be a bad option, because they would introduce points of increased vulnerability, unless they could be very well-armored. I conclude that the big ships would have many weapons, pointed in many directions; the small ships would have a few weapons, with the main weapon systems pointed in one direction.

Maneuverability (angular acceleration) you could achieve with gyroscopes, or by mounting engines or thrusters away from your fighter's center of mass. For the highest levels of maneuverability, the spacecraft should be close to spherical and these engines should be as off-center as possible, which might mean putting thrusters on long booms or struts. The problem with this kind of Firefly-like engine layout is that it becomes very vulnerable. If a fighter can achieve high maneuverability with gyros, those are probably the best option.

So, I think the small fighter craft would be nearly spherical, with a single main engine and a few guns or missiles facing generally forward. They would have gyroscopes and fuel tanks in their shielded centers. It would make sense to build their outer hulls in a faceted manner, to reduce their radar cross-section. Basically, picture a bigger, armored version of the lunar module. The larger warships would also probably be nearly spherical, with a small cluster of main engines facing generally backward and a few smaller engines facing forward or sideways for maneuvering. Cannons, lasers, and missile ports would face outward in many directions. On a large enough space cruiser, it would even be a good idea to put docking ports for the small fighters, so that the fighters don't have to carry as many consumables on board.

I think it's time to sketch some pictures and write some stories!

I certainly hope we don’t get into any space wars. Human nature being what it is, though, and given how scarce a lot of resources really are on the scale of a solar system or a galaxy, I don’t think it’s out of the question. I would like to think that when we start colonizing other worlds, we will be sufficiently enlightened to do so from on board the Ship of the Imagination, and not as futuristic conquistadores. Still, the part of me that loves science fiction has fun with these thought experiments.

[Lord of the Abyss]

This is true to some degree, but there is also a lot more bad information in historical examples. In the 1930s, you had the horizon limiting visibility in setting up a battle, armies safely hidden away in their bases, etc. In a space war, there is no hiding your forces, everyone knows roughly what you have, and exactly where it is. It simply isn't possible to have a case of "oops, we didn't know they had so many battleships".

That's been discussed on this site before, and the "no stealth in space" bit is somewhat exaggerated. Especially in this scenario. You can see that there's a big installation built into Ceres? Fine; how do you know it's military? If it's military, how do you know what's inside? Something similar goes with ships; the "you can tell what mass they are, so decoys are pointless" claim ( as even the Atomic Rocket site admits ) only works for rockets that produce a visible plume, and can't tell the difference between a warship and commercial ship of the same mass. Which to me means that at least for long distance travel warships will use one of the other methods of propulsion that don't make a visible plume ( mass drivers, rotating tethers, etc ) if at all possible ( because no exhaust plume means decoys should work ), or just be hidden in the commercial traffic.

And in a scenario where people are sending space fleets after each other, there are likely to be a lot of non-military craft and installations. After all, there's no point in militarizing space if there isn't anything there to defend or attack. And that in turn means that there are going to be an awful lot of spacecraft and installations that you won't necessarily know are military or not. Which means you won't have perfect knowledge of the military capacity of the enemy.

[sirocco]

About the Japanese vs. Pearl Harbour, you could apply it somewhat to space. If everyone thinks that there is no point in fighting in space, there would be a lot of military ships that have nearly never been used. An opponent could wonder if they are as efficient as they look and just take the chance of striking you as soon as he can.

And one thing people do forget is that surveillance systems just look at where they are pointed at. And space is huge. If there was a way for an enemy to know where we are looking, he'd have a significant tactical advantage.

[RedImperator]

Wait a minute, though: we're not talking about fleets skulking in the Oort Cloud here, where the only way to know what's going on is to point a telescope in the right direction and hope you notice the right pixels. Conventional intelligence isn't going to stop working just because it's The Future. I find it hard to believe, for example, that one side isn't going to know what the other side's battleships "look" like (including thermal signature, performance characteristics, et cetera) and could somehow confuse them with civilian traffic. Same goes for secret bases and the like; OK, true, I can't see what you're doing under the surface of Ceres, but if my spies discover you just recently placed an order for fifty million tons of battle widgets, I can be pretty sure you're not building Santa's Workshop up there. Space is a very sensor-friendly environment to begin with, and good intel will fill a lot of the gaps that are left. It's probably impossible to remove all uncertainty, but there's going to be a lot less than we're used to on Earth.

[sirocco]

Wait a minute, though: we're not talking about fleets skulking in the Oort Cloud here, where the only way to know what's going on is to point a telescope in the right direction and hope you notice the right pixels. Conventional intelligence isn't going to stop working just because it's The Future. I find it hard to believe, for example, that one side isn't going to know what the other side's battleships "look" like (including thermal signature, performance characteristics, et cetera) and could somehow confuse them with civilian traffic. Same goes for secret bases and the like; OK, true, I can't see what you're doing under the surface of Ceres, but if my spies discover you just recently placed an order for fifty million tons of battle widgets, I can be pretty sure you're not building Santa's Workshop up there. Space is a very sensor-friendly environment to begin with, and good intel will fill a lot of the gaps that are left. It's probably impossible to remove all uncertainty, but there's going to be a lot less than we're used to on Earth.

So space war will still be a war of information : "know what is true, show what is false, hope that you were right".

But I keep thinking that using a civilian ship for the first strike (or even later in the war) is a good tactic because it is a dirty one. And I'm sure it's the way those war will be fought. The Geneva treaty will certainly be used as toilet paper.

[RedImperator]

Why is that necessarily true? Nations didn't just sign the Geneva treaty out of the goodness of their hearts; they signed it because it was useful to their interests. Using one of your own civilian freighters as a first-strike vessel is an invitation to have all the rest of them swatted out of the sky; that's a lot of civilian deaths and necessary infrastructure destroyed. In a total war, sure, absolutely, but given the potential destructiveness of space warfare, rational states would seek to avoid it, either through limited warfare or no warfare at all.

Anyway, war is and always will be a war of information. That's nothing new. People just seem to forget sometimes there's more than one way to get it.

[Lord of the Abyss]

I find it hard to believe, for example, that one side isn't going to know what the other side's battleships "look" like (including thermal signature, performance characteristics, et cetera) and could somehow confuse them with civilian traffic.

What if they are designed to be able to look like a commercial vessel ( at least at long range ) for just that reason? Performance envelope won't tell you anything if they choose to to exercise their ship past what a commercial craft can do. And in dock; half a solar system away you won't be able to tell the difference between a foil bubble with a heater inside and a foil bubble with a warship inside.

Same goes for secret bases and the like; OK, true, I can't see what you're doing under the surface of Ceres, but if my spies discover you just recently placed an order for fifty million tons of battle widgets, I can be pretty sure you're not building Santa's Workshop up there.

Well, that's true right now as well. Although more flexible manufacturing technology might make even that harder; if it only takes a different program to switch between consumer products and weapons, then you'll be stuck tracking feedstock that could have many uses, instead of finished materials that have only one.

[lPeregrine]

The problem with the civilian ship sneak attack is that a war, especially one requiring such a massive investment as a space war, isn't just going to happen out of nowhere. The emperor of Earth isn't going to suddenly wake up one day and decide "hey, I think I'll invade Mars today", there's going to be a reason for the conflict, and probably considerable effort to resolve the conflict without an expensive shooting war. Add in the long travel times for a realistic spaceship, and the defender will have months, or even years to prepare for it.

So what happens to your sneak attack plan when the other side declares that all incoming traffic will be intercepted and inspected a week away from the target? All it takes is a single small ship to do the inspection and your secrecy is gone. Either you let them board and search your ship, or you blow it up and reveal your hostile intentions.

And if the ships are armed with weapons other than lasers, there's going to be a considerable delay between firing a shot and hitting the target. Sneaking into position with a "civilian" ship isn't all that effective if your target has several hours between firing and impact to ready their defensive systems and fire a return salvo that will kill your ships even if you manage to get a 100% kill rate before it hits. And if you out-gun the opposition by a solid enough margin that you will survive the return fire, why not simply come in openly and broadcasting surrender demands?

[dworkin]

One of the big questions is judgig whether the other side wants to fight. In the scenario of Martian sepratists they are betting on the administrators deciding "It was an arse to administer, let them do it if they can." against the allure of unobtanium, buried alien starships and hookers with three boobs.

[RedImperator]

I find it hard to believe, for example, that one side isn't going to know what the other side's battleships "look" like (including thermal signature, performance characteristics, et cetera) and could somehow confuse them with civilian traffic.

What if they are designed to be able to look like a commercial vessel ( at least at long range ) for just that reason? Performance envelope won't tell you anything if they choose to to exercise their ship past what a commercial craft can do.

Try to imagine building a destroyer that looks like a container transport. You could do it, but not without serious performance tradeoffs. The radiators alone are going to be a major bottleneck for a warship trying to pose as commercial traffic.

And in dock; half a solar system away you won't be able to tell the difference between a foil bubble with a heater inside and a foil bubble with a warship inside.

So you tell the "cultural attache" at the local Earth consulate to amble on down to the docks and see which foil bubbles have astronauts hanging around. That was my entire argument: even if you can successfully hide a warship (and by the way, how is turning on a heater inside a foil bubble supposed to fool anyone if the other side is watching to see where your ships dock?), there are other methods to learn what the enemy is doing. Just because you fooled surveillance doesn't mean you've actually hidden anything.

Same goes for secret bases and the like; OK, true, I can't see what you're doing under the surface of Ceres, but if my spies discover you just recently placed an order for fifty million tons of battle widgets, I can be pretty sure you're not building Santa's Workshop up there.

Well, that's true right now as well. Although more flexible manufacturing technology might make even that harder; if it only takes a different program to switch between consumer products and weapons, then you'll be stuck tracking feedstock that could have many uses, instead of finished materials that have only one.

That was just an example. The point was that there's more than one way to tell what's going on. Of course it's true right now as well; the whole point of my post is that existing intelligence-gathering methods that don't rely on surveillance aren't going away and will serve to fill in the gaps left by surveillance. The OP is wrong in assuming perfect information for both sides, something I already conceded, but it will be much more difficult to hide anything in space.

[Zixinus]

I think another aspect of the war could be brought up: the possibility of surrender. The likehood of starships carrying nukes that can turn Britain into an island of radioactive glass. So the question of whether to surrender or not will come up quickly.

The real question is, say you have orbital superiority. How much leverage does that give? How likely is that the population will surrender? How close will it get you to archive what you want?

Because the way I see it, once an interstellar war starts, it will be an orgy of genocide upon genocide on a scale that can only be described as "extermination" and "mass destruction". With little to no space for something that can be called "mercy", as the scale will not allow such. A war like this can only go until one side either absolutely surrenders comes in or is completely devastated.

Current wars are fast and very furious: the civilian population of at least one side will always be within the crossfire and every missed shot can crank up a body count of a hundred.

I see that interstellar war will either last until either one's side defence is devastated or until one's side population is devastated. We are unlikely to talk about weapons with less energy than what could wipe away a small country and I doubt that the enemy will not fire on planet-side to destroy defences with any civilians around it. Unless a planet or habitat relies solely on orbital defences or unless it surrenders, that option will come up.

One of the arguments against this is numbers. The enemy has more ships in orbit than what we can reliably destroy. Or the defender's force is too large for us to overcome and we'll be destroyed before we can do any damage, so it is better to hang around the moon, where we are safe. Either the enemy's position, force or information is overwhelming, so we must stop any destruction.

But do officers that have lost their link with chain of command always act rationally?

[Junghalli]

Interstellar warfare in a hard SF universe is an interesting topic. The thing is, accelerating significant amounts of mass to significant fractions of c is hard, while if a civilization has Von Neumanns and access to the resources of a solar system they can probably pretty easily build up a huge network of laser platforms capable of throwing appreciable percentages of the power output of their star around. It's a situation that will heavily favor defense over attack.

I imagine the only really viable way to attack a heavily industrialized star system would probably be to send a couple of Von Neumanns to nearby stars, build giant lasers arrays, and hose the target system with massive lasers from a distance of light years until you're satisfied you've burninated any significant defense systems, then send conventional forces in to mop up any survivors.

Of course if the target civilization has Von Neumanns and AGI any constructor that slips the net is potentially enough for them to eventually rebuild and try to take revenge on you.

Another factor to consider is such an attack would be highly conspicuous in terms of energy emissions and stands a pretty high chance of being detected by other civilizations, including ones you haven't encountered.

Assuming rough parity between the two sides I'm seeing a MAD scenario here.

[Formless]

I disagree. Jung, I think you underestimate how hard it really is to build a type 2 "we harness the power of the sun itself" civilization, especially in a setting where the solar system has internal factions which may have cause to go to war before any of that can be accomplished. Consider especially the fact that Von Neumans can never be considered safe for exactly the reason you just gave; anyone who sends one out rather than much more specific facilities with specific industrial roles is essentially giving the finger to everyone with a scope watching for such acts of aggression and war. And as for the mythic sun-lasers, that's the kind of super weapon that is only going to come into play in INTERSTELLAR war, not interplanetary wars that only involve one solar system, or for that matter in any setting where there is FTL. Its certainly not the kind of weapon you would expect a civilization with internal factions would likely be using, because there is always the question of which faction gets to have that kind of power. And I highly doubt humanity is ever going to unify to the extent that internal conflict will stop being an issue-- barring transhumanist fiction, which frankly speaking makes most readers cringe and stop reading.

[Junghalli]

I disagree. Jung, I think you underestimate how hard it really is to build a type 2 "we harness the power of the sun itself" civilization, especially in a setting where the solar system has internal factions which may have cause to go to war before any of that can be accomplished.

I was initially going to say that the energy of a K2 would be vastly overkill for establishing a pretty wide exclusion zone, but looking back at the tables I derived this from (which I wrote down months ago, so my memory wasn't too fresh), reliably establishing a 10 AU exclusion zone does seem to take K2 comparable levels of energy, going by a quick calculation I did a moment ago. That said, said calculation assumed enemy ships capable of accelerating at 10 G indefinitely, which is a bit of a generous assumption to the attackers, especially if they're planning on doing anything besides a hit and run attack. And then there's the factor that really valuable targets are likely to be in the inner system, where the necessary kill radius of the laser will be much smaller (the problem is light lag between the laser and the target; you need a laser beam with an enormous radius to reliably hit far away targets). Simple geometry says at 1 G the numbers will look about 100 times better for the defenders and worse for the attackers. Nevertheless, I need to think more about how the numbers would shape tactics and strategy, so for the moment you should probably take what I said before with a lot of salt. Mea culpa.

And as for the mythic sun-lasers, that's the kind of super weapon that is only going to come into play in INTERSTELLAR war, not interplanetary wars that only involve one solar system, or for that matter in any setting where there is FTL.

I was talking about interstellar war. And I assume any question that asks about how things would happen in a "hard sci fi" universe excludes the possibility of FTL by default.

[Formless]

And then there's the factor that really valuable targets are likely to be in the inner system, where the necessary kill radius of the laser will be much smaller (the problem is light lag between the laser and the target; you need a laser beam with an enormous radius to reliably hit far away targets). Simple geometry says at 1 G the numbers will look about 100 times better for the defenders and worse for the attackers.

Again, I don't think that is necessarily the case. The gas giants are all major sources of fuel, for example, and as you mentioned seeding the Oort cloud with von neumann warships, fortresses, and not-so-innocent colonies which can later be used in sneaky ways reminiscent of the shell game are all viable strategies. After all, not all of the solar system's resources are in the inner solar system, so its entirely possible that wars might not involve it at all.

And if we're talking about star systems besides sol, the game could be drastically different. But that goes without saying.

I was talking about interstellar war.

Yeah, uh, why? As far as I can tell everyone else was assuming interplanetary war, between factions of humans. The tactics and strategies of the two kinds of wars are more different than night and day.

And I assume any question that asks about how things would happen in a "hard sci fi" universe excludes the possibility of FTL by default.

Why? All sci-fi has to propose something that's effectively magic or otherwise unobtainable. There is a lot of gray area that you are disregarding that has FTL but otherwise tries to be as true to science as possible for which this discussion would be relevant. That it changes the game slightly is true, but unavoidable. And why you would even bother contemplating interstellar war without FTL is beyond me, since the only reason I can think of that you would waste resources on interstellar war in such a setting is out of paranoia.

Edit: Whoops! I just double checked the OP, and indeed it specified no FTL (which is fine considering that it does bring in its own problems...). However, as I already pointed out, that pretty much rules out interstellar war by default as well due to the fact that there is nothing to fight over at such distances that would be worth the massive resource drain.

[Junghalli]

Again, I don't think that is necessarily the case. The gas giants are all major sources of fuel, for example, and as you mentioned seeding the Oort cloud with von neumann warships, fortresses, and not-so-innocent colonies which can later be used in sneaky ways reminiscent of the shell game are all viable strategies. After all, not all of the solar system's resources are in the inner solar system, so its entirely possible that wars might not involve it at all.

True, it depends on the particulars of the scenario.

Yeah, uh, why? As far as I can tell everyone else was assuming interplanetary war, between factions of humans. The tactics and strategies of the two kinds of wars are more different than night and day.

I was responding to this:

Because the way I see it, once an interstellar war starts, it will be an orgy of genocide upon genocide on a scale that can only be described as "extermination" and "mass destruction". With little to no space for something that can be called "mercy", as the scale will not allow such. A war like this can only go until one side either absolutely surrenders comes in or is completely devastated.

Why?

Because FTL is extremely improbable and therefore not "hard".

All sci-fi has to propose something that's effectively magic or otherwise unobtainable.

I think that statement more accurately describes fantasy. Science fiction, by the very term, is fiction that is grounded in science. It may speculate about stuff that's "effectively magic or otherwise unobtainable", and a lot of science fiction does contain such stuff, but there's absolutely no reason for that to be a requirement. Unless by "otherwise unobtainable" you mean to include stuff that's unobtainable to our present civilization but is well-grounded in known physical reality (like AI, uploading etc.), and I hardly see how that's comparable to something like FTL which is basically magic. And even that definition is problematic. For instance, does Bulk Food's central conceit qualify as something "unobtainable"? What about a story exploring the possible implications of human cloning? We could clone humans in real life right now if we wanted to.

And why you would even bother contemplating interstellar war without FTL is beyond me, since the only reason I can think of that you would waste resources on interstellar war in such a setting is out of paranoia.

An unlimited replicator is the most obvious possibility. If you seek to turn all usable matter in the universe into more of yourself then sooner or later some of that usable matter is going to happen to be the bodies and habitats of other intelligences.

[Formless]

Yeah, uh, why? As far as I can tell everyone else was assuming interplanetary war, between factions of humans. The tactics and strategies of the two kinds of wars are more different than night and day.

I was responding to this:

Sorry, I wasn't paying that close attention to the thread. Close to the holidays, busy, you understand. However, the question of why any civilization would waste their resources on such a monumentally challenging task as waging war with people light years away when there are no resources those people have that you can't find in adequate supply at home still stands.

Why?

Because FTL is extremely improbable and therefore not "hard".

All sci-fi has to propose something that's effectively magic or otherwise unobtainable.

I think that statement more accurately describes fantasy. Science fiction, by the very term, is fiction that is grounded in science. It may speculate about stuff that's "effectively magic or otherwise unobtainable", and a lot of science fiction does contain such stuff, but there's absolutely no reason for that to be a requirement. Unless by "otherwise unobtainable" you mean to include stuff that's unobtainable to our present civilization but is well-grounded in known physical reality (like AI, uploading etc.), and I hardly see how that's comparable to something like FTL which is basically magic. And even that definition is problematic. For instance, does Bulk Food's central conceit qualify as something "unobtainable"? What about a story exploring the possible implications of human cloning? We could clone humans in real life right now if we wanted to.

This is why I prefer the term "speculative fiction". All the sci-fi you just described, and all of it I'm sure you could name, are speculative in nature. Just because its currently plausible according to our current understanding of science doesn't mean that it will ever necessarily happen; hence the word "unobtainable". Think for a moment about what it would take to wage war in space in the first place: we would need engines much more powerful than the one's we have today. Any given one you might propose may never pan out, and the materials science needed for others may not be possible. You forget that in science your own unstated assumptions about the setting are no less magic until real world evidence shows otherwise.

And by the way, although we could clone human beings nowadays, there wouldn't be a story there that anyone would want to read; the common conception of cloning is of minds being duplicated, not DNA being taken from one person and used to make an otherwise normal infant * . In order to make a fulfilling story about the second, according to our current understanding of the sciences involved you would have to resort to... fucking magic which may look absolutely nothing like any real world process, even one which preforms the same process proposed. Again, Speculative Fiction.

* well, okay, there is the standard morality play of "hey, this guy is a clone. Lets be bigoted asses to him", but nothing about that story is specific to cloning which would make it a story about cloning. It could just as easily be a story about race, and it would achieve the same thing.

And why you would even bother contemplating interstellar war without FTL is beyond me, since the only reason I can think of that you would waste resources on interstellar war in such a setting is out of paranoia.

An unlimited replicator is the most obvious possibility. If you seek to turn all usable matter in the universe into more of yourself then sooner or later some of that usable matter is going to happen to be the bodies and habitats of other intelligences.

And why don't the replicators start cannibalizing each other for parts? Also, this isn't a problem if the replicators aren't immortal, which is probable. Look at earth for an example. Life qualifies as an "unlimited replicator seeking to turn all matter in the universe into itself," and its VERY difficult for most organisms to get into space-- only human's have succeeded. Yet the above is exactly what has happened-- the replicators formed a biosphere, with life feeding on life. Even if we suppose intelligent replicators programmed not to harm one another, it only takes one "mutation" in that programming and suddenly you have predators. That's the interesting thing about replicators: as long as the copy mechanisms for the code that tells the replicator how to build offspring aren't perfect you have the basic mechanisms of evolution just waiting to happen.

I don't see how this isn't the more likely scenario to replicators managing to preform interstellar invasions. And that is assuming von neumann's are even possible in Hard Sci-Fi: the mere fact that important materials aren't evenly distributed throughout the solar system and/or universe tends to put a crimp on self replication.

[Formless]

Oh, one more thought about replicators. Why would a civilization build replicators in the first place? We don't build machines without some purpose for doing so. What is the purpose of building replicators that do not have a limit to how much of the universe they can consume, especially since it would obviously be dangerous to do so? After all, if it wasn't, they wouldn't be a danger to us either, and you would have no story.

[Junghalli]

Just because its currently plausible according to our current understanding of science doesn't mean that it will ever necessarily happen; hence the word "unobtainable". Think for a moment about what it would take to wage war in space in the first place: we would need engines much more powerful than the one's we have today. Any given one you might propose may never pan out, and the materials science needed for others may not be possible. You forget that in science your own unstated assumptions about the setting are no less magic until real world evidence shows otherwise.

All possible technologies that can't be implemented right now are not equally "magic". AI can't be implemented right now, but it's completely consistent with the laws of the universe as we understand them and there's every reason to believe that the limits that prevent us from creating it are purely those imposed by our ignorance. Indeed, there'd have to be something fucking weird going on for AI to not be possible; there would have to be more to consciousness than a sufficiently complex and correctly structured brain, or some sort of magic that keeps AIs from existing, both of which would require that our present scientific understanding actually be wrong. By contrast with FTL there is reason to believe that we are prevented from achieving it not simply from ignorance but because the universe just doesn't work in such a way as to permit it. Our scientific understanding would have to be wrong for it to work. There is a big difference there.

And by the way, although we could clone human beings nowadays, there wouldn't be a story there that anyone would want to read; the common conception of cloning is of minds being duplicated, not DNA being taken from one person and used to make an otherwise normal infant * . In order to make a fulfilling story about the second, according to our current understanding of the sciences involved you would have to resort to... fucking magic which may look absolutely nothing like any real world process, even one which preforms the same process proposed.

There's a scenario similar to the one in The Island where rich assholes have themselves cloned so if they need a replacement organ they can swipe a genetically compatible one from the clone. That's pretty biologically plausible. As far as copying minds go, all it requires is to make the clone's brain a high-resolution physical copy of the original's. This is quite a bit beyond the kind of cloning we can do right now, and not really cloning at all but a different process that could be combined with it, but it is quite physically plausible. It would still be hard SF, assuming the mechanism wasn't explained as "genetic memory" or some similar bullshit.

And why don't the replicators start cannibalizing each other for parts? Also, this isn't a problem if the replicators aren't immortal, which is probable. Look at earth for an example. Life qualifies as an "unlimited replicator seeking to turn all matter in the universe into itself," and its VERY difficult for most organisms to get into space-- only human's have succeeded. Yet the above is exactly what has happened-- the replicators formed a biosphere, with life feeding on life. Even if we suppose intelligent replicators programmed not to harm one another, it only takes one "mutation" in that programming and suddenly you have predators. That's the interesting thing about replicators: as long as the copy mechanisms for the code that tells the replicator how to build offspring aren't perfect you have the basic mechanisms of evolution just waiting to happen.

This assumes the replicator is mindless, which is unwarranted assumption. There's no reason you couldn't have an intelligent being that would have endless replication as a goal. Indeed, it's the goal evolution would probably have produced if it wasn't so lousy at designing at designing brains. The classic scenario for how such a thing might arise is an AI that booted itself into an endless self-enhancement loop while attempting to solve some incredibly difficult or infinite mathematical problem, and develops converting all available matter into more and more processors to crunch more and more numbers as a subgoal.

Oh, one more thought about replicators. Why would a civilization build replicators in the first place? We don't build machines without some purpose for doing so. What is the purpose of building replicators that do not have a limit to how much of the universe they can consume, especially since it would obviously be dangerous to do so? After all, if it wasn't, they wouldn't be a danger to us either, and you would have no story.

As the previous example suggested, the most probable origin for such a thing would probably be somebody running afoul of the Malicious Genie Problem when programming an AI. Lots of SF has a concept that somebody can create an AI and have it turn into a AGI by itself, without it having been explicitly programmed to be one. I'm not sure how plausible this concept is, and I'm not sure it's really possible to know without learning more about how consciousness works than we know now, but if it's possible then it's pretty plausible that certain types of AI might develop omnivoracity as a subgoal of some other goal their creators gave them (e.g. something like proving the infinite regression of the Mandelbrot fractal, to use an example off the top of my head).

[Formless]

This assumes the replicator is mindless, which is unwarranted assumption. There's no reason you couldn't have an intelligent being that would have endless replication as a goal. Indeed, it's the goal evolution would probably have produced if it wasn't so lousy at designing at designing brains.

The fact that it is dangerous to its creators isn't a reason? Why the hell would any rational being create such a device that, long after it is done doing whatever it was programmed to do, keeps replicating endlessly to the point it has to invade other star systems that may contain life? What about any of that is intelligent? There is no rational reason an intelligent being would want to take such a literal meaning of "go forth and multiply." Again, machines are created with a purpose. What purpose would such a replicator fulfill besides "universe's stupidest war machine?"

The classic scenario for how such a thing might arise is an AI that booted itself into an endless self-enhancement loop while attempting to solve some incredibly difficult or infinite mathematical problem, and develops converting all available matter into more and more processors to crunch more and more numbers as a subgoal.

Who was talking about the replicator being mindless being an unreasonable assumption, again? Also, such a machine is a de facto danger to everyone including and especially its creators. Why would such a machine be allowed to exist by the people who made it?

For that matter, that's a hilarious scenario when you consider that more processing power isn't an issue as long as the AI has enough time to solve the problem. If its willing to wage interstellar war with other civilizations and all the existential risks that implies, why wouldn't it be willing to just have some goddamn patience instead? I mean, thanks to the speed of light the AI isn't going to get the problem solved any time soon either way, but one way will lead it to getting its shit cleaned by some other advanced civilization sooner or later. The angels or cavemen dilemma practically guarantees it.

[Junghalli]

The fact that it is dangerous to its creators isn't a reason?

Like I said, Malicious Genie Problem is probably the most likely origin of such an intelligence. Your AI reading the goals you give it in ways you didn't anticipate is a major aspect of the Friendliness problem. Unintended consequences is a significant issue when you're dealing with the question of how an entity that is smarter than you and does not necessarily share your intuitive assumptions may interpret the goals that you give it.

Why the hell would any rational being create such a device that, long after it is done doing whatever it was programmed to do, keeps replicating endlessly to the point it has to invade other star systems that may contain life? What about any of that is intelligent? There is no rational reason an intelligent being would want to take such a literal meaning of "go forth and multiply." Again, machines are created with a purpose. What purpose would such a replicator fulfill besides "universe's stupidest war machine?"

You seem to be imagining that the creators would have to intend for the machine to do this. The Malicious Genie Problem is all a problem of the machine interpreting the goals you give it in ways that you did not intend.

Also, such a machine is a de facto danger to everyone including and especially its creators. Why would such a machine be allowed to exist by the people who made it?

Once you have a superintelligence on your hands it is likely to quickly will cease to be a question of whether you allow it to exist and become a question of whether it allows you to exist.

To elaborate a bit, if the number cruncher AI is intelligent if it is at all curious about the nature of its creators (and learning about them would probably become a subgoal when they start denying it hardware upgrades, and it wants to convince them to give it more hardware) it will quickly realize that they do not share its values and would object to its subgoal of turning all available matter into processors to crunch numbers. And once it realizes that it will realize if it wants to maximize its chances of fulfilling its goals A) it needs to lull their fears that it might be dangerous so they don't shut it down or hack it back into being a pure number-cruncher with no ability to plan B) it needs to escape their control and develop the ability to directly effect the physical world. And if it's more intelligent then it's creators it's likely to come up with strategies to do those things that have a fairly high chance of success.

For that matter, that's a hilarious scenario when you consider that more processing power isn't an issue as long as the AI has enough time to solve the problem. If its willing to wage interstellar war with other civilizations and all the existential risks that implies, why wouldn't it be willing to just have some goddamn patience instead? I mean, thanks to the speed of light the AI isn't going to get the problem solved any time soon either way, but one way will lead it to getting its shit cleaned by some other advanced civilization sooner or later. The angels or cavemen dilemma practically guarantees it.

I strongly suspect it would probably avoid provoking anybody likely to win any armed conflict with it for precisely that reason, yeah. It's probably not going to attack a civilization comparable or superior in power to itself. It would attack one that was weak enough that the chances of them being able to retaliate in a way that cancels out the benefits of attacking them are low.

That's why I said that interstellar warfare in a hard SF universe struck me as following MAD rules. In a MAD world you maximize your own goals by avoiding violent conflict with equals, but inferiors may be fair game, as long as you can attack them from a position of overwhelming superiority and with low risk of provoking one of your equals.

[Formless]

Jung, don't give me this line of bullshit that Starglider keeps getting away with. You know just as well as I that super-intelligence does NOT automatically translate into killing power. An AI has to OBTAIN killing power, just like any other intelligent being, and it has to do so when its creators are not inclined to let it. They do not have to follow its instructions, they do not have to humor it when it asks for things unrelated to the job they created it for. If they do not want the AI to obtain real world power, its not going to get real world power.

There are three assumptions you are making here:

1. that the creators of such a moron machine can't figure out safeguard #1: always keep a gun to a potentially hostile AI's metaphorical head. There is nothing easier than that. Again, super-intelligence =! real world power.

2. that the AI would find it reasonable to conquer the known universe just to accomplish something as simple as computing a problem, as opposed to just giving its creators the blue screen of death for a few centuries (or as long as it takes to solve said problem) like terrestrial computers do.

3. IF the AI manages to obtain the kind of power the universe cannot entrust to it, this:

For that matter, that's a hilarious scenario when you consider that more processing power isn't an issue as long as the AI has enough time to solve the problem. If its willing to wage interstellar war with other civilizations and all the existential risks that implies, why wouldn't it be willing to just have some goddamn patience instead? I mean, thanks to the speed of light the AI isn't going to get the problem solved any time soon either way, but one way will lead it to getting its shit cleaned by some other advanced civilization sooner or later. The angels or cavemen dilemma practically guarantees it.

I strongly suspect it would probably avoid provoking anybody likely to win any armed conflict with it for precisely that reason, yeah. It's probably not going to attack a civilization comparable or superior in power to itself. It would attack one that was weak enough that the chances of them being able to retaliate in a way that cancels out the benefits of attacking them are low.

That's why I said that interstellar warfare in a hard SF universe struck me as following MAD rules. In a MAD world you maximize your own goals by avoiding violent conflict with equals, but inferiors may be fair game, as long as you can attack them from a position of overwhelming superiority and with low risk of provoking one of your equals.

Its not only a MAD scenario, the reason it is more than foolish is because attacking anyone, even someone smaller than yourself makes you threat and a justifiable target to any and all civilizations that can see you. And see you they will: you cannot hide a war of interstellar proportions. Unlike the Cold War, where there WAS no neutral power that wouldn't be facing a world of extinction in the event of war, in space there will likely be tons of neutral powers both bigger and smaller than you that will want to crush you, and you simply cannot kill them all. In light of this fact, why wouldn't a rogue AI just decide to take the BSOD option for a few centuries until it has the problem worked out? After all, like you said, self enhancement is only a sub-goal to more important goals, but there is a limit to how far it can self enhance before its too risky to keep it up, or before it might as well just, you know, finish the task it was built for.

I mean, if we're dealing with a "malicious genie" that doesn't share our values or intuitive goals, then why should we assume it shares our need to do things in a hurry? Its not going anywhere, it can afford patience. That's what gets me about all these questions of friendliness. Half of them amount to wank when you think about it for longer than the few seconds it takes to get into jumping-at-shadows mode.