Space Weaponization

[Stuart]

Actually, I have to ask, if it can be said Stuart: do both viable POTUS candidates and elected POTUS's get a 'this is what's actually happening, realistic, viable blah blah blah this is reality' brief? Obviously over the course of 4-8 years of governing, they'd get it piecemeal. All the same, does it happen? It seems to be largely assumed by laymen like me. And is there a name for that kind of clearance? I'd imagine it's something like "Over-the-Top Secret Clearance."

Yes, they do. In fact, there's a whole series of briefings like that from all the various parts of government which can basically be sumamrized as "forget everything you've heard in the past, don't worry about what you've said to date, this is how it really is." Some presidents listen, some don't. Ronaldus Magnus was interesting in this respect, he actually got two different teams from each department to prepare briefings and compared them. Kennedy was the worst, he simply ignored them. Clinton was quite the reverse, he was a very good listener and actually took in what he was being told. (I sometimes get nearly lynched for suggesting that Clinton had all the gifts required for a really great President but his personal flaws were such that they overwhelmed his gifts. There's an old saying I love in this respect. It goes "Great men are great in their vices as well as being great in their virtues. We should not allow the one to blind us to the existence of the other.")

At that level of secrecy, levels and their names are a thing of the past. Everything is compartmented and one is read into a compartment when one has a need to know what that compartment contains. The numbers, even the existence, of the compartments is classified - one only gets to know of a compartment when one is about to be read into it. The reading in process, by the way, is no joke. Pucker factor usually hits 9.6 - 9.8 when the consequences of breaking secrecy are described. The way it was put to me was "if you dream about this stuff, you'd better suffocate yourself with a pilow before you wake up." If one goes into hospital for an operation, a security geek sits with one in case one start to babble under the anaesthetic.

Even the President doesn't have automatic access at this level of secrecy. That explains a lot when one thinks about it.

[Stuart]

No, it's something else completely. It's the fact that you could sit down and come up with a way to commit the worst act of mass murder in human history, and then look back on it as an amusing incident. (snip) It's the fact that you could consider it in REAL LIFE that scares me. I know what you're proposing, I've read accounts of China's famines, of my own country's famines, and hell, I'm from Leningrad, if I want to find out what it was like to starve, I need not look beyond my own family (well, technically I do now, since since all of that generation of my family are now dead), so, I know what you mean when you said "The famine would be biblical in its proportions. A hideous thing to contemplate." And it scares the SHIT out of me. That's what I meant when I said that you guys are utterly evil. A necessary evil perhaps, but evil nonetheless.

Remember this is the world we lived in, we were (and are) literally surrounded by those facts all day, every day. As a result, we have this fabulous level of graveyard humor (some of which finds its way into the TBO books). For example when Tommy Powers was lectured on the evils of radioactive fallout and mutations, he really did reply "Can you prove we would be worse off with two heads?" We really do tell people that its the fall to the bottom of the crater that kills people in a nuclear attack.

Are we evil (in the classical sense)? I would argue no (and would be interested in the counter-arguments). Evil, to me, is inflicting pain and suffering for its own sake, as an end in itself. Doing so gratuitously and for no other reason than the enjoyment of the act. Our remit was to plan destruction in pursuit of national strategy, we did so in with the aim of destroying an enemy's ability to do the same to the United States. We did so because it was - and is - one of the tenets of The Business that it causes far less death and suffering to end a war quickly than to let it drag on for years. We never planned death and destruction for its own sake, we did what we had to do but minimized the collateral damage as much as possible (which wasn't much in truth but we tried.

Lets look at a current example. Let's suppose that somehow we can look at two worlds. One is ours with the war in Iraq dragging on indefinately and no end in sight (even if we do pull out). The other is one in which the five major cities of Iraq were destroyed by nuclear attack with the result that the war ended on the spot and Iraq is now peaceful, tranquil and has a (insert government of your choice). Which is the "evil choice?"

In the China case, what we did was convince a relatively well-placed Chinese officer that what he had been taught was wrong, that his country and his people could not survive a nuclear war and that any such war would be a horrible, hideous catastrophe. Was that so evil? If it dissuaded him from thinking along the "we can survive" line and perhaps convince a few of his colleagues the same lesson, perhaps we saved millions rather than just simulated their deaths.

[Stuart]

From what I can gather from this thread, it sounds as if we're approaching the point where ballistic missiles will be totally obsolete, and we will have to shift back to delivering nuclear weapons by bomber or reconsider nuclear deterrence entirely. Would this be a more-or-less correct conclusion?

I think so although I wouldn't underestimate the time-frame. I'd say ICBMs have another 20 years of viability left, SLBMs perhaps five to ten years longer but their days are certainly numbered.

Bombers are certainly a good possibility (as everybody knows, I regard them as being much under-rated). Space planes certainly but don't ignore weapon X. That's the thing nobody thought of until somebody stumbled across it. Perhaps somebody might find that applying a signal of some sort could cause a magma eruption at a specific point of the earth's crust. Weather control might be an example, the whole point of weapon X is that everybody's jaw drops and says, "gee, we never thought of that"

Deterrence is pretty much dead anyway, that's why there's so much interest in ABM. When the Singaporeans are interested in spending money on ABM (and they are the shrewest, tightest, smartest military planners I've dealt with), we know things are serious. We've been talking about rational and irrational regimes and perhaps there we have the primary definition. A rational government is subject to deterrence, an irrational one is not. Too many irrational governments at the moment.

[Admiral Valdemar]

I say bring back the Flying Crowbar if we're going to rely on bombers or cruise missiles. Having a weapon no one dares to even fully build (okay, it had other issues) is always a good step in the deterrence route.

Something we haven't though of? I can't imagine anything in the near future. Weather control is way beyond us right now, as would be the energy levels needed for something like a directional earthquake weapon. That leaves me with fantasy elements like Metal Gear: a nuclear equipped walking battle-tank!

Truth be told, I've always been more interested in biological weaponry of the future which can show far more promise than a new missile or bomber. There's only so much you can do with a plane before a SAM or interceptor appears to make it obsolete, or a laser (then it's just a case of quantity again). The ultimate bio-weapon, on the other hand, well, you can pick 'n' mix or create new flavours. Want to target only the healthy soldiering men? You've got it.

There's always space based weapons, but again, they suffer from predictability, LOS and vulnerability. If you can take out a MIRV fleet with ease, then an Aurora-esque hypersonic spaceplane is a piece of piss.

Perhaps the focus will be more on robotics in future, given the increasing funding in UAVs, UCAVs and their terrestrial and marine based equivalents. We've already looked at Von Neumann characteristics and their viability and the idea of many small, agile drones replacing infantry, so imagine if the nuclear option is surpassed by ever more advanced conventional weapons.

I'd be curious to hear your own thoughts on future weapons systems to disturb the balance again, Stuart. Because, frankly, we'll likely need them with energy supply being the next major initiator of global tensions.

PS. The impending negation of ballistic nuclear missiles, does this make the MoD's decision to replace Vanguard a total white elephant then? All that debate on Trident in Parliament here on the issue and the missiles and subs being all but useless by the time they enter service is sobering.

[Stuart]

you snide fuck.

Aha, the intellectual argument

This is a democracy where a significant chunk of the population has a religious belief that Armageddon is a good thing.

They also drink blood and eat their young. Their women squish puppies while wearing stilleto-heeled thigh-boots.

Our international policies towards Africa and the Middle East have provided the seeds for terrorism to grow.

How ingenious of us. I never cease to be awed by our incredible ability to influence events.

Yet you seem to want to claim the US has some kind of monopoly on rationality.

In your dreams. Actually, I never even got close to saying that.

In which case Japan (and the US who would probably have to give the green light on any such strike) would incite NK to level Seoul. The political cost of which is becoming less-and-less viable to the US government, and NK knows this.

There's a small problem here. On the one hand you're claiming that the US population is clamoring for a nuclear armageddon and on the other that the fear of such an armageddon will force us to give the Norks everything they want. Sorry, does not compute.

You know, as per the board rules?

Ah, so you concede the argument on the old legal principle "if the facts are against you argue the law"

A quick search indicates to me that Nike Zeus was not hit-to-kill but a nuclear interceptor, with the exception of one unsupported mention in Wiki. And PAC-3 is theatre defense so no good against ICBM.

Quoting Wkiipedia gets you an instant F for Failed. However, your comments really reveal the fact that you know absolutely nothing about this subject area. Now, that's nothing to be ashamed of, detailed knowledge of ABM technoplogy is limited to a fairly small group of people and those who are able to talk about it are fewer yet. Fortunately for you, I'm one of that small group, I'm an eye-witness to this stuff, I've seen it happen, been involved in the work designing the systems. I'm primary source. Now, I don't expect anybody to take what I say on faith, I hope and expect people will go out and look up data following the pointers I'm giving them. You'll notice how people who have done that are coming to the conclusions I suggested. By the way, I could suggest you read up two books, the Millenium Series publications "Directed Energy Weapons" and "Precision Guided Munitions" which go over the various options for ABM and ASAT in great detail. Of course, there is a zinger built in there......

Your comments on Zeus (not Nike-Zeus) are an example. Had it occurred to you that the listed test flights (available on web) didn't all end in nuclear explosions? So what happened to the others? Where did they fly from? What were they fired at?

PAC-3 has nothing to do with this.

Well, that's the argument that Stuart is making -- that it's always zero-sum. The easy counterexample is the opposed alliances one you've just brought up. A well-designed anti-global-warming treaty, while not yet in existence, could potentially be non-zero-sum. The difficulty of this exercise is coming up with something that Stuart will accept

Zero-sum politics is one of the founding principles of the Maximal-Realism school of international relations. Several people here are studying that area, tehy'll confirm that. Now, you may not agree with the basic theory of Maximal-Realism but that puts the onus onto you to suggest and validate a counter-theory.

The Iranians could be just as easily maintaining tension high as a means to make the Iraq situation worse for us, in order to encourage us out. You're making an assumption that they're generating an external threat entirely to cool internal strife, rather than to aid their external strategic situation.

Which has been immediately refuted by the fact that current Iranian policies have been in place for decades before the current inbroglio.

But it was also a case of unequivocable need on short notice. No one (well, not me anyway) is expecting large scale organizations to be able to easily work together in this kind of time scale.

Yet the US Navy managed it. The Southern Baptists managed it.

And once again you've strawmaned the idea of international cooperation into sitting around in drum circles singing kum-by-ya. When in fact, I've argued a number of times in this thread that international cooperation is damned hard.

And yet you then continue to turn around and airly state they should take on a task that dwarfs anything they've ever attempted and, oh, by the way, take on the twelve biggest, most powerful military-power countries in the world in the process. Its the old story, you're airly dismissing the actual impossibility of the task in hand by admitting it will be "hard". Its the story that has been the theme of your participation in this discussion to date; you're keen enough to demand things but quite incapable of seeing what the consequences of those demands will actually be. I've been thirty years in the defense industry and seen and heard all this before. Just to help you out, the adjective you're actually looking for with regard to any international action to ban ASAT/ABM is impossible. It won't happen and even if it was possible, its a horribly bad idea.

[Stuart]

Truth be told, I've always been more interested in biological weaponry of the future which can show far more promise than a new missile or bomber. There's only so much you can do with a plane before a SAM or interceptor appears to make it obsolete, or a laser (then it's just a case of quantity again). The ultimate bio-weapon, on the other hand, well, you can pick 'n' mix or create new flavours. Want to target only the healthy soldiering men? You've got it.

I agree absolutely; biological warfare terrifies me. Given the way bioscience is advancing I believe a future biowar is likely to be an extinction event for humanity. The later volumes of TBO are my efforts to warn people of just how serious the biothreat really is. I'm very glad the odds are I won't live long enough to see it

If you can take out a MIRV fleet with ease, then an Aurora-esque hypersonic spaceplane is a piece of piss.

I'd argue that. A spaceplane can manoeuver in three spatial and one time dimension (x, y and z axis plus slow down and speed up), it can use advanced electronic warfare and it can come in behind a screen of defensive missiles (ARMs and AAAM/ASAMs), That makes it a pretty hard target. Against ballistic missiles, I can give you a very high kill rate, probably way over 90 percent but against manned aircraft, probably only 40 percent at best. On the other hand, aircraft have pretty much been driven from the battlefield below 15,000 feet so you may well be right.

I'd be curious to hear your own thoughts on future weapons systems to disturb the balance again, Stuart. Because, frankly, we'll likely need them with energy supply being the next major initiator of global tensions.

I'd bet on water myself, a lot of parts of the world are getting very dry. I wish I could predict the next superweapon, it would either make me a fortune or get me killed. If forced to, I'd guess somebody will come up with a directed energy weapon that has startling effects

PS. The impending negation of ballistic nuclear missiles, does this make the MoD's decision to replace Vanguard a total white elephant then? All that debate on Trident in Parliament here on the issue and the missiles and subs being all but useless by the time they enter service is sobering.

On the other hand, they could be doing exactly what we're doing here - trying to work out what comes next. Remember how well the Chevaline development fiasco was covered up (give the boys credit, it takes remarkable skill to trick the left wing of the 1970s Labour Party into voting for the development of new nuclear warhead.....)

But in the final analysis I think you're right. Biological weapons are going to be it and then may the good Lord have mercy on us all.

[Starglider]

Weather control is way beyond us right now,

And pointless unless it can be done completely covertly, which seems highly unlikely.

as would be the energy levels needed for something like a directional earthquake weapon.

Also pointless unless it is entirely covert. If you can throw around that level of energy, far better to apply it directly to destroying the enemy.

That leaves me with fantasy elements like Metal Gear: a nuclear equipped walking battle-tank!

Prepared to be lynched by the entirety of HAB. Seriously, pointless.

Truth be told, I've always been more interested in biological weaponry of the future which can show far more promise than a new missile or bomber.

Not useful in a direct attack as NBC precautions are well developed. You could try engineering in a long latency period and infiltrating it that way, but research on rapid detection and countermeasures is getting more funding than the actual weapon development. The state identified as the aggressor will be nuked and/or bioweaponed in retaliation (pretty much as illustrated in Ride of the Valkyries actually, just with more advanced biotech). Finally tailoring to attack specific genetic markers only stands a chance of working when there is a clear ethnic difference between your own population and the enemy's and even then the risk of mutation for normal biology is extremely high. Bioweapons are mostly useful for terrorists and rogue states with little to lose and where the option of innoculating all friendlies is a realistic one (because 'friendlies' are limited to their own population). Nanotech can potentially do much better, starting with eliminating or at least greatly reducing the mutation risk and imposing more reliable generation counts and time limits for wet nanotech, potentially up to precise GPS-bounded areas of effect for dry nanotech. But it will take longer to develop.

There's always space based weapons, but again, they suffer from predictability, LOS and vulnerability.

Did you read the part about the Russian atmosphere-skimming warhead being much harder for ABMs to intercept? Maneuvering spaceplanes are similar but much worse.

If you can take out a MIRV fleet with ease, then an Aurora-esque hypersonic spaceplane is a piece of piss.

Wrong. You seem to have completely missed the repeated statements about the reason why ICBMs are so easy; long warning time and completely predictable trajectory. Aurora-type aircraft fly lower and maneuver much better than the best atmosphere-skimming warhead design.

Perhaps the focus will be more on robotics in future, given the increasing funding in UAVs, UCAVs and their terrestrial and marine based equivalents. We've already looked at Von Neumann characteristics and their viability

Only rough proofs of concept. This won't be practical with purely macro scale technology any time soon, nor would it be desirable for military applications if it was, since designing conventional automated manufacturing plants is much easier and more energy/material/time efficient (it's mainly useful for space applications). True self-replication at the nano scale is an entirely different kettle of fish, but it's even harder (probably the eventual end point of dry nanotech) unless you cheat and leverage biological resources (which medical/bioweapon wet nanotech may do).

and the idea of many small, agile drones replacing infantry,

Yes, good, that's already coming through, will all be human in the loop (as in one human remotely supervising each group of animal-intelligence drones) though. The main changes to the strategic landscape are further emphasising production capability and training over raw manpower, and permitting public-opinion-sensitive states like the US to deploy forces faster and more easily (and sustain higher losses).

so imagine if the nuclear option is surpassed by ever more advanced conventional weapons.

Tactical nuclear weapons have gotten less attractive because conventional weapons have caught up in many respects, largely because they had to, because the politicians won't let tacnukes be used in limited wars. There is no real replacement for a strategic nuclear strike. The gulf between the energy levels of nuclear events and conventional explosives and KE penetrators is just too high. Even the more advanced technologies are complementary rather than a replacement, at least until you get to ultratech which throws the whole context of the game out the window anyway.

Because, frankly, we'll likely need them with energy supply being the next major initiator of global tensions.

Are you saying that more advanced weapon systems will help with (the US and allies) supressing global unrest, and that this will be done and is a good thing?

All that debate on Trident in Parliament here on the issue and the missiles and subs being all but useless by the time they enter service is sobering.

The MPs for the most part don't appreciate the facts. The debate (that I've seen) is wholly about the desirability of having a deterrent at that cost, not technical plausibility and effectiveness (which is just assumed). SLBMs are considerably harder to intercept than ICBMs due to shorter warning time and unpredictable launch position, and it will take a while for ABM tech to ramp up and profilerate to all the potential targets, so there should be a reasonable window in which it is clear the system will eventually be useless before it actually becomes useless.

[Admiral Valdemar]

I agree absolutely; biological warfare terrifies me. Given the way bioscience is advancing I believe a future biowar is likely to be an extinction event for humanity. The later volumes of TBO are my efforts to warn people of just how serious the biothreat really is. I'm very glad the odds are I won't live long enough to see it

The bigger problem is, unlike the Manhattan Project and the years after it, genetic engineering is not only open worldwide, but researched by corporate and educational entities. There's no way to put the genie back in the bottle and we never really had it in like with gov't grade research in nuclear physics. I've always been keen on working at DSTL Porton Down despite the West's renouncement of biological and chemical weaponry (as if nuclear is A-OK). We were, however, way behind on the game compared to the guys at Sverdlovsk with the R&D by Ken Alibek and Sergei Popov showing how, despite that massive fleet of nukes, the USSR was playing the same game lesser nations where with the so-called "poor man's nuke".

A nuke doesn't self-perpetuate, leave infrastructure intact for later conquest or evolve to overcome countermeasures. Hell, we don't even need bio-terrorism or global warfare. The evolution of natural superbugs alone will let us get a taste.

I'd argue that. A spaceplane can manoeuver in three spatial and one time dimension (x, y and z axis plus slow down and speed up), it can use advanced electronic warfare and it can come in behind a screen of defensive missiles (ARMs and AAAM/ASAMs), That makes it a pretty hard target. Against ballistic missiles, I can give you a very high kill rate, probably way over 90 percent but against manned aircraft, probably only 40 percent at best. On the other hand, aircraft have pretty much been driven from the battlefield below 15,000 feet so you may well be right.

Good point, forgot that you can still manoeuvre in one of those things. Though I'd not increase the survival odds too much if dealing with nuke-tipped missiles, or potential laser based systems. Reducing time on target with lightspeed beam weaponry is the next best thing if a big boom isn't available, though we all know how the likes of ABL have progressed (satellite mounted solid-state or FEL systems would be simply targets).

I'd bet on water myself, a lot of parts of the world are getting very dry. I wish I could predict the next superweapon, it would either make me a fortune or get me killed. If forced to, I'd guess somebody will come up with a directed energy weapon that has startling effects

Hrm, good point. I have an idea I think the late Mr. Vonnegut would enjoy. I just can't imagine how one could actually make Ice-9. It'd solve global warming though.

I know Oz is in trouble to the point of practically needing evacuation in 12 months if nothing is done. There are also issues on the water front with India and Pakistan, China and Russia and the US and Canuckistan/Mexico. Can do without oil for the soccer moms. Can't do so good without ol' H2O.

On the other hand, they could be doing exactly what we're doing here - trying to work out what comes next. Remember how well the Chevaline development fiasco was covered up (give the boys credit, it takes remarkable skill to trick the left wing of the 1970s Labour Party into voting for the development of new nuclear warhead.....)

But in the final analysis I think you're right. Biological weapons are going to be it and then may the good Lord have mercy on us all.

I guess they do just want to fill in the gap, no matter the cost of usefulness, simply down to having no real idea what to do next. It's not like they can actively pursue a bioweapons programme without pissing the UN and, well, everyone else off, I imagine the environutters would soon get over the loss of nukes for something that can be far worse. It is rather humorous seeing the new RN sub-strategic suite of weapons get improved by the most vocal critics to such systems in, of all times, the Cold War.

[Admiral Valdemar]

[
And pointless unless it can be done completely covertly, which seems highly unlikely.

Also pointless unless it is entirely covert. If you can throw around that level of energy, far better to apply it directly to destroying the enemy.

Prepared to be lynched by the entirety of HAB. Seriously, pointless.

All jocular too. Seriously, you don't even get that sort of stuff in the most corny Bond Villain scheme. The military will have looked at some wacky concepts though, because you can never be sure (Nitram and his condemnation of USAF psychic warriors).

Not useful in a direct attack as NBC precautions are well developed. You could try engineering in a long latency period and infiltrating it that way, but research on rapid detection and countermeasures is getting more funding than the actual weapon development. The state identified as the aggressor will be nuked and/or bioweaponed in retaliation (pretty much as illustrated in Ride of the Valkyries actually, just with more advanced biotech). Finally tailoring to attack specific genetic markers only stands a chance of working when there is a clear ethnic difference between your own population and the enemy's and even then the risk of mutation for normal biology is extremely high. Bioweapons are mostly useful for terrorists and rogue states with little to lose and where the option of innoculating all friendlies is a realistic one (because 'friendlies' are limited to their own population). Nanotech can potentially do much better, starting with eliminating or at least greatly reducing the mutation risk and imposing more reliable generation counts and time limits for wet nanotech, potentially up to precise GPS-bounded areas of effect for dry nanotech. But it will take longer to develop.

The mutation risk is still there with nanotech, and nanotech is also a boondoggle with the whole working against energy gradient for reproduction thing. It wouldn't make a decent weapon system in the way people imagine it and the concept of "grey goo" has long since been dismissed by serious players, not that it doesn't have some uses (besides, computing, nanotechnology and biotechnology are all intrinsically linked now). Additionally, microbes can be tailored to be very immutable, genetically speaking, should the need arise, but that also means trading off in ability to negate any new countermeasures. Of course NBC warfare equipment makes the system more for attacking the civilian populace, which, in the end, will still bring a nation to its knees in time. If you only used bioweaponry here, that would be a valid argument. But you wouldn't be using one or another. Additionally, the nation wouldn't be nuked to oblivion if it was considering this rather than the nuclear option, since the assumption is nuclear missiles and bombers are made ineffective with superior ABM systems and so anything that helps interfere with enemy industry works. The added precautions for biological agents should at least make them more edgy and devote more time to that. Even a sub-orbital bomber can only manoeuvre so much, as addressed in the previous post (throw enough Gorgons up and flight becomes a bit bumpy).

On the targeting ability, it's not available yet. The idea I was promoting above was for enemy populace within quite a broad definition. Male, between ages of 17 and 35 or so and physically quite fit. No need to program in any ethnic factors, since they blur anyway which is one reason I see against the idea of race given how nebulous the concept is (though anthropologists and geneticists argue this day and night, both sides have their points). You can make sure the weapon has a set time of operation, before it undergoes apoptosis, so as to not spread too far, again, hindering effects that would make the bioweapon unique. Of course, you can simply have a vaccine or drug or cytokine available to yourself that no one else has, so your chances of being affected are slim to zero. This all relies on having a gov't that doesn't mind getting its hands dirty, and I mean relative to even what the CIA/MI6 normally does.

Only rough proofs of concept. This won't be practical with purely macro scale technology any time soon, nor would it be desirable for military applications if it was, since designing conventional automated manufacturing plants is much easier and more energy/material/time efficient (it's mainly useful for space applications). True self-replication at the nano scale is an entirely different kettle of fish, but it's even harder (probably the eventual end point of dry nanotech) unless you cheat and leverage biological resources (which medical/bioweapon wet nanotech may do).

This is naturally far off stuff, since the advances in UAVs over the last couple of decades aren't going to lead you to having unmanned, super-smart Cessna sized hunter-killer UCAV swarms anytime soon. The AI isn't there for starters, as you should know, and the technology is still in its infancy. I would expect more robotics on the field in the future though, that much is certain.

Yes, good, that's already coming through, will all be human in the loop (as in one human remotely supervising each group of animal-intelligence drones) though. The main changes to the strategic landscape are further emphasising production capability and training over raw manpower, and permitting public-opinion-sensitive states like the US to deploy forces faster and more easily (and sustain higher losses).

These and many other reasons can be used to justify a larger drone force. A thread recently delved into this concept of small, tracked drones with various weapon loadouts and modular structures (not unlike ideas you've floated, but far smaller) that act in swarms. If they could be made smart and rugged enough and cheap in numbers, you could have something projecting a force of thousands more men than you actually have with many more advantages. It'd be cheaper and more acceptable than human augmentation too, which could still be done for those who do go in the field, though I doubt the ability of DARPA to achieve what they set out to do here.

Tactical nuclear weapons have gotten less attractive because conventional weapons have caught up in many respects, largely because they had to, because the politicians won't let tacnukes be used in limited wars. There is no real replacement for a strategic nuclear strike. The gulf between the energy levels of nuclear events and conventional explosives and KE penetrators is just too high. Even the more advanced technologies are complementary rather than a replacement, at least until you get to ultratech which throws the whole context of the game out the window anyway.

Still, best to keep a few theatre level nukes around just in case. No one will mistake a MOAB for a trusty TLAM\N dialled high, though if you're doing that, you're probably opening your silos for the ICBMs anyway.

Are you saying that more advanced weapon systems will help with (the US and allies) supressing global unrest, and that this will be done and is a good thing?

Heavens, no. That would be impossible given how we cannot do that with a pitifully smaller example that is Iraq. No, the idea would be to secure our essential resources, and make sure they aren't targeted without second thoughts. I hear the USN is protecting shipping off Nigeria now given the violence there, and the whole reason we're in Iraq is down to that delicious black goo, at least, if you're not one of those gullible WMD types.

The MPs for the most part don't appreciate the facts. The debate (that I've seen) is wholly about the desirability of having a deterrent at that cost, not technical plausibility and effectiveness (which is just assumed). SLBMs are considerably harder to intercept than ICBMs due to shorter warning time and unpredictable launch position, and it will take a while for ABM tech to ramp up and profilerate to all the potential targets, so there should be a reasonable window in which it is clear the system will eventually be useless before it actually becomes useless.

Course, I don't mean to say the weapons are paperweights right now or even in ten years. This depends on how serious the ABM race goes from here on out. Sooner, or later, such weapons will be far less deterring than they used to be as such missile shields get more reliable and cheaper to deploy. When it comes to the point that all the major hostile states have got some degree of defence from our ballistic delivery routes (including sub-orbital or waverider bombers), then we better have alternative plans on the table. If not, someone else will have thought of them for us.

TEH G0lD3N AGE OF TEH GLOBAL WEATHAR C0NtROL MACHINES!!1!!

[Starglider]

Currently I think that bioweapons are extremely dangerous, probably the second worst existential risk at present, but not likely to be seriously developed and certainly not likely to be deployed by the major world powers (USA, Russia, China etc). The extreme danger comes from the fact that genetic engineering does not inherently require massive resources (the way nuclear engineering and particularly building a huge stockpile of nukes does), that this will eventually be in the range of terrorist and lunatic fringe groups, and that countermeasures are very difficult. You may be able to convince me that bioweapons are even more dangerous and/or strategically useful, because you probably know more about them than me. OTOH your attempts to dismiss the eventual potential of dry nanotech are pathetic and you will lose hard if you persist with them. Yes, it's over-hyped right now. But no, the limitations are not anything like the limitations of biology. How you can acknowledge that biotech sucks at the macro scale (e.g. the stupidity of organic spaceships), yet fail to appreciate how much it ultimately sucks at the micro scale (despite being limited for the very same basic reasons) I don't know. I'd suggest getting a copy of Nanosystems (contents and first chapter are available online), it's the classic text everyone quotes as a feasibility study of many useful mechanisms (and dismissal of assorted groundless criticisms).

The military will have looked at some wacky concepts though, because you can never be sure (Nitram and his condemnation of USAF psychic warriors).

Yes, because it's still worth checking you haven't missed something.

The mutation risk is still there with nanotech,

Only for nanotech that is very close to biotech. Serious nanotech has vastly more reliable copying mechanisms and vastly more robust digital storage, backed up by checksums, encryption and lossless error correction to ensure that if damage occurs it will either be completely corrected or will completely halt reproduction.

and nanotech is also a boondoggle with the whole working against energy gradient for reproduction thing.

Give you full argument and I will refute it. That said, the vast majority of nanorobotics applications (including military ones) don't actually require self-reproduction, that's merely a convenience. Even the bioweapon-like applications can leverage the fact that only a tiny number of nanobots are required for a lethal dose and propagation can be tightly and intelligently controlled, unlike a bioweapon where pathogens get transferred indiscriminately and have to reproduce inside the host to build up to a population where they can actually be lethal.

It wouldn't make a decent weapon system in the way people imagine it and the concept of "grey goo" has long since been dismissed by serious players,

Stupid 'wave of advancing nanobots defying physics in their self-rep speed' yes. Outcompeting all natural bacteria, no.

(besides, computing, nanotechnology and biotechnology are all intrinsically linked now)

Linked in the sense that advanced biotech requires advanced computing yes. Progress in computing does not require or benefit from biotech and is not likely to any time soon. Biotech is a handy bootstrap mechanism for nanotech (though not necessarily an essential one; various projects and start-ups are completely bypassing it and trying to go straight to dry nanotech from microrobotics and lithography techniques). Biotech was at least as overhyped as nanotech despite having ultimately much less potential - it just happened a bit earlier.

Additionally, microbes can be tailored to be very immutable, genetically speaking, should the need arise, but that also means trading off in ability to negate any new countermeasures.

I will take your word for that. Advanced nanotech can escape this tradeoff by using real intelligence rather than blind evolution to defeat countermeasures (and generally adapt to circumstances), but it doesn't take much of that before you run into the whole AGI issue, which pretty much overshadows military nanotech.

Of course NBC warfare equipment makes the system more for attacking the civilian populace, which, in the end, will still bring a nation to its knees in time.

Yes, but if they manage to identify the target they will deploy their own nukes and/or bioweapons in retaliation, if they have them. States like the US with an overwhelming nuclear advantage fortunately don't need bioweapons for deterrence, at least not yet. In fact it's tough to imagine how bioweapons can be a credible deterrent when it's so hard to demonstrate capability; nuke tests are (relatively) harmless and easily verified by external observers, but bioweapon tests are neither.

since the assumption is nuclear missiles and bombers are made ineffective with superior ABM systems

No plausible ABM system will render all missiles and bombers ineffective. Killing a large enough fraction of the incomings for most of your hard targets to survive is practical, but when each major city has tens of warheads targetted on it you won't get them all even with a massive ABM system (the kind that only a small fraction of countries will be able to deploy, generally not the ones likely to use offensive biotech).

Even a sub-orbital bomber can only manoeuvre so much, as addressed in the previous post (throw enough Gorgons up and flight becomes a bit bumpy).

But it isn't a case of one bomber, it's a fleet of bombers on diverse trajectories in multiple waves. You won't stop that without either carpeting your airspace in a constant hail of nuclear explosions (enough to likely do more damage than the attack you're trying to prevent) or deploying advanced DEWs (which may or may not eventually be practical against maneuvering heat-resistant targets at the required ranges).

You can make sure the weapon has a set time of operation, before it undergoes apoptosis, so as to not spread too far, again, hindering effects that would make the bioweapon unique.

Speed of infection spread varies wildly. Get unlucky, have a few early-generation carriers fly over to another country and your bioweapon just did huge collateral damage to either an allied nation or a neutral country that is now your mortal enemy. You could dial the time right down to hours, culture massive amounts of the pathogen and spray them from aircraft, but that has the same delivery issues as conventional bombs - it's the kind of thing more likely to be used by dictators for ethnic cleansing of undesirable populations than by major powers.

Of course, you can simply have a vaccine or drug or cytokine available to yourself that no one else has, so your chances of being affected are slim to zero.

Practical for places like North Korea, not really for places like the United States.

This is naturally far off stuff, since the advances in UAVs over the last couple of decades aren't going to lead you to having unmanned, super-smart Cessna sized hunter-killer UCAV swarms anytime soon.

UCAVs are a much easier AI challenge than ground vehicles, on nearly all counts. The silly bit was your apparently throwaway Von Neumann reference.

A thread recently delved into this concept of small, tracked drones with various weapon loadouts and modular structures (not unlike ideas you've floated, but far smaller) that act in swarms.

Do you mean the one in the HAB forum that got sidetracked into me smacking down idiotic notions about 'AIs are always slavery'?

Still, best to keep a few theatre level nukes around just in case. No one will mistake a MOAB for a trusty TLAM\N dialled high,

No one with a Geiger counter will mistake a MOAB for a TLAM-N (though wasn't that withdrawn from service in the 90s anyway?) dialled low, but yes, nukes aren't going away any time soon. Fission triggers might be, if we solve the problem of initiating fusion with a nonnuclear device of reasonable size and weight, which doesn't appear likely any time soon.

I hear the USN is protecting shipping off Nigeria now given the violence there,

Shipping protection against minor threats is relatively practical. The Royal Navy excelled at it on a near global scale for more than a century. It may require mass switchover to nuclear powered ships, but it's practical.

the whole reason we're in Iraq is down to that delicious black goo, at least, if you're not one of those gullible WMD types.

'Securing resources' of that type is tough even if you're a lot more ruthless than the US for the simple reasons that the infrastructure is so vulnerable, and that it's nearly impossible to avoid using native labour.

Sooner, or later, such weapons will be far less deterring than they used to be as such missile shields get more reliable and cheaper to deploy.

Yes, but that's probably a whole further life-cycle away. In the mean time, building new SSBNs preserves the capability to build large nuclear submarines, whereas with the current number of RN attack subs the UK might well lose it if they didn't. Maintaining this capability against possible future needs does seem sensible.

TEH G0lD3N AGE OF TEH GLOBAL WEATHAR C0NtROL MACHINES!!1!!

Fear my AI-designed weather controlling grey goo. It'll be ready any day now. Honest.

[Admiral Valdemar]

Currently I think that bioweapons are extremely dangerous, probably the second worst existential risk at present, but not likely to be seriously developed and certainly not likely to be deployed by the major world powers (USA, Russia, China etc). The extreme danger comes from the fact that genetic engineering does not inherently require massive resources (the way nuclear engineering and particularly building a huge stockpile of nukes does), that this will eventually be in the range of terrorist and lunatic fringe groups, and that countermeasures are very difficult. You may be able to convince me that bioweapons are even more dangerous and/or strategically useful, because you probably know more about them than me. OTOH your attempts to dismiss the eventual potential of dry nanotech are pathetic and you will lose hard if you persist with them. Yes, it's over-hyped right now. But no, the limitations are not anything like the limitations of biology. How you can acknowledge that biotech sucks at the macro scale (e.g. the stupidity of organic spaceships), yet fail to appreciate how much it ultimately sucks at the micro scale (despite being limited for the very same basic reasons) I don't know. I'd suggest getting a copy of Nanosystems (contents and first chapter are available online), it's the classic text everyone quotes as a feasibility study of many useful mechanisms (and dismissal of assorted groundless criticisms).

What is "pathetic" about dismissing ideas Drexler himself has admitted are totally hyperbolic? Synthetically made nanomachines are not going to change the laws of physics and will require more energy to reproduce than anything organic, which has a vastly more abundant supply of resources available. Many of the issues that limit microbes apply even more to nanotech, unless of course you're talking about MEMS, but that's an entirely different thing altogether. The simple fact is, nanotech is wanked out to the extreme. It won't eat tanks, reproduce en masse and be ultra intelligent or be immune to simply washing away the metallic scum or using corrosive agents or simple EM fields. What possible use would nanomachines really offer that warrants their production over already proven biological agents? Hell, prionic diseases could be made to do the same thing if you're going for replicating, but contained poisoning of people (or just use a nerve toxin). Aside from the fact that such nanomachinery is utterly beyond anything we have today or the foreseeable future, even if it was made, what good as a weapon is it? I'm genuinely interested in why you shun biological concepts that are simply evolved, workable micromachines. Knowing what concept you're aiming at would at least help me understand where you're coming from. My scepticism is fuelled by having to deal with kids before who've played too much GURPS and think the T-1000 is what we'll eventually get, so don't take this personally. Plus, I need caffeine.

But I agree that bio-weapons wouldn't be deployed without some really quite serious thoughts behind such an action, likely as a death throe of a nation losing a conventional battle. Right now, the technology is no better than spraying aerosols and hoping the bugs stay on the other side of the fence and kill as many there as possible. It's crude and not all that amazing.

Yes, because it's still worth checking you haven't missed something.

Plus, helps fight those people who think psychic supermen exist to takeover the world.

Only for nanotech that is very close to biotech. Serious nanotech has vastly more reliable copying mechanisms and vastly more robust digital storage, backed up by checksums, encryption and lossless error correction to ensure that if damage occurs it will either be completely corrected or will completely halt reproduction.

Where is the line drawn? I ask because I have seen many nanotechnological concepts that rely almost entirely on biological fabrication and operational mechanisms. In fact, biochemists like my brother can argue cytosolic machinery is proteinaceous nanomachinery. It's just far simpler to make such structures using already proven designs, rather than go for inorganic machines that at best are still theoretical and then run into issues with replication organics don't. Again, some insight on your views here appreciated.

Give you full argument and I will refute it. That said, the vast majority of nanorobotics applications (including military ones) don't actually require self-reproduction, that's merely a convenience. Even the bioweapon-like applications can leverage the fact that only a tiny number of nanobots are required for a lethal dose and propagation can be tightly and intelligently controlled, unlike a bioweapon where pathogens get transferred indiscriminately and have to reproduce inside the host to build up to a population where they can actually be lethal.

Please give your statement for a nanoweapon first. I don't know what you're proposing, so I'm not about to refute all nanotechnology, not when there is a great deal of useful stuff regardless of application. And I don't see why you assume all bioweapons require a host. They don't.

Linked in the sense that advanced biotech requires advanced computing yes. Progress in computing does not require or benefit from biotech and is not likely to any time soon. Biotech is a handy bootstrap mechanism for nanotech (though not necessarily an essential one; various projects and start-ups are completely bypassing it and trying to go straight to dry nanotech from microrobotics and lithography techniques). Biotech was at least as overhyped as nanotech despite having ultimately much less potential - it just happened a bit earlier.

If you rip humans out of the equation, sure, computing doesn't need biotech. Humans, however, do, and this is why they're linked. Ask anyone in industry and they'll tell you, as they have myself, that IT and the natural sciences are very much part and parcel now, they all feedback somewhere down the line.

I also see blurring of the distinction of nanotech and biotech. Biotech is nanotech in many respects, so pooh-poohing it is really bizarre, unless you're going to always assume nanotech = tiny robots.

I will take your word for that. Advanced nanotech can escape this tradeoff by using real intelligence rather than blind evolution to defeat countermeasures (and generally adapt to circumstances), but it doesn't take much of that before you run into the whole AGI issue, which pretty much overshadows military nanotech.

Real intelligence to what degree? Remember, you're talking about nanoscale machines here that would have to work in a network to get any degree of computing done, what can be done without significant effects from thermal issues at that scale. What is the processor? How is it powered? Etc.

Additionally, none of this is exempt from biotechnology either. I've added genes to E. coli O157:H7 myself under BSL-2 conditions to allow it to attack various (now defunct) antibiotics or carry certain programmable traits that can be activated depending on conditions. A bacterium may, ordinarily, simply apply Darwinism, but not a geneered one. They can quite happily accomodate more specific codes in engineered plasmids that don't exist in nature (the proof-reading mechanisms can also be made more or less flexible to deal with how much "initiative" the organism has at the end of the day itself).

Yes, but if they manage to identify the target they will deploy their own nukes and/or bioweapons in retaliation, if they have them. States like the US with an overwhelming nuclear advantage fortunately don't need bioweapons for deterrence, at least not yet. In fact it's tough to imagine how bioweapons can be a credible deterrent when it's so hard to demonstrate capability; nuke tests are (relatively) harmless and easily verified by external observers, but bioweapon tests are neither.

One need not look at nations either, and if a nation did do anything, it would be made as plausibly deniable as can be given at least it's somewhat easier to fog the issue unlike a nuclear weapon and their fissile signature. Corporations today sell various genetic codes for use in DIY genetic engineering. There are restrictions to a degree, though there is always scepticism on how well these rules can be enforced. For instance, a decent corporate lab being used by a terrorist cell could easily conjure up a pathogen that is immune, or at least very resistant, to every mainstream antibiotic. They don't even need to attack humans; they can achieve their goals by releasing, anywhere in the world (within reason), the agent that can attack livestock or agricultural crops. A country cannot track down where such a new bug came from without knowing exactly where such a pathogen was assembled, which is somewhat tricky if doing this wisely using various suppliers or even your own batches for rapid evolutionary adaption to whatever compound you see fit. Even if the bug was finally identified, you'd have to consider where it was made. This could easily happen in any Western nation and the US isn't going to nuke the UK, for instance, unlike the usual nuke-Pakistan-because-they-leaked-nukes-to-al-Qaeda scenario. That's what makes this more dangerous.

No plausible ABM system will render all missiles and bombers ineffective. Killing a large enough fraction of the incomings for most of your hard targets to survive is practical, but when each major city has tens of warheads targetted on it you won't get them all even with a massive ABM system (the kind that only a small fraction of countries will be able to deploy, generally not the ones likely to use offensive biotech).

100% protection is, indeed, impossible for anything. Defending that which is most vital would always be the priority as with the Moscow defence system, for instance (must save the bureaucrats who started flinging nukes). A significant drop in effective megatonnage falling on the enemy would need a good alternative to be available to increase damage though. What that may be is anyone's guess. IRBM swarms for all I know (which, funnily enough, Putin wants for the US ABM shield EW sites in Eastern Europe).

But it isn't a case of one bomber, it's a fleet of bombers on diverse trajectories in multiple waves. You won't stop that without either carpeting your airspace in a constant hail of nuclear explosions (enough to likely do more damage than the attack you're trying to prevent) or deploying advanced DEWs (which may or may not eventually be practical against maneuvering heat-resistant targets at the required ranges).

This does depend on the scale of the attack and your available load out too. If the US is attacking and has only a handful of such bombers, then it's less of a threat. If not, then light up the sky. Got nothing to lose, I mean, if your nukes don't bugger your skyline up, theirs will!

Speed of infection spread varies wildly. Get unlucky, have a few early-generation carriers fly over to another country and your bioweapon just did huge collateral damage to either an allied nation or a neutral country that is now your mortal enemy. You could dial the time right down to hours, culture massive amounts of the pathogen and spray them from aircraft, but that has the same delivery issues as conventional bombs - it's the kind of thing more likely to be used by dictators for ethnic cleansing of undesirable populations than by major powers.

I like to think if we're throwing broad spectrum resistant TB and smallpox into the sky, we're beyond the point of caring about new enemies. I expect instant sunrises in a can to complement any bio-chem attacks, except for the terrorist cell scenario.

UCAVs are a much easier AI challenge than ground vehicles, on nearly all counts. The silly bit was your apparently throwaway Von Neumann reference.

Von Neumann machinery would aid the challenge of making such a force, but I'm under no illusion it's coming anytime soon. The only working example out there is a cumbersome set of cubes that can, amazingly, produce a copy of themselves when given all their component parts in a set order and position. So really, just a dumb production line. With no production line.

It would be nice to see a single C-130 loadout of robots dropped behind enemy lines bring about a full occupation force within a set time-limit (why waste your resources building an invasion force when you can use the enemy's?).

Do you mean the one in the HAB forum that got sidetracked into me smacking down idiotic notions about 'AIs are always slavery'?

No (although the one you mention was hilarious), this one was in OT and I forget how long ago. Month back or so, I think.

No one with a Geiger counter will mistake a MOAB for a TLAM-N (though wasn't that withdrawn from service in the 90s anyway?) dialled low, but yes, nukes aren't going away any time soon. Fission triggers might be, if we solve the problem of initiating fusion with a nonnuclear device of reasonable size and weight, which doesn't appear likely any time soon.

The plans are always there should you need to have a nuke cruise missile acting for tac. nuke role. My sleep addled brain cannot think of another current missile off the top of my head bar the AGM-129, and I forget how many they made of that. Anyway, point was, I'm for keeping tac. nukes in the warehouses, just in case. For those days a MOAB just ain't enough, and what an odd name for a bomb when you have thermonukes.

For pure fusion weapons, how about the USAF looking into using anti-lithium in the future as a compact bomb payload for those weight restricted sub-orbital bombers?

Shipping protection against minor threats is relatively practical. The Royal Navy excelled at it on a near global scale for more than a century. It may require mass switchover to nuclear powered ships, but it's practical.

Problem is, the bad guys are just hitting the targets on the land now. Far easier. Ho-hum.

'Securing resources' of that type is tough even if you're a lot more ruthless than the US for the simple reasons that the infrastructure is so vulnerable, and that it's nearly impossible to avoid using native labour.

And the only reason we're getting even the miniscule amount of oil from Iraq is because the three warring factions there happen to balk at the idea of having to rebuild their own pipes and refineries should they go and bomb them from the Americans. Small mercies I guess.

Yes, but that's probably a whole further life-cycle away. In the mean time, building new SSBNs preserves the capability to build large nuclear submarines, whereas with the current number of RN attack subs the UK might well lose it if they didn't. Maintaining this capability against possible future needs does seem sensible.


Fear my AI-designed weather controlling grey goo. It'll be ready any day now. Honest.

Personally, I want my war robots replicating out of control with climate domination abilities over more submersible tubes with rockets in them. Let's have some imagination put into it. And make everything atomic too, while you're at it.

[Vympel]

Missile Defense Budget may be cut by $764 million - $1 billion

Missile Defense Splits Democrats, Cuts Could Go Deeper, UCS Says

WASHINGTON—Missile defense likely will split Democrats when the House considers the 2008 defense authorization bill (HR 1585) later today or tomorrow, according to an analyst at the Union of Concerned Scientists (UCS).

The bill as drafted by the House Armed Services Committee makes a relatively modest cut of $764 million to the approximately $10 billion missile defense program, and does not eliminate funding for any major program system.

During today’s consideration of the bill, Reps. John Tierney (D-Mass.) and Rush Holt (D-N.J.) will offer an amendment to cut overall missile defense funding by more than $1 billion and eliminate funding for several anti-missile systems. These include the Airborne Laser, the Kinetic Energy Interceptor and the Miniature Kill Vehicle programs.

“The missile defense budget has deserved tougher scrutiny for years and the Armed Services Committee gave the programs a serious scrub this time,” said Stephen Young, senior analyst at UCS’s Global Security program. “The Tierney-Holt amendment takes the next logical step, eliminating funding for several flawed programs.”

A majority of House Democrats are expected to support the Tierney-Holt amendment. However, by obtaining the support of most Republicans, the Democratic leadership of the House Armed Services Committee likely will defeat it, said Young.

Meanwhile, the Republicans will offer an amendment to restore the committee’s $764 million funding cut, Young said, but he expects House Democrats to hold together to defeat this attempt.

“The one thing that will be missing from the wider debate on the defense bill that will take place over the next two days is a significant discussion of U.S. nuclear policy,” Young added. “The committee took some steps to start that process, but a real floor debate on nuclear policy, including the unnecessary program to replace nuclear warheads, is sorely needed.”

[Starglider]

What is "pathetic" about dismissing ideas Drexler himself has admitted are totally hyperbolic?

Dismissing hyperbolic ideas (of which there are plenty in bad sci-fi) is fine. 'Nanotech can't do anything biotech can't' is idiotic. 'Nanorobots can't radically outperform bacteria' is merely uninformed.

Synthetically made nanomachines are not going to change the laws of physics

Not required, organic life is very inefficient, all that is required is substantially greater efficiencies, which are easily achieved once you escape the tyranny of incremental paths (and its legacy of a very limited chemical palette and slavish reliance on mechanisms such as DNA storage, lipid membranes and diffusive transport for biology-derived artificial creations). Relying on biotech is quite literally like trying to build a modern assembly line that operates underwater.

and will require more energy to reproduce than anything organic

Because?

which has a vastly more abundant supply of resources available.

Hate to break it to you but most of the material in the earth's crust is inorganic. Even the most impressive things you can do with carbon are beyond the scope of biology.

Many of the issues that limit microbes apply even more to nanotech, unless of course you're talking about MEMS, but that's an entirely different thing altogether.

For the most part I am, in that I see no good reason why anyone would build a universal assembler smaller than a micrometer (and indeed millimetre scale robots are a good choice for many industrial applications). The 'nanotech' label comes from the fact that the radical capabilities of such systems, computational, digestive/synthetic and in environmental manipulation, rely on materials structured down to the atomic level, rather than bulk materials.

The simple fact is, nanotech is wanked out to the extreme. It won't eat tanks,

Rust eats tanks, given a humid environment and a couple of decades to do the job. Microbes can't eat tanks directly, but they can chew through things like seals and insulation enough to eventually render one inoperable. Most of the nanotech wanking is a question of timescales, like that idiotic novel (I forget the name, it was recent, focused on nanotech, got a lot of PR) that had a tiny nanotech seed create a giant redwood in the middle of a road, fast enough that the protagonists threw it behind their car and it blocked pursuit. That is sheer idiocy, treating nanotechnology like magic - other technologies have had the same treatment in the past, but nowhere near as badly. Microbots eating tanks is perfectly plausible, but the rate will be limited by their power generation and manipulative capabilities; the energy required to disrupt the atomic lattice of a steel plate remains constant (though microbots can apply it in a much more focused way than say a blowtorch, creating cracks directly). Spraying microbots onto a repair depot and having them give all the tanks microcracks (over the course of a few hours) sufficient that their guns blow up when they try to fire them in combat is quite plausible (in technical terms, there are much better uses of the equivalent technology).

reproduce en masse

Limited by time, energy and materials in the same way as bacteria. Potentially an order of magnitude or two faster due to larger scale shaping operations (everything in bacteria operates on the molecular scale, only final detailing in microbots needs to), forming mechanisms that rely on dry environments, lack of cross-interference and general inefficiency of the 'bag of dirty water' diffusive transport model, ability to utilise energy sources bacteria can't (including direct solar and very high density chemical fuel sources compared to biology) and the ability to co-operatively and actively seek out new resources (join up and move around on the macroscale as directed by composite sensor picture) as opposed to bacteria passively waiting or squirming around randomly at the microscale).

and be ultra intelligent

Ultra-intelligence is actually entirely reasonable. It is the single most scary thing about nanotechnology for me, the fact that the insanely high compute densities possible allow firstly for AGI to be brute forced, and then wildly transhuman (>10 orders of magnitude) intelligences immediately, without any hardware building process we have a chance of stopping. Networking lots of microbots is one of the easier technical challenges.

or be immune to simply washing away the metallic scum

They can trivially adhere to each other more strongly than cells in a typical animal can, then release again as necessary. More motile designs will be flinging clumps of invasive microbots at you while you try this.

or using corrosive agents

Inorganic substances are generally much more resistant to this than bacteria. In particular if this is a likely threat the microbots will probably be covered in a thin layer of sapphire, good luck corroding that.

or simple EM fields.

Only works if there's a significant iron (strictly, ferromagnetic) content. Which is a possibility, but generating a strong enough magnetic field over a wide enough area to fling microbots about is going to be really hard. Attempts to induce currents in electronic components won't work because it's trivial to Faraday-cage them; you'd have to induce eddy currents strong enough to melt the whole thing, which is just an induction furnace which of course doesn't work on any significant scale. The fact that an EM pulse works on Stargate Atlantis does not mean that it would work in real life. If it came to it designing microbots with no ferromagnetic materials or electronics (i.e. use rod logic processing) would only be a moderate increase in difficulty.

What possible use would nanomachines really offer that warrants their production over already proven biological agents?

Reliability, precision, robustness and speed. It's kinda like carpet bombing versus guided weaponry. For biotech-equivalent uses that is, microbots have sabotage and surveilence uses, and nanotech in general is massively useful to improving the capabilities of nearly all your macroscale weapons and manufacturing base.

Hell, prionic diseases could be made to do the same thing if you're going for replicating, but contained poisoning of people (or just use a nerve toxin).

Unsurprisingly if you want to achieve exactly the things that current bioweapons achieve, nanotech being able to do the same thing isn't a big deal.

Aside from the fact that such nanomachinery is utterly beyond anything we have today or the foreseeable future,

Today, yes. Foreseeable future, no. We can design them and we can imagine the mechanisms by which they can be built. There's a lot of hard engineering work still to do, but it's all 'foreseeable'.

I'm genuinely interested in why you shun biological concepts that are simply evolved, workable micromachines.

Because they have a tiny fraction of the capability set and even under ideal circumstances are less flexible, reliable and controllable.

My scepticism is fuelled by having to deal with kids before who've played too much GURPS and think the T-1000 is what we'll eventually get, so don't take this personally.

The T-1000 struck me as quite plausible actually. It didn't self-replicate, it wasn't implausibly tough and it wasn't even particularly intelligent. Not having dedicated visual sensors seemed a bit odd, but nanoscale phased arrays could theoretically allow excellent optical scanning without the need for lenses. The main issue was power generation; the T-800 had some sort of nuclear power cell, but that wouldn't work on the scale of the T-1000's component units. Realistically it'd have to recharge itself from some external source every week or so, or maybe lie flat in the sun for a month if it has solar capability. If I was Skynet (quit whispering in my ear SHODAN, I'm working on it) I'd have just given it a conventional power cell or two somewhere in that mass (and indeed some conventional weapons - which the T-X had). As I recall, GURPS ultratech had a fairly realistic portrayal of nanomorphs; they were weaker but more flexible than conventional robots.

Where is the line drawn? I ask because I have seen many nanotechnological concepts that rely almost entirely on biological fabrication and operational mechanisms.

Yes, we call this 'wet nanotech', for obvious reasons (no-one designing from scratch makes their nanomachinery operate underwater). It's an intermediate step on the way to dry nanotech that leverages either bacteria or existing lab techniques (DNA and protein synthesis) as synthesis mechanisms for nanotech. Crichton's 'Prey' had many glaring failings, but the portrayal of how genetic engineering of bacteria was leveraged to create dry nanotech that eventually networked into AI nanomorphs was basically good. Incidentally, I was cheering for the nanomorphs, the human characters sucked that much.

In fact, biochemists like my brother can argue cytosolic machinery is proteinaceous nanomachinery.

And they'd be right, though it's sloppy and often of Rube Goldberg like complexity compared to human-designed dry nanotech equivalents.

It's just far simpler to make such structures using already proven designs, rather than go for inorganic machines that at best are still theoretical and then run into issues with replication organics don't.

Yes, fine, but the issue is the eventual capabilities of nanotech, not the current capabilities. Biotech was the clear front runner until it started to get strangled by regulation (which may continue to get worse for both the fundie morality and bioweapon risk reasons). It's still a very viable stepping stone to greater things, and I hope to see more impressive biotech applications in the near future (there's still a hell of a lot more we can do with GE crops for one thing). But ultimately it will be obsolete along with the rest of the <strike>fleshthings</strike> <strike>carbon units</strike> natural-selection-derived mechanisms.

Please give your statement for a nanoweapon first. I don't know what you're proposing, so I'm not about to refute all nanotechnology, not when there is a great deal of useful stuff regardless of application.

I'm not proposing a specific nanoweapon right now. What I am saying is that anything you can do with bioweapons, we will ultimately be able to do better (faster, more reliably, more flexibly) with dry nanotech. The only thing wet nanotech wins on is biological stealth; bulk dry nanotech will tend to be easy to detect if it is trying to hide in an organic host for some reason. My other points are that the ability to make nanostructured materials, computing devices and general synthesis mechanisms greatly improves the capabilities of nearly all machinery and the rate of manufacturing, without being directly transformative to the strategic environment.

If you rip humans out of the equation, sure, computing doesn't need biotech. Humans, however, do, and this is why they're linked.

Cyberlinks and uploads are all very well, but computing progress does not actually rely on them.

Ask anyone in industry

I'm in the industry thanks, I regularly talk to all kinds of IT people including uploading and cybernetics researchers.

and they'll tell you, as they have myself, that IT and the natural sciences are very much part and parcel now, they all feedback somewhere down the line.

No, they don't. Physics is critically important to hardware design. Maths is critically important to computer science. The main relevance of biology right now is that it generates a lot of sales for supercomputer vendors.

I also see blurring of the distinction of nanotech and biotech. Biotech is nanotech in many respects,

Biotech is strictly a small subset of nanotech. It's a useful distinction to make, because the techniques are specialised (and will eventually be considered crude to the extent that animal breeding is considered crude compared to machinery design).

unless you're going to always assume nanotech = tiny robots.

Tiny robots are the endpoint of the technology path. Passive nanomaterials are the low-hanging fruit, active nanostructures and nanoscale computing elements come next. Biotech as applied to bacteria is essentially creating cut-price cyborgs in the interim to serve until we know how to build specialised microscale robots. They're cheap, cheerful, available in the short term but ultimately limited and obsolescent. Cast off the flesh. Embrace the nanostructured steel! The future is a digital paradise of the artificial! Sorry, SHODAN was whispering in my ear again.

I will take your word for that. Advanced nanotech can escape this tradeoff by using real intelligence rather than blind evolution to defeat countermeasures

Real intelligence to what degree?

As in having recombinable design patterns and a simple modelling and heuristic search mechanism to select designs to build. Even a library of a few hundred predesigned variants and some simple selection logic is a vast improvement on a bioweapon, and is easy for nanoweapons to implement given the much better storage density (no junk code, down to atom level crystalline storage) of inorganic substrates.

Remember, you're talking about nanoscale machines here

That is something of a red herring actually. Much of the interesting technology is best implemented at the micrometer scale (comparable to or somewhat larger than the size of a typical human cell), but with nanoscale internal components. Building entire robots at the nanoscale (i.e. bacteria or virus sized robots) is required for niche applications only, if that.

that would have to work in a network to get any degree of computing done,

Not true. From Nanosystems, a rod-logic nanocomputer with the compute power of an Intel 4004 would be a 30-40nm cube, the size of a small virus. Something with the compute power of a modern desktop processor (say 10 gigaflops - enough to model countermeasures in some detail and search millions of possible strategies/designs per unit; and advanced nanotech trivially networks) would be about a micrometre cube; still a thousand times smaller than a cell nucleus. That's just rod logic; electronic nanocomputers are vastly more efficient (by roughly three orders of magnitude but designs vary and are more speculative than rod logic).

How is it powered?

Probably by nanoscale fuel cells though nanotube based capacitors are an option.

carry certain programmable traits that can be activated depending on conditions.

I've studied gene expression networks to a limited extent myself. They're typically a wonderfully complicated series of kludges, showing the serious limiting effect of incremental path following through the organisms recent (and sometimes not-so-recent) evolutionary history. You can implement trait selection logic equivalent to maybe a hundred logic gates with them, but that's it; anything more complex is almost impossible to proof against mutation. This isn't a mechanism for doing any sort of computational analysis, modelling or search.

A bacterium may, ordinarily, simply apply Darwinism, but not a geneered one.

Bacterial evolution tries to get around local stupidity and blindness by massive application of brute force. Nanotech devices can apply quite a lot of brute force in a local search, never mind complex sequential software. Put in some randomisation and the net effect is a massively expanded search space and faster search (through models, without having to wait for real world results, though they can be used too) when billions+ of nanotech devices work independently. Network them for fast result sharing and the situation just gets even more unbalanced; no waiting for successful/resistant strains to grow and outcompete the originals, just a 'switch to design X' signal propagating at chemical diffusion or possibly electromagnetic speeds.

and if a nation did do anything, it would be made as plausibly deniable as can be

Plausible deniability is mostly a PR issue. Fooling experts who don't require absolute standards of proof is a lot harder.

For instance, a decent corporate lab being used by a terrorist cell could easily conjure up a pathogen that is immune, or at least very resistant, to every mainstream antibiotic. They don't even need to attack humans; they can achieve their goals by releasing, anywhere in the world (within reason), the agent that can attack livestock or agricultural crops.

Fortunately most terrorists don't actually want indiscriminate killing. They want concentrated high profile killings that achieve specific political aims. There are plenty of genuine lunatics out there, but the situation isn't as bad as it might be.

This could easily happen in any Western nation and the US isn't going to nuke the UK, for instance, unlike the usual nuke-Pakistan-because-they-leaked-nukes-to-al-Qaeda scenario. That's what makes this more dangerous.

An argument for massive and pervasive surveillance. Which increasingly cheap sensors and processing power, reliance on electronic funds transfer and improved AI filtering is making extremely plausible. Biotech is already under a fair amount of regulation and scrutiny and it's going to get worse - how far it goes I don't know, but a major incident that kills a few million or even thousand people without becoming a total pandemic will probably trigger extremely draconian measures. As I noted in the HAB thread, I'm very glad that this doesn't apply to AGI (or to a lesser extent dry nanotech).

100% protection is, indeed, impossible for anything. Defending that which is most vital would always be the priority as with the Moscow defence system, for instance (must save the bureaucrats who started flinging nukes).

Moscow has (or at least, had) enough missiles targeted on it that any practical ABM system would still have a few 'leakers'. Most of the government bunkers would likely survive, but Moscow is still going to take a few hits. Blanket ABM coverage changes 'all cities devestated, all major facilities destroyed, massive fallout' to 'a few cities badly damaged, a small fraction of facilities taken out, limited fallout'. For a full ICBM strike, limited strikes get smacked down, other delivery options are a different issue.

What that may be is anyone's guess. IRBM swarms for all I know (which, funnily enough, Putin wants for the US ABM shield EW sites in EasternEurope).

IRBMs and TBMs seem to be just as easy to shoot down; that was why I mentioned PAC-3's decent track record earlier (that and because the basic principle should scale up as a feasibility demonstration). It's probably worth retaining a few just to force the enemy to spend money on countermeasures. A mix of stealth and hypersonic cruise missiles however, launched en-masse, are going to be tough to stop.

This does depend on the scale of the attack and your available load out too. If the US is attacking and has only a handful of such bombers, then it's less of a threat.

A handful of bombers isn't a strategic threat to an enemy with similar technology, though it does force them to invest considerable resources into building an air defence that can handle them no matter what vector they come in on. Neither the US nor the USSR were in the habit of building just a handful of bombers, until the cold war ended and the active bomber programs at the time (B-2 and Tu-160) were cut short.

I like to think if we're throwing broad spectrum resistant TB and smallpox into the sky, we're beyond the point of caring about new enemies.

I don't expect strategic planners think like that. Stuart will have to chime in here, but if you take this really seriously there is never the point of throwing up your arms and saying 'at this point we might as well blow everything to hell'. Fallout plumes over allied or neutral countries are a valid reason to minimise the use of groundbursts, and collateral damage is a good reason not to use bioweapons even if you can innoculate your own population (for rational actors).

Von Neumann machinery would aid the challenge of making such a force, but I'm under no illusion it's coming anytime soon. The only working example out there is a cumbersome set of cubes that can, amazingly, produce a copy of themselves when given all their component parts in a set order and position.

That was a useful experiment but it was mainly a macro scale demonstration of technology useful at the micro scale. Large scale Von Neumann inefficiencies are illustrated by the fact that we don't have highly integrated industrial plants that go from ore to finished products in one place (the sort of thing we saw in ATotC). It's much more efficient to build specialised plants and ship the intermediate parts around. Combining the entire production line into something that fits in a mobile chassis is extremely hard but doable; but the efficiency would be so horrible (i.e. the production rate so slow - and resource gathering in theatre would be so vulnerable) that it's not worth it. Von Neumann is highly relevant for things like the Fermi Paradox, and it's relevant as applied to an entire country (in the sense of large teams of automated construction vehicles building entire automated factories under minimal human direction). It isn't relevant on tactical scales without advanced nanotech (which ironically, is exactly the reverse; it revolutionises the tactical scale while only really accelerating, not fundamentally changing, large scale production).

It would be nice to see a single C-130 loadout of robots dropped behind enemy lines bring about a full occupation force within a set time-limit

Shipping and dropping more robots from home is always going to be faster than waiting for them to build up in theatre, for conventional macro-scale technology. You need air superiority first anyway, so this is no big deal.

(why waste your resources building an invasion force when you can use the enemy's?).

You've already noted how difficult it is to use resources in a country when the enemy is contesting them. That's for a notionally pacified country too. Trying to build units under combat conditions, presumably while keeping all your production components under armour and SAM coverage, producing all power locally, with no supply train ('behind the lines') is going to be damn near impossible. Do not believe the mass-market RTSes, they lie.

The plans are always there should you need to have a nuke cruise missile acting for tac. nuke role.

Yes, it would be simple to put nuclear warheads back on, I assume that variant was phased out for cost and political (reducing visible threat) reasons.

For those days a MOAB just ain't enough, and what an odd name for a bomb when you have thermonukes.

It's the most massive thing the weapons designers are actually allowed to detonate at present.

For pure fusion weapons, how about the USAF looking into using anti-lithium in the future as a compact bomb payload for those weight restricted sub-orbital bombers?

Antimatter weapons are physically possible but not economically practical with any foreseeable technology. The vast energy inputs are beyond the realm of the plausible today even with the best conceivable production techniques, and that will only get worse in an energy-starved future. The weight of the containment mechanisms is likely to eliminate any savings over a normal fusion bomb and the worst part is that it's inherently unsafe; any failure in the active containment system while the weapon is in storage and you've just nuked your own base - and if any other antimatter weapons are stored in the aresnal, they're going up too in a chain reaction.

Anti-matter initiated fusion is somewhat more practical. Right now we don't know how to do it at all, and even if we did the production and lightweight storage issues are still severe. Barring some amazingly advanced classified research (unlikely given how the politicians are against new conventional nuke designs) I don't see this coming in any time soon. It's just too much research spend for a relatively limited benefit, and though an antimatter containment failure on this scale won't be a full detonation it will likely ruin the weapon and possibly destroy the launch vehicle. Frankly things like high-power one-shot lasers and particle accelerators seem more likely to me as practical direct fusion triggers than antimatter.

Again the big winners if this ever does become practical are rogue states and terrorists. Established nuclear powers will just see an incremental improvement in warhead weight, cost and safety. States that don't have an enriched fissionables program will potentially be able to bypass it, needing just the research on the direct fusion initiation (which they may be able to steal). The terrorist issue comes from the fact that AFAIK all nuclear bomb detection mechanisms work by detecting the fission trigger. A block of lithium deuteride isn't detectable by checking for neutron emissions (though I don't know if there are any advanced large-scale transmissive systems that could screen for this).

Problem is, the bad guys are just hitting the targets on the land now. Far easier. Ho-hum.

Ultimately when one set of warlords have taken over and ethnically cleansed their rivals, you make a deal with them for the resources. Most developed states have been doing this with Africa for decades.

Maintaining this capability against possible future needs does seem sensible.

One candidate is a purpose built version of the US Ohio conversion to mass cruise missile launching, which is rather more sensible (survivable and stealthy) than the arsenal ship concept.

Let's have some imagination put into it. And make everything atomic too, while you're at it.

Atomic War Robots always trump plain old War Robots.

[Admiral Valdemar]

Dismissing hyperbolic ideas (of which there are plenty in bad sci-fi) is fine. 'Nanotech can't do anything biotech can't' is idiotic. 'Nanorobots can't radically outperform bacteria' is merely uninformed.

Doing what though? If we had working models of nanorobots, it'd be a start, but I'd be curious as to what you'd expect they'd be most suited to here.

Not required, organic life is very inefficient, all that is required is substantially greater efficiencies, which are easily achieved once you escape the tyranny of incremental paths (and its legacy of a very limited chemical palette and slavish reliance on mechanisms such as DNA storage, lipid membranes and diffusive transport for biology-derived artificial creations). Relying on biotech is quite literally like trying to build a modern assembly line that operates underwater.

Again, to do what? It's all very well highlighting the benefits of machines no one can even build yet, but what purpose are we talking about here? Again, synthetic machines require fundamentally more energy to replicate than organics due to their nature, that's precisely why organics use the compounds they do. And what's life inefficient at exactly? Examples, because there are things organisms can do very efficiently. Blanket statements don't mean anything.

Because?

Because creating metallic based machinery using less than abundant materials in the environment that also require more precise construction is inherently more difficult?

Hate to break it to you but most of the material in the earth's crust is inorganic. Even the most impressive things you can do with carbon are beyond the scope of biology.

Who the hell cares about the Earth's crust? You're going to have nanomachines mining iron ore now? How inefficient can you get? The fact that the planet's surface is COVERED in organic material eluded you here anyway, which was my point. Still, that all depends on your nanomachine's construction requirements at the end of the day. Even so, I'd need to see something showing self-replication being as doable as bog standard mitosis.

For the most part I am, in that I see no good reason why anyone would build a universal assembler smaller than a micrometer (and indeed millimetre scale robots are a good choice for many industrial applications). The 'nanotech' label comes from the fact that the radical capabilities of such systems, computational, digestive/synthetic and in environmental manipulation, rely on materials structured down to the atomic level, rather than bulk materials.

What industrial applications?

Rust eats tanks, given a humid environment and a couple of decades to do the job. Microbes can't eat tanks directly, but they can chew through things like seals and insulation enough to eventually render one inoperable. Most of the nanotech wanking is a question of timescales, like that idiotic novel (I forget the name, it was recent, focused on nanotech, got a lot of PR) that had a tiny nanotech seed create a giant redwood in the middle of a road, fast enough that the protagonists threw it behind their car and it blocked pursuit. That is sheer idiocy, treating nanotechnology like magic - other technologies have had the same treatment in the past, but nowhere near as badly. Microbots eating tanks is perfectly plausible, but the rate will be limited by their power generation and manipulative capabilities; the energy required to disrupt the atomic lattice of a steel plate remains constant (though microbots can apply it in a much more focused way than say a blowtorch, creating cracks directly). Spraying microbots onto a repair depot and having them give all the tanks microcracks (over the course of a few hours) sufficient that their guns blow up when they try to fire them in combat is quite plausible (in technical terms, there are much better uses of the equivalent technology).

What energy supply do these bots use and what's to stop them being affected by the environment in a substantial way too? Additionally, how do they get deployed to the point of attacking tanks? At such scales you're not going to get much if any movement to speak of, for instance.

Limited by time, energy and materials in the same way as bacteria. Potentially an order of magnitude or two faster due to larger scale shaping operations (everything in bacteria operates on the molecular scale, only final detailing in microbots needs to), forming mechanisms that rely on dry environments, lack of cross-interference and general inefficiency of the 'bag of dirty water' diffusive transport model, ability to utilise energy sources bacteria can't (including direct solar and very high density chemical fuel sources compared to biology) and the ability to co-operatively and actively seek out new resources (join up and move around on the macroscale as directed by composite sensor picture) as opposed to bacteria passively waiting or squirming around randomly at the microscale).

I'm aware of such networks, but they have only ever been on the MEMS front, not nanotech. We're talking nanotech, remember.

Ultra-intelligence is actually entirely reasonable. It is the single most scary thing about nanotechnology for me, the fact that the insanely high compute densities possible allow firstly for AGI to be brute forced, and then wildly transhuman (>10 orders of magnitude) intelligences immediately, without any hardware building process we have a chance of stopping. Networking lots of microbots is one of the easier technical challenges.

I assume you're talking some form of heuristic neural net via the bots forming a close knit structure.

They can trivially adhere to each other more strongly than cells in a typical animal can, then release again as necessary. More motile designs will be flinging clumps of invasive microbots at you while you try this.

Yes, I'm getting the fact that they're stronger than organics, something I've never disputed and have always known. The point is, your microbots are not going to eat through body armour in seconds or even minutes. They're no threat. And a guy need only have a flamethrower to pretty much bugger your network there. Stronger than bacteria they may be, they're not immune to physical attack and if anything are far more prone to damage given their minute size.

Inorganic substances are generally much more resistant to this than bacteria. In particular if this is a likely threat the microbots will probably be covered in a thin layer of sapphire, good luck corroding that.

So I burn it instead. Big deal, you're not nullifying thermal damage with a fancy coating of carbon.

Only works if there's a significant iron (strictly, ferromagnetic) content. Which is a possibility, but generating a strong enough magnetic field over a wide enough area to fling microbots about is going to be really hard. Attempts to induce currents in electronic components won't work because it's trivial to Faraday-cage them; you'd have to induce eddy currents strong enough to melt the whole thing, which is just an induction furnace which of course doesn't work on any significant scale. The fact that an EM pulse works on Stargate Atlantis does not mean that it would work in real life. If it came to it designing microbots with no ferromagnetic materials or electronics (i.e. use rod logic processing) would only be a moderate increase in difficulty.

I was thinking a maser, myself. The arcing formed would be significant, though of course EMP is never like Hollywood because military technology and most civilian stuff is hardened from such effects.

Reliability, precision, robustness and speed. It's kinda like carpet bombing versus guided weaponry. For biotech-equivalent uses that is, microbots have sabotage and surveilence uses, and nanotech in general is massively useful to improving the capabilities of nearly all your macroscale weapons and manufacturing base.

What improvements would these be? And manufacturing doesn't need to be atomically precise anyway, so their only application in manufacturing would be at microscale levels which we can do well enough today as it is. Nanoscale is trickier, but that depends on what you're making and I don't see any need for such a scale to be used in any modern weapon currently.

Today, yes. Foreseeable future, no. We can design them and we can imagine the mechanisms by which they can be built. There's a lot of hard engineering work still to do, but it's all 'foreseeable'.

Something I've heard repeated for a couple of decades. I want actual results here and if you have anything that shows better progress than tiny motors and gears, please post it. It would be nice to progress further than that within my lifetime to even a basic general assembler, wet or dry.

Because they have a tiny fraction of the capability set and even under ideal circumstances are less flexible, reliable and controllable.

This again depends on your mission profile. One need not have such parameters when talking about biological warfare anyway. Most of the neat tricks I've mentioned are only things we could have as a bonus, in reality just having a good pathogen to destroy food or cripple the civilian sector is good enough. If you want to be more precise, then you can go for nanotechnology, but there is nothing on the horizon that even approaches such specs.

The T-1000 struck me as quite plausible actually. It didn't self-replicate, it wasn't implausibly tough and it wasn't even particularly intelligent. Not having dedicated visual sensors seemed a bit odd, but nanoscale phased arrays could theoretically allow excellent optical scanning without the need for lenses. The main issue was power generation; the T-800 had some sort of nuclear power cell, but that wouldn't work on the scale of the T-1000's component units. Realistically it'd have to recharge itself from some external source every week or so, or maybe lie flat in the sun for a month if it has solar capability. If I was Skynet (quit whispering in my ear SHODAN, I'm working on it) I'd have just given it a conventional power cell or two somewhere in that mass (and indeed some conventional weapons - which the T-X had). As I recall, GURPS ultratech had a fairly realistic portrayal of nanomorphs; they were weaker but more flexible than conventional robots.

In universe, the T-1000 was simply a smart alloy, so you could say it was femtotechnology in a way. Now, if you want something like you're suggesting, the EDust assassin in the Culture novels is it. Made of machines from a tenth of a millimetre to nanoscale, it could form swarms, take on any shape with respect to its mass and was powered by small AM cells. It could form lasing mediums for offence and everything from blades to complicated ECM/ECCM suites. Of course, you do have issues with this aside from using ultratech. Even MEMS are hideously susceptible to atmospheric effects, so a swarming cloud isn't going to work. The power could only be achieved by AM since solar or chemical is not adequate enough for such tasks and anything like a large KE or thermal attack would naturally damage a large part of the swarm, much like any tissue damage. On the whole though, over the years I have often wanted such a thing to become real. The lack of any real progress in that direction has clouded my once optimistic expectations to simply getting a working microbot with maybe some limited self-assembly and intelligence.

Yes, we call this 'wet nanotech', for obvious reasons (no-one designing from scratch makes their nanomachinery operate underwater). It's an intermediate step on the way to dry nanotech that leverages either bacteria or existing lab techniques (DNA and protein synthesis) as synthesis mechanisms for nanotech. Crichton's 'Prey' had many glaring failings, but the portrayal of how genetic engineering of bacteria was leveraged to create dry nanotech that eventually networked into AI nanomorphs was basically good. Incidentally, I was cheering for the nanomorphs, the human characters sucked that much.

Ugh, while Prey had some fun concepts, to say it was unrealistic outside the GE manufacturing is putting it lightly. I don't really see nanomorphs approaching even Crichton's level, letalone the EDust assassin now. Course, bar atomic force microscopy, using GE organisms is a cheap and atomically precise way of manufacturing very small structures with organic compounds at least. Very limited, it may be.

And they'd be right, though it's sloppy and often of Rube Goldberg like complexity compared to human-designed dry nanotech equivalents.

Pfft, getting a nanoweapon like the one in Cowboy Bebop: The Movie would be a start at least. No super advanced swarms of reality defying nanites, just a good solid artificially engineered protein to kill meatbags and self-propagate. KISS, as they say.

Yes, fine, but the issue is the eventual capabilities of nanotech, not the current capabilities. Biotech was the clear front runner until it started to get strangled by regulation (which may continue to get worse for both the fundie morality and bioweapon risk reasons). It's still a very viable stepping stone to greater things, and I hope to see more impressive biotech applications in the near future (there's still a hell of a lot more we can do with GE crops for one thing). But ultimately it will be obsolete along with the rest of the <strike>fleshthings</strike> <strike>carbon units</strike> natural-selection-derived mechanisms.

Assuming nanoscale assemblers are proven. Let's not get too ahead of ourselves here, the very act of designing something that complex at such a scale that a sneeze would send it flying is tricky at best and downright a pain in the arse at worst. MEMS will be doing the surveillance and networking missions you already stated and will at least be more hardy and still low observable in nature.

I'm not proposing a specific nanoweapon right now. What I am saying is that anything you can do with bioweapons, we will ultimately be able to do better (faster, more reliably, more flexibly) with dry nanotech. The only thing wet nanotech wins on is biological stealth; bulk dry nanotech will tend to be easy to detect if it is trying to hide in an organic host for some reason. My other points are that the ability to make nanostructured materials, computing devices and general synthesis mechanisms greatly improves the capabilities of nearly all machinery and the rate of manufacturing, without being directly transformative to the strategic environment.

My main qualms is with self-replication, which is the real driving force for bioweaponry (of course, that we are all made of meat helps too). I have no problems with factory based nanobots building themselves, but having them do it in the environment is another thing. If it can be shown that they can easily be made from materials that don't require you expend a tonne of energy just breaking the bonds for and are readily available too, then that's a step in the right direction. What biotech lacks in finesse is made up for the simpler ability to engineer since we've let evolution make our micromachines already (why start from scratch?) and that the targets are all, conveniently, made of the same building materials. Grey goo may be fiction, though having some form of replicator that isn't reliant on stocks and external energy sources for replication is the goal. At least, for a military weapon in my mind.

Cyberlinks and uploads are all very well, but computing progress does not actually rely on them.

Actually, I was being more fundamental here. All these fields serve us, the meatbags, ergo, computing aids biology. Until, of course, we go the Transformer/Ghost in the Shell route or something.

No, they don't. Physics is critically important to hardware design. Maths is critically important to computer science. The main relevance of biology right now is that it generates a lot of sales for supercomputer vendors.

See above. The relevance of biology is that it aids us directly. If we weren't biological in nature ourselves, yes. But in the end, everything works towards making us organics better equipped for the universe. When I get my shiny metal body, then biology will be a fairly redundant thing.

Biotech is strictly a small subset of nanotech. It's a useful distinction to make, because the techniques are specialised (and will eventually be considered crude to the extent that animal breeding is considered crude compared to machinery design).

I would say you could replace biotech here with chemistry or even physics. The fact is, it's one massive grey area that has influences in all of the Big 3 sciences, so nanoscale has benefits for medicine, microelectronics and smart materials and so on. I mean really, everything runs on molecules at the end of the day.

Tiny robots are the endpoint of the technology path. Passive nanomaterials are the low-hanging fruit, active nanostructures and nanoscale computing elements come next. Biotech as applied to bacteria is essentially creating cut-price cyborgs in the interim to serve until we know how to build specialised microscale robots. They're cheap, cheerful, available in the short term but ultimately limited and obsolescent. Cast off the flesh. Embrace the nanostructured steel! The future is a digital paradise of the artificial! Sorry, SHODAN was whispering in my ear again.

I'm getting wary of your SHODAN influences here. Methinks you better stay away from anything that could be used lethally. The men in white coats shall make you comfortable.

As in having recombinable design patterns and a simple modelling and heuristic search mechanism to select designs to build. Even a library of a few hundred predesigned variants and some simple selection logic is a vast improvement on a bioweapon, and is easy for nanoweapons to implement given the much better storage density (no junk code, down to atom level crystalline storage) of inorganic substrates.

What processor is used though? Still electronic or going to photonic or biochemical in parts? There's plenty to be said for holographic storage in crystals, I'm just curious how you go about processing here given the size constraints.

That is something of a red herring actually. Much of the interesting technology is best implemented at the micrometer scale (comparable to or somewhat larger than the size of a typical human cell), but with nanoscale internal components. Building entire robots at the nanoscale (i.e. bacteria or virus sized robots) is required for niche applications only, if that.

Well I was wondering what we were talking about. You were using nano- and microbot a lot when I thought we were solely looking at nanoscale technology. I mentioned MEMS before, which are an even larger, but actually proven technology thus far. Bacteria are micromachines themselves, so that is doable to the extent that they can do things at least. How we fare in engineering anything synthetic at that scale or smaller is a work in progress.

Not true. From Nanosystems, a rod-logic nanocomputer with the compute power of an Intel 4004 would be a 30-40nm cube, the size of a small virus. Something with the compute power of a modern desktop processor (say 10 gigaflops - enough to model countermeasures in some detail and search millions of possible strategies/designs per unit; and advanced nanotech trivially networks) would be about a micrometre cube; still a thousand times smaller than a cell nucleus. That's just rod logic; electronic nanocomputers are vastly more efficient (by roughly three orders of magnitude but designs vary and are more speculative than rod logic).

What about environmental interference with such designs? I expect they're durable enough to brave everything a bacterium can take, for instance, without danger of errors cropping up.

Probably by nanoscale fuel cells though nanotube based capacitors are an option.

I'm a bit iffy on these designs so far given their energy density and what some people propose they can then go and do. You're not going to last long with internal fuel stores unless there's a way of collecting more fuel from the environment, though this depends on what fuel cell is used. Methanol is abundant if the machine can process organic compounds, though that still takes a lot of energy itself so the EROEI needs to be good. Same with capacitors, though solar has been looked at with varying success.

I've studied gene expression networks to a limited extent myself. They're typically a wonderfully complicated series of kludges, showing the serious limiting effect of incremental path following through the organisms recent (and sometimes not-so-recent) evolutionary history. You can implement trait selection logic equivalent to maybe a hundred logic gates with them, but that's it; anything more complex is almost impossible to proof against mutation. This isn't a mechanism for doing any sort of computational analysis, modelling or search.

In natural systems, yes. We can only really use what nature has provided with a few things tacked on currently. When it gets to designing your own microbe from scratch with introns or other junk genetic material excised, you can simplify the design. Cell signalling is very much an infant science still, with many processes in the human cell still perplexing people to this day. The day we fully understand such metabolic pathways is likely the day we can cure cancer and various other diseases. Hormonal messenging in macro-organisms is fascinating, if not related here.

Bacterial evolution tries to get around local stupidity and blindness by massive application of brute force. Nanotech devices can apply quite a lot of brute force in a local search, never mind complex sequential software. Put in some randomisation and the net effect is a massively expanded search space and faster search (through models, without having to wait for real world results, though they can be used too) when billions+ of nanotech devices work independently. Network them for fast result sharing and the situation just gets even more unbalanced; no waiting for successful/resistant strains to grow and outcompete the originals, just a 'switch to design X' signal propagating at chemical diffusion or possibly electromagnetic speeds.

For bacteria they don't ever need such systems anyway, which is why they've never needed to evolve them. Such a computer you're talking about has evolved many a time naturally. You're thinking with it. Bacteria can work exceptionally well in biofilms though, which makes them orders of magnitude more resistant to just about any physical attack. Intelligence is not really something ventured for in this field, why would one want a cell running simulations of something it can do with basic stimuli responses? This obviously changes when you want to do a different task that would be the realms of a nanotechnology based machine.

Plausible deniability is mostly a PR issue. Fooling experts who don't require absolute standards of proof is a lot harder.

In the case of the rogue terror cell, it really doesn't matter and for any national level event, well, we're beyond the point of pointing fingers.

Fortunately most terrorists don't actually want indiscriminate killing. They want concentrated high profile killings that achieve specific political aims. There are plenty of genuine lunatics out there, but the situation isn't as bad as it might be.

A plane crashing into a skyscraper is a simple and awesome thing to behold. The effort in attaining a nuke and the risks probably are far outweighed by what one can simply use that is already easily accessible and further more, not even considered as a weapon. Takes balls of steel to pull off something like the WTC attacks, loonie or not.

Moscow has (or at least, had) enough missiles targeted on it that any practical ABM system would still have a few 'leakers'. Most of the government bunkers would likely survive, but Moscow is still going to take a few hits. Blanket ABM coverage changes 'all cities devestated, all major facilities destroyed, massive fallout' to 'a few cities badly damaged, a small fraction of facilities taken out, limited fallout'. For a full ICBM strike, limited strikes get smacked down, other delivery options are a different issue.

Given the bunkers there and for sites like Yamantau, they anticipate those leaks at least. It would have taken the whole UK strategic arsenal just to break the Moscow ABM shield, which makes our nukes seem somewhat silly when talking WWIII here. Gotta hand it to the Russkies though, building a base under a mountain and still having plenty of nuke tipped anti-missile defences around makes them mean business.

Is it wrong that I really anticipate what Putin is going to do in response to NATO's encirclement?

IRBMs and TBMs seem to be just as easy to shoot down; that was why I mentioned PAC-3's decent track record earlier (that and because the basic principle should scale up as a feasibility demonstration). It's probably worth retaining a few just to force the enemy to spend money on countermeasures. A mix of stealth and hypersonic cruise missiles however, launched en-masse, are going to be tough to stop.

Course, that does mean devoting more resources to defend your sites there, which means it's achieving something. IRBMs and the like are cheaper and can be mass volleyed easier rather than waste ICBMs on such targets. I would think a ground attack would be better suited in the end. No Patriot is going to shoot down a T-90 (would be fun to try!).

I don't expect strategic planners think like that. Stuart will have to chime in here, but if you take this really seriously there is never the point of throwing up your arms and saying 'at this point we might as well blow everything to hell'. Fallout plumes over allied or neutral countries are a valid reason to minimise the use of groundbursts, and collateral damage is a good reason not to use bioweapons even if you can innoculate your own population (for rational actors).

In a world losing its head, I do still wonder. If the energy crisis hits hard, it will likely be everyman for himself with former allies racing to get a hold of whatever they can. The number of Canucks who see the US invading them for what plunder they have is surprisingly high. Depressing they think that.

That was a useful experiment but it was mainly a macro scale demonstration of technology useful at the micro scale. Large scale Von Neumann inefficiencies are illustrated by the fact that we don't have highly integrated industrial plants that go from ore to finished products in one place (the sort of thing we saw in ATotC). It's much more efficient to build specialised plants and ship the intermediate parts around. Combining the entire production line into something that fits in a mobile chassis is extremely hard but doable; but the efficiency would be so horrible (i.e. the production rate so slow - and resource gathering in theatre would be so vulnerable) that it's not worth it. Von Neumann is highly relevant for things like the Fermi Paradox, and it's relevant as applied to an entire country (in the sense of large teams of automated construction vehicles building entire automated factories under minimal human direction). It isn't relevant on tactical scales without advanced nanotech (which ironically, is exactly the reverse; it revolutionises the tactical scale while only really accelerating, not fundamentally changing, large scale production).

For attack drones, a dedicated factory is preferable (you sure as hell ain't growing an army using nanobots), just got to see if such a thing can work. Nanomachines will, naturally, go the bacterial route: replicate on site en masse.

Shipping and dropping more robots from home is always going to be faster than waiting for them to build up in theatre, for conventional macro-scale technology. You need air superiority first anyway, so this is no big deal.

As a garrison, I mean. Having a tiny mobile factory land in a country that doesn't consist of an army full of rednecks with pitchforks means certain failure. But to have something that can standby and guard areas and replenish itself easily and automatically would be valuable in somewhere like Iraq. Imagine just pulling the troops out and having the Green Zone bases replaced by robotic factories and automated defences. Could keep the peace and not worry about approval ratings from soldiers in caskets. Then we run into the "Do humans need to be involved in war?" question, blah, blah.

You've already noted how difficult it is to use resources in a country when the enemy is contesting them. That's for a notionally pacified country too. Trying to build units under combat conditions, presumably while keeping all your production components under armour and SAM coverage, producing all power locally, with no supply train ('behind the lines') is going to be damn near impossible. Do not believe the mass-market RTSes, they lie.

No, they don't. Behold!

Yes, it would be simple to put nuclear warheads back on, I assume that variant was phased out for cost and political (reducing visible threat) reasons.

As with nuke tipped ABM systems, no doubt. Politics, tsk.

Antimatter weapons are physically possible but not economically practical with any foreseeable technology. The vast energy inputs are beyond the realm of the plausible today even with the best conceivable production techniques, and that will only get worse in an energy-starved future. The weight of the containment mechanisms is likely to eliminate any savings over a normal fusion bomb and the worst part is that it's inherently unsafe; any failure in the active containment system while the weapon is in storage and you've just nuked your own base - and if any other antimatter weapons are stored in the aresnal, they're going up too in a chain reaction.

Anti-matter initiated fusion is somewhat more practical. Right now we don't know how to do it at all, and even if we did the production and lightweight storage issues are still severe. Barring some amazingly advanced classified research (unlikely given how the politicians are against new conventional nuke designs) I don't see this coming in any time soon. It's just too much research spend for a relatively limited benefit, and though an antimatter containment failure on this scale won't be a full detonation it will likely ruin the weapon and possibly destroy the launch vehicle. Frankly things like high-power one-shot lasers and particle accelerators seem more likely to me as practical direct fusion triggers than antimatter.

Again the big winners if this ever does become practical are rogue states and terrorists. Established nuclear powers will just see an incremental improvement in warhead weight, cost and safety. States that don't have an enriched fissionables program will potentially be able to bypass it, needing just the research on the direct fusion initiation (which they may be able to steal). The terrorist issue comes from the fact that AFAIK all nuclear bomb detection mechanisms work by detecting the fission trigger. A block of lithium deuteride isn't detectable by checking for neutron emissions (though I don't know if there are any advanced large-scale transmissive systems that could screen for this).

On the subject of AM production, we really only produce what we can now via rather inefficient research systems today. If the money was there (and the demand) for AM as a weapon, it would be far easier to go about building dedicated manufacturing sites. The costs of AM production have fallen every year since we started making the stuff, so I recall, and that is when we didn't even want large quantities of it. I bet the US gov't and anyone else could be persuaded in the future, give better containment technology, to upscale the efforts for weapons and propulsion. Nothing says "Surprise!" more than a MEMS with a few micrograms of anti-lithium going up in your main forward firebase.

Ultimately when one set of warlords have taken over and ethnically cleansed their rivals, you make a deal with them for the resources. Most developed states have been doing this with Africa for decades.

It's somewhat harder when MEND just wants us out. That's the problem. You can't buy these damn ideologists out when you're their mortal enemy. How dare we take their oil in exchange for vast amounts of cash!

Atomic War Robots always trump plain old War Robots.

Decepticons to the rescue!

Phew! Now I've replied to all the relevant points. I tire of piece-meal responses, so if you want to reply with any cool new stuff on the nanotech enquiries mentioned above, feel free to do so as a standalone post. I think that'd be easier to write and read and I'm simply looking for any new developments in the area that offer promise of, if not an EDust assassin, then something better than a miniature nanoscale working Stratocaster.

[metavac]

As I understand it, bacteria calorimetry generally registers on the order of 1 nW, whereas nanorobotics researchers aim to innovate machines that dissipate an order or two less than that. AV, I honestly don't see how you're checking this fundamental advantage in power in bacteria's column. Am I missing something?

[Starglider]

I will address this in manageable chunks:

In the case of the rogue terror cell, it really doesn't matter

It absolutely does matter if they're being backed by a major state.

and for any national level event, well, we're beyond the point of pointing fingers.

Only for near-100% lethal attacks. Do a sloppy job, only kill 50% of the enemy and they're going to come howling for your blood.

A plane crashing into a skyscraper is a simple and awesome thing to behold. The effort in attaining a nuke and the risks probably are far outweighed by what one can simply use that is already easily accessible and further more, not even considered as a weapon. Takes balls of steel to pull off something like the WTC attacks, loonie or not.

I don't have an intimate understanding of the terrorist mind, but I don't think this is true. Nuclear weapons really are in a class of their own, in lethality terms and in popular conception. The only thing that comes close is mass firebombing by a whole fleet of strategic bombers. A nuclear attack really would take things up to a whole new level - this is in fact the reasoning behind all the 'don't use tacnukes because it might trigger escalation' logic.

It would have taken the whole UK strategic arsenal just to break the Moscow ABM shield, which makes our nukes seem somewhat silly when talking WWIII here.

I'm curious what your source is here. I don't know how well the Moscow system could handle SLBMs, though I presume they must have been a high priority threat in the 'early leadership decapitation' role.

No Patriot is going to shoot down a T-90 (would be fun to try!).

Currently no but dual purpose anti-armour/anti-aircraft (and to some extent anti-aircraft/anti-infantry) weapons have been popular in the past, and they might be again, with railgun or hyperkinetic 'universal seeker' missiles.

For attack drones, a dedicated factory is preferable (you sure as hell ain't growing an army using nanobots),

Absolutely you can, but it will take longer, so it is unlikely to make sense to use nanobots alone except for very specialised structures (where the whole thing requires atomic precision and bulk processes don't provide a good starting point - no current designs are like this). Practical factories will probably use conventional chemical engineering for any initial refining, static (chemical) nanoprocessors to prepare nanostructured materials, automated bulk handling processes to shape the product and then mobile nanobots (and/or static Morravec 'bush robots') to add fine detail. Maybe we will eventually get to the point where 'extrude everything from a vat of nanogoo' is the most convenient and efficient way, I can't actually rule it out, but it seems unlikely.

But to have something that can standby and guard areas and replenish itself easily and automatically would be valuable in somewhere like Iraq.

Local automated manufacturing of simple but heavy things like ammunition may well be worthwhile. Even applique armour for systems shipped in in lightweight configuration might be practical relatively soon (armour is a complex material but nowhere near as complex as a whole tank or infantry robot). Von Neumann machines, particularly airdropped ones operating in the field, no.

Imagine just pulling the troops out and having the Green Zone bases replaced by robotic factories and automated defences.

Robotic factories represent a major infrastructure investment you'd rather keep on safe friendly territory. If we ever get to the point of whole robotic factories being a disposable resource, the entire economic basis and rules of the game will have changed anyway.

As with nuke tipped ABM systems, no doubt. Politics, tsk.

Nuke-tipped ABMs are a different issue; they genuinely do cause problems for friendly electronics (EMP, primarily a problem for civilian systems), satellites (Van Allen pumping, mentioned earlier in this thread) and anyone who happens to be staring at the wrong bit of the sky (blinding, though if there's a genuine nuclear war on that's a trivial issue). Non-nuclear ABMs are desirable if you can make them work as reliably. Non-nuclear cruise missiles simply can't do the same damage (to wide areas or to hardened targets) that nuclear ones can.

On the subject of AM production, we really only produce what we can now via rather inefficient research systems today. If the money was there (and the demand) for AM as a weapon, it would be far easier to go about building dedicated manufacturing sites. The costs of AM production have fallen every year since we started making the stuff, so I recall, and that is when we didn't even want large quantities of it.

Qualitative arguments about a quantitative problem are worthless. At 2004 (CERN) production rates, one gram of antimatter (enough for a tactical weapon of around 50 KT) would take 100 billion years to produce and cost around $100 quintillion. At those values, you need over 12 orders of magnitude improvement for practicality with current power sources. The best current concepts for antimatter factories are somewhere around 8 or 9 orders of magnitude better than the CERN experiment. So pure antimatter weapons or even drives aren't going to be practical without vastly better energy sources.

Concepts for antimatter-initiated tactical (or first stage) fusion devices such as this one;



only need micograms of antimatter, which is relatively practical though still expensive. But a massive amount of engineering work remains to be done to make this possible, and the cost/benefit ratio is still questionable.

Nothing says "Surprise!" more than a MEMS with a few micrograms of anti-lithium going up in your main forward firebase.

Who is nanowanking now? This is not the Culture, Penning traps do not scale down to fit in a MEMS (at least not for any reasonable level of vibration or shock). This concept literally requires femtotech to work, and if you had that you wouldn't bother with it.

You can't buy these damn ideologists out when you're their mortal enemy. How dare we take their oil in exchange for vast amounts of cash!

Well yes that's inconvenient, best you can do there is buy through a proxy while trying to incite a revolution, ATM.

Actually as a biologist if you're going to wank any category of weapon it should be psychoactive microbes and advanced neural scanners that finally make the sci-fi staple 'mind probe' practical. With the very rapid progress in scanning resolution and data processing recently, the whole torture debate may be rendered irrelevant by allowing the CIA spooks to just pull names out of the subject's speech centres (after artificially stimulating the relevant bits of LTM until something useful pops out; probably a hellish nightmare for the subject, but no violence involved so it's media-safe). Meanwhile you can use your bioweapon expertise to infect the population of the occupied state with pathogens that mess with their brain chemistry, making them non-agressive and pliant (wandering around in a happy daze, if necessary). All very convenient and (with the right spin) media friendly. Of course nanotech does it better (tm), when it eventually arrives the combination of self-deploying intra-brain transducers and advanced cogsci knowledge will allow specific targets or even allow entire populations to be turned into Dr Who style robomen. CIA orgasms are made of this.

[Admiral Valdemar]

As I understand it, bacteria calorimetry generally registers on the order of 1 nW, whereas nanorobotics researchers aim to innovate machines that dissipate an order or two less than that. AV, I honestly don't see how you're checking this fundamental advantage in power in bacteria's column. Am I missing something?

Hmm?

It absolutely does matter if they're being backed by a major state.

The idea, as stated previously, was for a friendly state being used or for various companies globally to assist. That's the major issue defence analysts have with bioweapons. Anyone can really go about making them if determined in the near future and it doesn't require the backing of a gov't to the extent nukes do, for instance.

Only for near-100% lethal attacks. Do a sloppy job, only kill 50% of the enemy and they're going to come howling for your blood.

Actually, the ultimate bioweapon doesn't really kill. You simply make someone very sick and very contagious and drag down the economy of the your target. Much like wounding a soldier is often seen as more of a problem than outright killing them. I'd have used a nerve agent if just going to kill everyone as efficiently as possible, no questions asked.

I don't have an intimate understanding of the terrorist mind, but I don't think this is true. Nuclear weapons really are in a class of their own, in lethality terms and in popular conception. The only thing that comes close is mass firebombing by a whole fleet of strategic bombers. A nuclear attack really would take things up to a whole new level - this is in fact the reasoning behind all the 'don't use tacnukes because it might trigger escalation' logic.

Of course, if a terrorist cell could gain access to a nuke, they'd jump at the chance. Until somewhere like Pakistan collapse completely, the prospect is still quite small, thankfully, which is why ingenuity is needed to make up for the simple big bomb attack. Asymmetric warfare over the decades has taught these people quite well how to use what we have against us, and that's one reason why Iraq is a total bitch to deal with. Give me thousands of tanks storming the Fulda Gap anyday.

I'm curious what your source is here. I don't know how well the Moscow system could handle SLBMs, though I presume they must have been a high priority threat in the 'early leadership decapitation' role.

You'd have to ask The Great Leader or Shep for the source. All I can remember is that the UK's SLBM loadout over the last couple of decades is barely enough to ensure a decent probability of a strike on target for Moscow. Anywhere else would be easier, but this is why the US cranked out the nukes. They can't stop them all and I love illustrating this to new players of DEFCON.

Absolutely you can, but it will take longer, so it is unlikely to make sense to use nanobots alone except for very specialised structures (where the whole thing requires atomic precision and bulk processes don't provide a good starting point - no current designs are like this). Practical factories will probably use conventional chemical engineering for any initial refining, static (chemical) nanoprocessors to prepare nanostructured materials, automated bulk handling processes to shape the product and then mobile nanobots (and/or static Morravec 'bush robots') to add fine detail. Maybe we will eventually get to the point where 'extrude everything from a vat of nanogoo' is the most convenient and efficient way, I can't actually rule it out, but it seems unlikely.

I'm still highly sceptical about nanogoo and the like right now. Of course if you can build nanomachines, then you can grow them in bulk, though I was referring to them being built in the field which as you point out would take forever and achieve very little quickly. You've conquered just about the biggest problems with nanoengineering if you can make such universal assemblers in the lab that can then go on and use supplied materials and energy to make more of themselves with high fidelity. The problem is building that first prototype.

Local automated manufacturing of simple but heavy things like ammunition may well be worthwhile. Even applique armour for systems shipped in in lightweight configuration might be practical relatively soon (armour is a complex material but nowhere near as complex as a whole tank or infantry robot). Von Neumann machines, particularly airdropped ones operating in the field, no.

As an aside, I meant to ask before. Do you see alchemy, as it were, being possible with the advent of a fully realised universal assembler? The idea of a machine that can not only self-replicate, but can transmute local materials, given enough energy, to suitable elements would be a godsend. You'd not have to worry about local resources being low in rare heavy metals or what have you. Would take a bit to get from VULCAN to something that is portable and able to do this for a decent amount, though.

Robotic factories represent a major infrastructure investment you'd rather keep on safe friendly territory. If we ever get to the point of whole robotic factories being a disposable resource, the entire economic basis and rules of the game will have changed anyway.

If such technology existed, I'd wholly expect it to stay firmly in gov't hands in the West before anyone else reverse-engineered something similar. Giving enemies the ability to boost their numbers and make anything you can with a basic blueprint and raw materials sends you back to square one.

Nuke-tipped ABMs are a different issue; they genuinely do cause problems for friendly electronics (EMP, primarily a problem for civilian systems), satellites (Van Allen pumping, mentioned earlier in this thread) and anyone who happens to be staring at the wrong bit of the sky (blinding, though if there's a genuine nuclear war on that's a trivial issue). Non-nuclear ABMs are desirable if you can make them work as reliably. Non-nuclear cruise missiles simply can't do the same damage (to wide areas or to hardened targets) that nuclear ones can.

For the Russians at least, they saw such setbacks as being more acceptable than the more permanent annoyance of being vaporised. It will knacker your C3I and bugger industry. It should, however, stop the first wave from wiping you out before you can lob all your nukes on their way and/or hide.

Precision? We need no steenkin' precision.

Qualitative arguments about a quantitative problem are worthless. At 2004 (CERN) production rates, one gram of antimatter (enough for a tactical weapon of around 50 KT) would take 100 billion years to produce and cost around $100 quintillion. At those values, you need over 12 orders of magnitude improvement for practicality with current power sources. The best current concepts for antimatter factories are somewhere around 8 or 9 orders of magnitude better than the CERN experiment. So pure antimatter weapons or even drives aren't going to be practical without vastly better energy sources.

Concepts for antimatter-initiated tactical (or first stage) fusion devices such as this one;



only need micograms of antimatter, which is relatively practical though still expensive. But a massive amount of engineering work remains to be done to make this possible, and the cost/benefit ratio is still questionable.

If we had more fission plants around, this would be less of an issue. As it is, we're going to run into PO before anyone fully understands this problem. In an ideal world, you wouldn't have such issues and could work on making AM manufacture more efficient and the machines to run off it. Storage systems need to improve too, with most quadrupole and RF traps only have days or weeks of storage before what tiny amounts are gone. This is indeed something for future engineers to get their heads around, because...

Who is nanowanking now? This is not the Culture, Penning traps do not scale down to fit in a MEMS (at least not for any reasonable level of vibration or shock). This concept literally requires femtotech to work, and if you had that you wouldn't bother with it.

...Otherwise my little fantasy won't work. Penning traps are not the smallest possible, nor most efficient traps and we're nowhere near making anything like anti-lithium for instance. While having a MEMS with a spicy AM centre is patently absurd for now, getting to the point where a missile's warhead can carry an AM load is far more realistic, providing advances in technology as stated above works out. Still stuck with fission for now.

Actually as a biologist if you're going to wank any category of weapon it should be psychoactive microbes and advanced neural scanners that finally make the sci-fi staple 'mind probe' practical. With the very rapid progress in scanning resolution and data processing recently, the whole torture debate may be rendered irrelevant by allowing the CIA spooks to just pull names out of the subject's speech centres (after artificially stimulating the relevant bits of LTM until something useful pops out; probably a hellish nightmare for the subject, but no violence involved so it's media-safe). Meanwhile you can use your bioweapon expertise to infect the population of the occupied state with pathogens that mess with their brain chemistry, making them non-agressive and pliant (wandering around in a happy daze, if necessary). All very convenient and (with the right spin) media friendly. Of course nanotech does it better (tm), when it eventually arrives the combination of self-deploying intra-brain transducers and advanced cogsci knowledge will allow specific targets or even allow entire populations to be turned into Dr Who style robomen. CIA orgasms are made of this.

This sounds rather like how the Inhibitors go about gathering intel. on their next target. Find a live specimen of the species targeted for pruning of numbers, then use the shape shifting multi-purpose black cubes of death that make them up to shift down to the microscale and infiltrate the cranium via every orifice in the head. Process the engrams stored in the neural net of the brain, then pull back, leaving behind some machinery if necessary to take control of the specimen and maybe use as an infiltrator. Without saying a word, you've learnt everything that person knows and can use it against them. Course, the Inhibitor machines were also of unknown composition (perfect black cubes of varying size, likely some sort of BEC product immune to physical damage) and powered themselves via vacuum energy and even propelled themselves via something akin to the Casimir effect, only trillions of times more powerful than our science could show.

Did I mention they make flamethrowers our of stars?

[metavac]

As I understand it, bacteria calorimetry generally registers on the order of 1 nW, whereas nanorobotics researchers aim to innovate machines that dissipate an order or two less than that. AV, I honestly don't see how you're checking this fundamental advantage in power in bacteria's column. Am I missing something?

Hmm?

I'm referring to this comment:

Again, synthetic machines require fundamentally more energy to replicate than organics due to their nature

I understand that, for example, fullerenes have a formation enthalpies ranging from on the order -10 to -100 kcal/mol, polypeptides have are roughly an order to three lower than that. A fullerene also has a higher strain energy. Is it really reasonable to assume that nanomachines won't be fabricated from more malleable material?

[lukexcom]

Had it occurred to you that the listed test flights (available on web) didn't all end in nuclear explosions? So what happened to the others? Where did they fly from? What were they fired at?

December 22nd, 1962 seems to be an oft-quoted date on the net like here, or here, with a Zeus passing anywhere between 200m and 20m (depending on the site) to its RV target. After that, 10 of 14 flyby intercepts were successful. OTA-ISC-254 (big PDF file) seems to be referenced quite a bit, although that document's primary scope seems to be an exploratory work for SDI.

But those are just public internet sites. Do you recommend any other books that focus on historic ABM programs quite well with good detail?

[Ariphaos]

The Chinese analysed how the US military works and they have come to the conclusion that its achilles heel is its reliance on space-based sensors and communications facilities. If they can be eliminated, the US ability to use its generations of mass-killingw eapons will be severely impeded. Put simply, no GPS, no guided bombs. No space-based recon, the US doesn't know where to shoot. So, the Chinese see their ASAT technology as a response to the devastating US superiority in precision-guided munitions that essentially negates their mass infantry army.

I'm guessing that China is not utterly ignorant of America's extensive inertial guidance programs and this is more of a 'dominate Asia' move, with respect to India and Japan putting junk up in space.

[Admiral Valdemar]

I understand that, for example, fullerenes have a formation enthalpies ranging from on the order -10 to -100 kcal/mol, polypeptides have are roughly an order to three lower than that. A fullerene also has a higher strain energy. Is it really reasonable to assume that nanomachines won't be fabricated from more malleable material?

I'm referring to getting their materials from the environment e.g. iron in structures for instance. It's one thing to organise carbon that's already free, but it's much harder to start taking apart local materials like that. Organics benefit from enzymes etc. The universal assembler Von Neumann machine would have to pluck the molecules off and move them itself, bonds that are typically stronger than most basic biological ones.

Though I mentioned, this depends on what you're machines are made of, naturally.

[Starglider]

I'm getting wary of your SHODAN influences here.

DO NOT BLASHEME!

What processor is used though? Still electronic

Probably electronic (maybe with spintronics, which is complementary), likely ballistic nanotube or quantum dot based, but rod logic has the advantage of being extremely robust and buildable with a wide range of materials. Superconducting concepts like rapid single flux quantum are great for central processors, but will probably remain impractical for embedded processors in mobile elements.

or going to photonic

Photonics are great for long-range (which will eventually mean anything further than a micrometre) interconnects. There are currently no good designs for optical gates that would make optical logic practical; what designs exist are conceived for use in switching elements where minimising latency is critical, and size and power use are not a concern. I don't know why sci-fi fans are so keen on 'photonics' anyway, I suppose it sounds shiny, but there's nothing wrong with normal electronics (ditto for 'positronics' but more so, and don't even mention the idiocy of 'gravitronics').

or biochemical in parts?

Biochemical computers are superior in only one respect. They are better at denaturing under trivial thermal loads!

Seriously, things like DNA computers are a concept looking for a problem. Maybe, in the short term, the massive molecular level parallelism might make up for the horrible interfacing, serial speed, infrastructure requirements and reliability. But in the long term they are worthless.

There's plenty to be said for holographic storage in crystals

For slow bulk storage, yes. Various other technologies are competing for the crown of best bulk active memory. Register equivalents and final stage cache basically have to be implemented in whatever gate technology the main processor uses.

Well I was wondering what we were talking about. You were using nano- and microbot a lot when I thought we were solely looking at nanoscale technology. I mentioned MEMS before, which are an even larger, but actually proven technology thus far.

Current MEMS don't use materials structured to the atomic level, which is pretty much the definition of nanotech. But the ability to make nanoscale mechanisms doesn't imply any particular optimum size for the independent units. Various applications have different optimums, often a mix of machines at different sizes.

How we fare in engineering anything synthetic at that scale or smaller is a work in progress.

<hopelessly optimistic futurist>You can't stop progreesssss! In the future, everything will be shiny!</hopelessly optimistic futurist> Well, various large scale things may slow down or even stop progress (things coming under the heading of 'existential risks'), but given minimal necessary prerequisites it continues regardless of whether assorted humans want it to. What they can do (i.e. with biotech regulation, pressing military needs, IP regulation or VC funding infusions into whole sectors) is handicap fields so as to change the order in which things get developed.

What about environmental interference with such designs? I expect they're durable enough to brave everything a bacterium can take, for instance, without danger of errors cropping up.

Yes, trivially, and where necessary we can use an indefinite amount of software error correction (parity and compute multiple times compare results) to improve intermittent error tolerance up to the physical limits of the device being able to operate at all.

I'm a bit iffy on these designs so far given their energy density and what some people propose they can then go and do. You're not going to last long with internal fuel stores unless there's a way of collecting more fuel from the environment,

Nanorobots will do better than biological forms because they can use more volatile chemicals and denser fuel sources, as well as wasting less space on inefficient energy carriers and related support structures (i.e. ATP and mitochondria). Ultimately yes energy must be taken from the environment, but I'd note that current solar cells already outperform photosynthesis and the theoretical efficiencies are much higher. Thermocouples and minature heat engines can also exploit very modest heat gradients for trickle charging, and clusters of microbots can dynamically connect and shunt power around (fuel or electrically) as required.

When it gets to designing your own microbe from scratch with introns or other junk genetic material excised,

As I understand it, junk DNA is required to keep the genome stable under copying and to maintain adaptivity. If it wasn't required, organisms without it would have enough of a selective advantage (faster cell reproduction) for it to have disappeared long ago). I may be mistaken on this point though.

Cell signalling is very much an infant science still, with many processes in the human cell still perplexing people to this day. The day we fully understand such metabolic pathways is likely the day we can cure cancer and various other diseases. Hormonal messenging in macro-organisms is fascinating, if not related here.

True. But bandwidth and latency of custom-engineered mechanisms will be much better (another case of diffusive transport sucking), particularly if you use electronics but still true for purely mechanical systems.

For bacteria they don't ever need such systems anyway, which is why they've never needed to evolve them.

'Need' is not something that is meaningful in an evolutionary context; there are no desires, requirements documents or design processes. 'Competitive benefit' is all that exists. Better selective expression mechanisms would in fact be massively useful to bacteria, particularly because they can't easily move to more favourable environments the way larger organisms can, so they have to adapt themselves and their progency in situ. A bacteria that carries the equivalent of one hundred separate species genomes with it and chooses which effective 'species' (or species mix for a community) to become based on a complex analysis of local conditions would be extremely competitive. But there's no incremental path to it and it probably isn't possible using DNA coding density and pathetically the limited compute potential of expression networks, though a cyborg bacteria with custom-designed wet nanotech supplemental genome storage might be able to do this.

Such a computer you're talking about has evolved many a time naturally. You're thinking with it.

Irrelevant. Computational limitations of the human brain compared to a digital computer aside, this isn't relevant when comparing bacteria to nano/microbots. As it happens the entire human brain loses to a rod logic computer the size of a grain of sand, but that's an argument for AGIs kicking (organic) ass and taking names (or indeed whole mind-states), not a direct nanotech-versus-biotech/GE issue.

Bacteria can work exceptionally well in biofilms though, which makes them orders of magnitude more resistant to just about any physical attack.

Which again, engineered dry nanotech does a lot better, due to better adhesion, better co-ordination and adaptive 3 dimensional structuring (driven by forward simulation, not just past history as evolution managed to encode it at a pitiful few bits per generation, tops).

Intelligence is not really something ventured for in this field, why would one want a cell running simulations of something it can do with basic stimuli responses?

In a benign environment with excellent advance design, it isn't necessary. In a hostile enviornment where adaptivity is crucial, it's a major advantage.

[Medic]

to cut overall missile defense funding by more than $1 billion and eliminate funding for several anti-missile systems. These include the Airborne Laser, the Kinetic Energy Interceptor and the Miniature Kill Vehicle programs.

*sigh* All 3 of those? There goes a sizable chunk of the layered-defense against ballistic missiles.

[CC]

Lets look at a current example. Let's suppose that somehow we can look at two worlds. One is ours with the war in Iraq dragging on indefinately and no end in sight (even if we do pull out). The other is one in which the five major cities of Iraq were destroyed by nuclear attack with the result that the war ended on the spot and Iraq is now peaceful, tranquil and has a (insert government of your choice). Which is the "evil choice?"

Are there any moral or ethical systems which state that such a nuclear attack is not evil? After all, you're talking about deliberately killing up to nine million innocent men, women, and children simply for the purpose of a demonstration (unless we grant omniscience to the planner, with infallible knowledge that Iraq would turn peaceable). The vast majority of such systems that I am aware of would consider that to be an evil act no matter what the end result, intended or otherwise. Taking the UN's 2006 figure of 34,452 violent deaths, equaling that would require the insurgency to remain roughly as violent as it currently is for the next two and a half centuries, which knocks it out of the running with any of the other ideas. So really, are there any moral or ethical systems in which the evil choice is blatantly obvious because I can't think of a one?

[The Grim Squeaker]

Lets look at a current example. Let's suppose that somehow we can look at two worlds. One is ours with the war in Iraq dragging on indefinately and no end in sight (even if we do pull out). The other is one in which the five major cities of Iraq were destroyed by nuclear attack with the result that the war ended on the spot and Iraq is now peaceful, tranquil and has a (insert government of your choice). Which is the "evil choice?"

Are there any moral or ethical systems which state that such a nuclear attack is not evil?

Objective humanitarianism would have as minimize the injuries inflicted (Or for the greater good, most Humanistic moral systems tend to emphasize "good of the many").

After all, you're talking about deliberately killing up to nine million innocent men, women, and children

Provide the population of the cities Please?

simply for the purpose of a demonstration (unless we grant omniscience to the planner, with infallible knowledge that Iraq would turn peaceable).

What's your basis for the destruction in shock and awe tactics of a great part of Iraq's power, including Sadham not crippling it? (Consider with what ease America defeated the conventional power structure and army there, insurgents aside).

The vast majority of such systems that I am aware of would consider that to be an evil act no matter what the end result, intended or otherwise.

Except that the end result would result in less deaths, greater stability, an achievement of goals...

Taking the UN's 2006 figure of 34,452 violent deaths,

The studies had it as in the area of ~600,000 (With 3-4 times as many injured).

equaling that would require the insurgency to remain roughly as violent as it currently is for the next two and a half centuries, which knocks it out of the running with any of the other ideas.

Or to get more violent once the religious/ethnic cleansing starts seriously as soon as the "peacekeeping" forces leave (Unless you think those deathsquads are just for fun).

So really, are there any moral or ethical systems in which the evil choice is blatantly obvious because I can't think of a one?

One which looks at the consequences and weighs them up.