Are you being sarcastic? Those "magic batteries" can be purchased today. They do recharge in 10 minutes and would have a 300 mile range if you could afford enough of them in a pack. I'm referring to some of the high end cells by A123 systems, there are other competitors.
Self driving cars on the highway are technically possible, and in fact could have been done decades ago if the highway system were built to ONLY allows self driving cars on it.
We do have a lot of natural gas left it appears, albeit with some environmental risks, and as time passes solar IS getting cheaper. Dramatically so. (dunno about wind, while you can optimize manufacturing you are still build and installing an enormous machine. Solar is small modules that can get cheaper and cheaper)
If you had real 3d printing that was accurate to the molecular level, you could print out nuclear fission reactors to provide enough energy to run all this. Or enough solar panels to cover entire deserts. However, private industry will not be able to develop this technology : it would take an enormous investment in time and resources with no guarantee of success to develop molecular manufacturing.
What is wrong with doing this? China isn't banning knowledge about useless majors, it's simply declining to pay people to study majors that don't train people to be contributing members to society.
The USA should absolutely do the same. We need more engineers and less psychology majors.
Here's a simple and reasonable explanation for this shift.
The reason for young male scientists making their big breakthroughs before age 30 probably is caused by hormonal levels (they work extremely hard to create a novel solution to a problem in order to attract a mate) and possibly some brain aging. The brain is most likely a bit more plastic and higher performance between age 20 and 30 than it is between age 40 and 50.
HOWEVER, what has happened is that a stroke of brilliance is no longer sufficient. All the easy pickings in physics have already been found. Now, the significant discoveries are much more complex endeavors, requiring far more knowledge and experience before someone could even be in a situation to make one. Just like how major inventions can't really happen in garages anymore. (sure, you can hack something together in a garage with Arduino boards...but you won't have made anything that hasn't already been prototyped in lots of places elsewhere) Contrast the present day with, say, the Wright Brothers building a powered aircraft with only limited resources. Today to make spacecraft able to take a man to Mars you'd need the resources of entire country.
So, yes, I think that physicists that age probably become less effective due to aging, but due to more knowledge and experience and resources they became able to make these big discoveries AT ALL.
Well, you didn't have to insult me if you knew what I was talking about. As I just mentioned, the one time pad is a removable module. If you ever run out of keyspace, the avionics in the drone will refuse to allow a flight (they will keep functioning until the end of the current flight by looping back to the beginning of the pad but no further) until the module containing the pad is replaced.
And just think about it for a minute. How large are these control packets going to be? A few hundred bytes each? The major user of bandwidth for a drone is going to be the video streams, which can be encrypted with more conventional public/private key methods. While it would be entirely practical to give a 10 million dollar drone a 1 terrabyte or more long pad, exceeding any plausible lifespan for the drone.
The key to use the drone is a module with the drone's serial number that is stored in a vault. I sort of imagine it made of red plastic for the module that allows the drone to use it's weapon systems. Any command console can accept the key to any drone. Perhaps there might be two-three identical keys per drone.
These drone aircraft are at least 10 million dollars each. None of the things I have described are all that difficult or expensive to implement, and would use commodity parts. It wouldn't add a significant amount to the cost.
1. A one time pad is a a data file that stores a string a random numbers taken from a hardware random number generator. At a central location, exactly 2 devices are created with the same data file. Each "device" has a microcontroller front end that blocks the pad from ever being copied off of the device, written onto the same silicon wafer as the portion containing the flash. One device is installed inside the drone, the other one is the key to access the drone and is inserted into the control console when you are ready to use the drone. All control data packets are XORed against this pad when they leave the control console and XORed again when they are received by the drone. Only valid packets that match an md5 of the same packet that is appended at the end of the packet are executed by the drone.
If you are at all knowledgable about computing, you will see how this both makes interception and decryption of the control packets impossible as well as ensures that no one can insert unauthorized commands with more than an infitesimal chance of success (without physical access to one or other pad and the technical means to copy them, which is made as difficult as possible).
2. If the control system is run on read-only media, and the part that handles uploads of data to external drives is a separate computer system without a 2 way communication link to the control system, then uploading a virus is by definition impossible as there is no path for such an upload to take.
3. A hardware filter is a microcontroller or small embedded computer running a separate OS from the main computer that inspects each data packet coming and and leaving and ensures that it meets certain criteria. (such as no strings longer than the allowed buffer size, etc valid checksums, etc)
Ok, so you get some interns in a room and ask them to draw on the whiteboard the things to consider when designing a remote controlled killer robot.
What do you suppose the FIRST thing any intern is going to write up there in terms of things you need to worry about?
Make SURE the enemy can't hack your robots and turn them against you!
Well, when you start writing up how to accomplish that, you would want
1. A completely secure system for authenticating commands sent from the control system. The only form of encryption that is completely secure is one time pad.
2. NO POSSIBLE WAY for someone to load viruses or gain access to the control system!!! That means NO network access to anything but the systems that send and receive signals from the drone! And one heck of a hardware filter on those information packets!
Yes but if you publish the code + proofs, and the mathematical analysis you used to formulate the proofs is flawed, and an attacker is able to see that but others aren't...Then you have just given him or her the means to break in.
Same goes for encryption. You can't generally crack an encryption algorithm, even a flawed one, if you only have the encrypted data and plaintext but no idea at all what algorithm was used.
But the OS in projects like that was probably a one-off written JUST for the application. And the software probably won't RESPOND to most packets, nor support modern networking methods. It's one thing if a true hacker who knows everything had something to work with. But if he doesn't know what computer it is he is trying to hack into is using, and even if he did it he wouldn't be able to find any information about how it works, being a one-off project with the books being top secret...
I am not saying that such security is perfect, merely that in this example there might BE serious security holes that no one would ever be able to exploit since they wouldn't have any experience with the OS in question.
What about true obscurity. What kind of OS or software runs on the computers in a nuclear missile silo? Do those computers even use an OS? The point is, with little or nothing published, an attacker who was able to access systems like those would have little realistic hope of hacking them. There's no 0 day lists, no marketplace to pick up working cracks, no books describing how the internals of such a system.
With the Times Square bomber, he also had a half-assed ANFO bomb in the car as well. Why half-assed? Because he didn't follow even the most basic and logical testing routine. If you want to make homemade explosives that you are not sure will go off, you need to
1. Make SURE you have the right ingredients (he had the wrong kind of Fertilizer because apparently he can't read labels)
2. TEST your concoction, with a large enough sample to be sure it works.
Oh, and multiple detonators are probably also a good idea. Don't just rely on a single fuse that might go out.
The various 'terrorists attempts' over the last 10 years have been so pathetic they make me angry. (because we are spending trillions of dollars to fight enemies who are so incompetent they couldn't shoot someone in the back without blowing off their own foot)
How would you destroy something like that? A machine like this used right would be a more powerful weapon than an arsenal of ICBMs. And the power of exponential growth means that the 'powers that be' would have to destroy ALL copies and the blueprints for making them before at least one copy spreads to a place that they can't get to. Not to mention all someone has to do is wikileaks the blueprints and no one will ever be able to squash the technology then.
What if we determine that based upon all verifiable evidence, "my brain made me do it" is an difficult to dispute fact. There's an objective reality out there independent of what humans think it is, and it might in fact be that in that reality there is no free will. Whether or not you or I or anyone else wish there were.
I think the space elevator is a horrible engineering idea. I am not impressed by space elevator cultists. You are talking about thousands of miles of cable and that the entire system will fail if a single piece breaks anyone. Also, you beam all of the energy needed to reach orbit from the ground to the climber using lasers. The climber takes days to make 1 trip, preventing the elevator from being used for anything else during that time. A single missile, a single hit from a high energy weapon, and you lose 100% of your space elevator. This makes the correct answer freaking obvious : scale up your lasers a couple of orders of magnitude and blast off from the ground directly with their energy.
Africa : as I said, if we had the technology to force the warlords to disappear, we could stop it. We do not have that kind of technology : with today's military, we would have to take hundreds of thousands of soldiers killed in all the wars, more money than the USA can afford at all (even our big creditors would probably stop lending to the U.S. if we did something this stupid), and we would not gain enough control over africa to really prevent the deaths and murders much. (look at Iraq : more people died in the resulting disruption than died under Sadaam)
Obviously, with exponentially replicating manufacturing, we could make enough remotely controlled semi-automated soldier bots, armed with less than lethal weapons (no reason to kill the enemy if they cannot do more than force you to recycle a robot) to invade every country at once.
Slavery : the reason the North one was they had overwhelming advantages from the beginning in production and manpower. The South were idiots for even fighting.
It's a myth that hyper expensive medical care is more effective. The gains are extremely small, and in the USA as a whole, lifespans are actually shorter both due to misdistribution of resources (large numbers of people get almost no medical care at all) and poor lifestyles for the majority of the population.
Fossil fuels : we can't afford nuclear reactors not JUST because of the greens, but also because the manufacturing cost to build reactors is prohibitive. Problem, meet solution : if we could print atomically near perfect reactor parts in mass quantities, nuclear power plants would be cheap. And we could replace every part used in a reactor right on schedule rather than running the same facility for 50 years.
I'd like to hear more about your ideas. What would we be able to accomplish if we had the "political and social will" to do so. My assumption about our reality is that human nature is unchanging because evolution is a glacially slow process. The reason why human societies change at all is driven by the new rules created by technology. The world we have now is better than before : longer lifespans, more toys, etc only because of technology. Even social reforms : elimination of slavery/civil rights were driven by technology.
Thus, I assume that given this hypothesis, long standing, near intractable problems that exist now : Africa, no real space travel, medical care too expensive and ineffective, running out of fossil fuels, climate change, and so forth can only be fixed if we substantially surpass our current technology.
Small incremental improvements to what we have are not going to be a sufficient driver of change. We'd need a huge breakthrough on the scale of atomic energy or greater to really make a dent.
? It need not be energy efficient or quick, relatively speaking. Even a very slow machine that consumed a lot of energy would be incredibly useful. Precision has to be high enough for self replication to succeed at least some of the time, and the machine by definition IS flawless since every atom in the machine would have a purpose.
With limitations on the "universal" part. Aka the constructor takes a long time, is extremely slow for anything but thin and flat objects, uses a lot of energy, and can't work for certain molecular structures that are too unstable during construction. (although it could probably build another variant of itself that COULD handle special cases)
And said machine would only work on chemically different elements : it would not be able to distinguish between isotopes. So if you wanted to make a nuclear weapon from pitchblende you'd still need to do isotope separation first.
Oh, and don't forget it only makes solid objects and they must be made at cryogenic temperatures. Could you make food? Yes, but it better handle thawing well...
I'm not Kurzweil. I don't think that his desired scenario is at all inevitable, and technical challenges may mean that it takes centuries to come about. (although once the full Singularity really started i don't see why it wouldn't reach the end-game conditions within a decade or two. )
And for fixing Africa and third world governments : the fix is to print millions of remotely controlled drone soldiers and invade and depose every government in the third world. Possibly at the same time. The only way you can stop the genocides and the mass starvations is to arrest or kill the people responsible for them.
Once that is done, you'd create vast, automated police states. Total surveillance everywhere, the police use robots heavily, everyone is being watched by software 24/7. Basic food and medicine would be extremely cheap.
One distopic possibility is that these societies truly could become prisons for all. With so many put out of work (since there would be very little need for factory workers, only designers of new equipment which only a small percentage of the population at any given time have the talent for) you might create vast bureaucracies and make a large percentage of the population cops to watch everyone else. If 20% of the population were cops, you really could have the 1984 scenario where everyone IS being watched all of the time. (when I say "watched", I mean there would be a person employed to watch every action taken by 5-10 other people who would know everything there is to know about them. It's not just that everything would be recorded by surveillance...someone would be checking the tapes)
But by eliminating most freedom, you would also eliminate most crime. At least crimes committed by the unwashed masses... Important people would have special privileges so that no one could surveil them.
Molecular manufacturing technology may be decades away, but things like this are enheartening to read about. Out of all of the possible technologies we might develop in the foreseeable future, molecular manufacturing nanotechnology is the most promising. (that we know are possible within the laws of physics...antigravity or free energy would be nice but we don't know of any physical principles that would allow them)
Essentially, "all" we have to do is develop a nanoscale machine that is made of gears and motor systems like this one and has sensors and electronics packages. It has to duplicate the functions of a 3d printer under extremely controlled conditions.
Limitations on the machine : conditions will need to be as predictable as humans can make them. That means cryogenic temperatures, a vacuum chamber, a steady and consistent power source, and a steady supply of completely pure feedstock to work with.
The machine's function would be to place a single atom in one of several possible locations, and/or to stabilize a structure with some kind of atomic clamp that injects or removes electric charge. Each machine would probably only be able to work in a single case...say a carbon atom in a single bonding scenario for one machine.
You'd build arrays of these machines, and with a few hundred variants of the machine (each one only slightly different than the others) you'd have a complete printing system able to print nearly any structure you have the atomic bonding map for, including COPIES OF THEMSELVES.
That last bit is everything. Nanotechnology is merely a hyper-expensive way to make high end electronics and other expensive items without self replication.
WITH it, the sky is the limit. With self replication, we could very soon make huge arrays of these 3d printers, big flat plates with trillions and trillions of individual, identical subunits. These machines could gradually produce, inside big vacuum chambers and at cryogenic temperatures, almost anything you have the resources and means to make.
All those kids who claim they want to help the unfortunate in Africa? We need this kind of technology to really make a dent in the world's problems. Anyone want to go to outer space? The only real way we could ever make rocket rides cheap enough for the average man is if building high end spacecraft was as easy as printing out the parts, with near complete automation (and the parts would be atomically near-perfect, eliminating the need for most quality control)
We could even use disposable rockets this way...just send out a tractor to pick up the spent stages, melt down the metals in a plasma furnace to separate the different elements, and reform the atomic feedstock you need to print out new spacecraft.
Want artificial intelligence? Ain't going to happen with today's software methods nor today's neuroscience. But if you could look at atomically perfect scans of a perfectly preserved human brain (through careful cryogenic freezing and fixation) you could actually steal the firmware of human intelligence right from the hardware. With automated tools, you'd convert neural maps of human beings you KNOW were sentient (before they died) and emulate them on molecular computing circuitry. It probably would still be an incredible challenge, but with these kinds of tools I think working AI would be merely a matter of time.
Tired of being born, growing up, enjoying a brief period of good health and sexual function, and then gradually declining decade after decade until death? In the long run, this same technology could be used to repair human bodies or even eliminate the need for them entirely.
Current spacesuits take 2 hours to even put on. Even assuming they fix that problem for moon suits, who would risk an EVA (and go through the process of suiting up to go to the restraunt) just to get a meal? Especially for Domino's quality pizza...
And you'd probably track moon dust into the restraunt which would cause other problems.
And that restraunt is on the surface of the moon, with windows no less.
A much better design would be to have the pizza kitchen as an adjunct to the normal cafeteria, underground, as part of the main lunar base. And hiring an employee and sending them to the moon is a terrible idea. For 21 billion dollars, you could easily fit a couple of billion worth of robotics R&D to develop a reasonably reliable robot to prepare the pizzas. The robot would be teleoperated from the ground.
Despite what I said, I skimmed your article. I do agree that bulk processing is more efficient than atom-by-atom processing. And a plasma centrifuge or something similar would make for a good universal separation step to prepare the feedstocks for the molecular printer arrays. (the concept I have in mind...and remember this is a specific implementation of molecular manufacturing that may or may not be practical is flat, extremely thin plates that are arrays of trillions of individual 'print head' subunits, with rotary mechanisms for positioning groups of a few hundred heads and internal pathways for transporting feedstocks to the needed locations)
Each plate would of course be able to print a clone of itself. Eventually, low probability events (damage from radiation, random chance, etc) would jam pieces of the machinery and a plate would wear out to the point that it would need replacement.
One key limitation is this technology would be extremely good for making huge amounts of very thin objects with molecular precision but the 'print times' for a thick object would become extremely long. Eric Drexler estimates that it would take a year in the best case to produce a meter - thick object.
So you'd have to build smaller, very thin, subunits in mass and assemble big things out of little pieces - or be prepared to wait a long time.
I don't think you understand the power of exponential growth.
Bacteria already do exactly what I am describing. The simplest bacteria use about 2,800 mechanical parts, some of which are fiendishly complex. Those bacteriums can self replicate in about 40 minutes if supplied with ONLY : 1. Sugar. (energy source and feedstock) 2. Water and C02 and oxygen (feedstocks and solvent) and 3. trace amounts of other elements they need (iron, magnesium, etc)
In a week, given unlimited resources, the bacteria would out-mass the planet.
For various technical reasons, the machinery of life are not as fast nor as efficient as they theoretically could be. If we build machinery with internal parts made of graphene or diamond, each part rationally optimized for maximal performance, we could in theory make bacteria seem slow.
So yes, the first molecular printer would be pathetically slow. But if it could print itself in even a week we'd have enough of them to cover entire planets in a year.
I'm not really interested in reading your diaries given your obvious ignorance.
Problem : to manufacture everything we use right now requires a vast and complex infrastructure. To produce any high tech good requires a complex process with thousands of steps with many steps unique to that particular good.
Solution : molecular manufacturing. Living creatures are far more complex internally than the most complicated machine humans have built to date. Yet, it is possible for living creatures to produce every single component they use, atom by atom, when supplied with nothing more than energy and the raw elements they need in a suitable feedstock. (the fact that larger creatures such as mammals lose the ability to be self sufficient like that is a limitation of evolution as a design algorithm and not a technical limit)
Molecular manufacturing is a technology that, through some technical means (there are hundreds of possible solutions) allows us to produce products with control over the bonding of all individual atoms in the product. By definition, since the machinery that performs molecular manufacturing is made of individual atoms bonded a particular way, said machinery will be able to manufacture a new copy of itself, leading to exponential growth.
The most common concept for this technology would be a 3d printer that is extremely small and can place individual atoms. It is accepted by all credible scientists that biochemistry already does this, so said printers are at least theoretically possible. (some have argued that we can't make a printer like this work without a suitable solvent)
Anyways, once we have this technology, space colonization would become quite simple. We would not even attempt to terraform planets - doing so would deprive us of the valuable elements located inside the celestial bodies. We would instead land a few probes equipped with high speed molecular manufacturing equipment, some ore gathering robots, and a power source. Those probes would produce more copies of the robots, the power plants, and the molecular printers which would in turn...
Once you have an entire surface covered with machinery, you order it to produce something to put cargo into orbit (a linear electromagnetic accelerator, or a laser launch array, or a few million space elevator tethers, or whatever is appropriate) and you throw your waste and your finished products into space. In space, you use the same technology to produce space habitats for humans.
Assuming you still have human beings, that is. This same technology would make many other things possible.
Slashdot Mirror
Wait a second.
Option 1 : simply leave enormous numbers of people unemployed. Society gets 0 benefit from those people.
Option 2 : give those unemployed money to live on so they don't die. Same thing.
Option 3 : pay those people to build a high speed train. Society gets a high speed train, which provides some benefit.
Option 4 : pay those people to fight in a war and die. Society gets no benefit and enrages people of other countries who may fight back.
Are you being sarcastic? Those "magic batteries" can be purchased today. They do recharge in 10 minutes and would have a 300 mile range if you could afford enough of them in a pack. I'm referring to some of the high end cells by A123 systems, there are other competitors.
Self driving cars on the highway are technically possible, and in fact could have been done decades ago if the highway system were built to ONLY allows self driving cars on it.
We do have a lot of natural gas left it appears, albeit with some environmental risks, and as time passes solar IS getting cheaper. Dramatically so. (dunno about wind, while you can optimize manufacturing you are still build and installing an enormous machine. Solar is small modules that can get cheaper and cheaper)
If you had real 3d printing that was accurate to the molecular level, you could print out nuclear fission reactors to provide enough energy to run all this. Or enough solar panels to cover entire deserts. However, private industry will not be able to develop this technology : it would take an enormous investment in time and resources with no guarantee of success to develop molecular manufacturing.
What is wrong with doing this? China isn't banning knowledge about useless majors, it's simply declining to pay people to study majors that don't train people to be contributing members to society.
The USA should absolutely do the same. We need more engineers and less psychology majors.
Here's a simple and reasonable explanation for this shift.
The reason for young male scientists making their big breakthroughs before age 30 probably is caused by hormonal levels (they work extremely hard to create a novel solution to a problem in order to attract a mate) and possibly some brain aging. The brain is most likely a bit more plastic and higher performance between age 20 and 30 than it is between age 40 and 50.
HOWEVER, what has happened is that a stroke of brilliance is no longer sufficient. All the easy pickings in physics have already been found. Now, the significant discoveries are much more complex endeavors, requiring far more knowledge and experience before someone could even be in a situation to make one. Just like how major inventions can't really happen in garages anymore. (sure, you can hack something together in a garage with Arduino boards...but you won't have made anything that hasn't already been prototyped in lots of places elsewhere) Contrast the present day with, say, the Wright Brothers building a powered aircraft with only limited resources. Today to make spacecraft able to take a man to Mars you'd need the resources of entire country.
So, yes, I think that physicists that age probably become less effective due to aging, but due to more knowledge and experience and resources they became able to make these big discoveries AT ALL.
Well, you didn't have to insult me if you knew what I was talking about. As I just mentioned, the one time pad is a removable module. If you ever run out of keyspace, the avionics in the drone will refuse to allow a flight (they will keep functioning until the end of the current flight by looping back to the beginning of the pad but no further) until the module containing the pad is replaced.
And just think about it for a minute. How large are these control packets going to be? A few hundred bytes each? The major user of bandwidth for a drone is going to be the video streams, which can be encrypted with more conventional public/private key methods. While it would be entirely practical to give a 10 million dollar drone a 1 terrabyte or more long pad, exceeding any plausible lifespan for the drone.
The key to use the drone is a module with the drone's serial number that is stored in a vault. I sort of imagine it made of red plastic for the module that allows the drone to use it's weapon systems. Any command console can accept the key to any drone. Perhaps there might be two-three identical keys per drone.
These drone aircraft are at least 10 million dollars each. None of the things I have described are all that difficult or expensive to implement, and would use commodity parts. It wouldn't add a significant amount to the cost.
You obviously have no idea what these things are.
1. A one time pad is a a data file that stores a string a random numbers taken from a hardware random number generator. At a central location, exactly 2 devices are created with the same data file. Each "device" has a microcontroller front end that blocks the pad from ever being copied off of the device, written onto the same silicon wafer as the portion containing the flash. One device is installed inside the drone, the other one is the key to access the drone and is inserted into the control console when you are ready to use the drone. All control data packets are XORed against this pad when they leave the control console and XORed again when they are received by the drone. Only valid packets that match an md5 of the same packet that is appended at the end of the packet are executed by the drone.
If you are at all knowledgable about computing, you will see how this both makes interception and decryption of the control packets impossible as well as ensures that no one can insert unauthorized commands with more than an infitesimal chance of success (without physical access to one or other pad and the technical means to copy them, which is made as difficult as possible).
2. If the control system is run on read-only media, and the part that handles uploads of data to external drives is a separate computer system without a 2 way communication link to the control system, then uploading a virus is by definition impossible as there is no path for such an upload to take.
3. A hardware filter is a microcontroller or small embedded computer running a separate OS from the main computer that inspects each data packet coming and and leaving and ensures that it meets certain criteria. (such as no strings longer than the allowed buffer size, etc valid checksums, etc)
See for yourself
http://www.psychologytoday.com/blog/the-scientific-fundamentalist/201103/criminals-look-different-noncriminals
Ok, so you get some interns in a room and ask them to draw on the whiteboard the things to consider when designing a remote controlled killer robot.
What do you suppose the FIRST thing any intern is going to write up there in terms of things you need to worry about?
Make SURE the enemy can't hack your robots and turn them against you!
Well, when you start writing up how to accomplish that, you would want
1. A completely secure system for authenticating commands sent from the control system. The only form of encryption that is completely secure is one time pad.
2. NO POSSIBLE WAY for someone to load viruses or gain access to the control system!!! That means NO network access to anything but the systems that send and receive signals from the drone! And one heck of a hardware filter on those information packets!
Yes but if you publish the code + proofs, and the mathematical analysis you used to formulate the proofs is flawed, and an attacker is able to see that but others aren't...Then you have just given him or her the means to break in.
Same goes for encryption. You can't generally crack an encryption algorithm, even a flawed one, if you only have the encrypted data and plaintext but no idea at all what algorithm was used.
But the OS in projects like that was probably a one-off written JUST for the application. And the software probably won't RESPOND to most packets, nor support modern networking methods. It's one thing if a true hacker who knows everything had something to work with. But if he doesn't know what computer it is he is trying to hack into is using, and even if he did it he wouldn't be able to find any information about how it works, being a one-off project with the books being top secret...
I am not saying that such security is perfect, merely that in this example there might BE serious security holes that no one would ever be able to exploit since they wouldn't have any experience with the OS in question.
What about true obscurity. What kind of OS or software runs on the computers in a nuclear missile silo? Do those computers even use an OS? The point is, with little or nothing published, an attacker who was able to access systems like those would have little realistic hope of hacking them. There's no 0 day lists, no marketplace to pick up working cracks, no books describing how the internals of such a system.
With the Times Square bomber, he also had a half-assed ANFO bomb in the car as well. Why half-assed? Because he didn't follow even the most basic and logical testing routine. If you want to make homemade explosives that you are not sure will go off, you need to
1. Make SURE you have the right ingredients (he had the wrong kind of Fertilizer because apparently he can't read labels)
2. TEST your concoction, with a large enough sample to be sure it works.
Oh, and multiple detonators are probably also a good idea. Don't just rely on a single fuse that might go out.
The various 'terrorists attempts' over the last 10 years have been so pathetic they make me angry. (because we are spending trillions of dollars to fight enemies who are so incompetent they couldn't shoot someone in the back without blowing off their own foot)
How would you destroy something like that? A machine like this used right would be a more powerful weapon than an arsenal of ICBMs. And the power of exponential growth means that the 'powers that be' would have to destroy ALL copies and the blueprints for making them before at least one copy spreads to a place that they can't get to. Not to mention all someone has to do is wikileaks the blueprints and no one will ever be able to squash the technology then.
What if we determine that based upon all verifiable evidence, "my brain made me do it" is an difficult to dispute fact. There's an objective reality out there independent of what humans think it is, and it might in fact be that in that reality there is no free will. Whether or not you or I or anyone else wish there were.
I think the space elevator is a horrible engineering idea. I am not impressed by space elevator cultists. You are talking about thousands of miles of cable and that the entire system will fail if a single piece breaks anyone. Also, you beam all of the energy needed to reach orbit from the ground to the climber using lasers. The climber takes days to make 1 trip, preventing the elevator from being used for anything else during that time. A single missile, a single hit from a high energy weapon, and you lose 100% of your space elevator. This makes the correct answer freaking obvious : scale up your lasers a couple of orders of magnitude and blast off from the ground directly with their energy.
Africa : as I said, if we had the technology to force the warlords to disappear, we could stop it. We do not have that kind of technology : with today's military, we would have to take hundreds of thousands of soldiers killed in all the wars, more money than the USA can afford at all (even our big creditors would probably stop lending to the U.S. if we did something this stupid), and we would not gain enough control over africa to really prevent the deaths and murders much. (look at Iraq : more people died in the resulting disruption than died under Sadaam)
Obviously, with exponentially replicating manufacturing, we could make enough remotely controlled semi-automated soldier bots, armed with less than lethal weapons (no reason to kill the enemy if they cannot do more than force you to recycle a robot) to invade every country at once.
Slavery : the reason the North one was they had overwhelming advantages from the beginning in production and manpower. The South were idiots for even fighting.
It's a myth that hyper expensive medical care is more effective. The gains are extremely small, and in the USA as a whole, lifespans are actually shorter both due to misdistribution of resources (large numbers of people get almost no medical care at all) and poor lifestyles for the majority of the population.
Fossil fuels : we can't afford nuclear reactors not JUST because of the greens, but also because the manufacturing cost to build reactors is prohibitive. Problem, meet solution : if we could print atomically near perfect reactor parts in mass quantities, nuclear power plants would be cheap. And we could replace every part used in a reactor right on schedule rather than running the same facility for 50 years.
I'd like to hear more about your ideas. What would we be able to accomplish if we had the "political and social will" to do so. My assumption about our reality is that human nature is unchanging because evolution is a glacially slow process. The reason why human societies change at all is driven by the new rules created by technology. The world we have now is better than before : longer lifespans, more toys, etc only because of technology. Even social reforms : elimination of slavery/civil rights were driven by technology.
Thus, I assume that given this hypothesis, long standing, near intractable problems that exist now : Africa, no real space travel, medical care too expensive and ineffective, running out of fossil fuels, climate change, and so forth can only be fixed if we substantially surpass our current technology.
Small incremental improvements to what we have are not going to be a sufficient driver of change. We'd need a huge breakthrough on the scale of atomic energy or greater to really make a dent.
? It need not be energy efficient or quick, relatively speaking. Even a very slow machine that consumed a lot of energy would be incredibly useful. Precision has to be high enough for self replication to succeed at least some of the time, and the machine by definition IS flawless since every atom in the machine would have a purpose.
With limitations on the "universal" part. Aka the constructor takes a long time, is extremely slow for anything but thin and flat objects, uses a lot of energy, and can't work for certain molecular structures that are too unstable during construction. (although it could probably build another variant of itself that COULD handle special cases)
And said machine would only work on chemically different elements : it would not be able to distinguish between isotopes. So if you wanted to make a nuclear weapon from pitchblende you'd still need to do isotope separation first.
Oh, and don't forget it only makes solid objects and they must be made at cryogenic temperatures. Could you make food? Yes, but it better handle thawing well...
I'm not Kurzweil. I don't think that his desired scenario is at all inevitable, and technical challenges may mean that it takes centuries to come about. (although once the full Singularity really started i don't see why it wouldn't reach the end-game conditions within a decade or two. )
And for fixing Africa and third world governments : the fix is to print millions of remotely controlled drone soldiers and invade and depose every government in the third world. Possibly at the same time. The only way you can stop the genocides and the mass starvations is to arrest or kill the people responsible for them.
Once that is done, you'd create vast, automated police states. Total surveillance everywhere, the police use robots heavily, everyone is being watched by software 24/7. Basic food and medicine would be extremely cheap.
One distopic possibility is that these societies truly could become prisons for all. With so many put out of work (since there would be very little need for factory workers, only designers of new equipment which only a small percentage of the population at any given time have the talent for) you might create vast bureaucracies and make a large percentage of the population cops to watch everyone else. If 20% of the population were cops, you really could have the 1984 scenario where everyone IS being watched all of the time. (when I say "watched", I mean there would be a person employed to watch every action taken by 5-10 other people who would know everything there is to know about them. It's not just that everything would be recorded by surveillance...someone would be checking the tapes)
But by eliminating most freedom, you would also eliminate most crime. At least crimes committed by the unwashed masses... Important people would have special privileges so that no one could surveil them.
Molecular manufacturing technology may be decades away, but things like this are enheartening to read about. Out of all of the possible technologies we might develop in the foreseeable future, molecular manufacturing nanotechnology is the most promising. (that we know are possible within the laws of physics...antigravity or free energy would be nice but we don't know of any physical principles that would allow them)
Essentially, "all" we have to do is develop a nanoscale machine that is made of gears and motor systems like this one and has sensors and electronics packages. It has to duplicate the functions of a 3d printer under extremely controlled conditions.
Limitations on the machine : conditions will need to be as predictable as humans can make them. That means cryogenic temperatures, a vacuum chamber, a steady and consistent power source, and a steady supply of completely pure feedstock to work with.
The machine's function would be to place a single atom in one of several possible locations, and/or to stabilize a structure with some kind of atomic clamp that injects or removes electric charge. Each machine would probably only be able to work in a single case...say a carbon atom in a single bonding scenario for one machine.
You'd build arrays of these machines, and with a few hundred variants of the machine (each one only slightly different than the others) you'd have a complete printing system able to print nearly any structure you have the atomic bonding map for, including COPIES OF THEMSELVES.
That last bit is everything. Nanotechnology is merely a hyper-expensive way to make high end electronics and other expensive items without self replication.
WITH it, the sky is the limit. With self replication, we could very soon make huge arrays of these 3d printers, big flat plates with trillions and trillions of individual, identical subunits. These machines could gradually produce, inside big vacuum chambers and at cryogenic temperatures, almost anything you have the resources and means to make.
All those kids who claim they want to help the unfortunate in Africa? We need this kind of technology to really make a dent in the world's problems. Anyone want to go to outer space? The only real way we could ever make rocket rides cheap enough for the average man is if building high end spacecraft was as easy as printing out the parts, with near complete automation (and the parts would be atomically near-perfect, eliminating the need for most quality control)
We could even use disposable rockets this way...just send out a tractor to pick up the spent stages, melt down the metals in a plasma furnace to separate the different elements, and reform the atomic feedstock you need to print out new spacecraft.
Want artificial intelligence? Ain't going to happen with today's software methods nor today's neuroscience. But if you could look at atomically perfect scans of a perfectly preserved human brain (through careful cryogenic freezing and fixation) you could actually steal the firmware of human intelligence right from the hardware. With automated tools, you'd convert neural maps of human beings you KNOW were sentient (before they died) and emulate them on molecular computing circuitry. It probably would still be an incredible challenge, but with these kinds of tools I think working AI would be merely a matter of time.
Tired of being born, growing up, enjoying a brief period of good health and sexual function, and then gradually declining decade after decade until death? In the long run, this same technology could be used to repair human bodies or even eliminate the need for them entirely.
Current spacesuits take 2 hours to even put on. Even assuming they fix that problem for moon suits, who would risk an EVA (and go through the process of suiting up to go to the restraunt) just to get a meal? Especially for Domino's quality pizza...
And you'd probably track moon dust into the restraunt which would cause other problems.
And that restraunt is on the surface of the moon, with windows no less.
A much better design would be to have the pizza kitchen as an adjunct to the normal cafeteria, underground, as part of the main lunar base. And hiring an employee and sending them to the moon is a terrible idea. For 21 billion dollars, you could easily fit a couple of billion worth of robotics R&D to develop a reasonably reliable robot to prepare the pizzas. The robot would be teleoperated from the ground.
Despite what I said, I skimmed your article. I do agree that bulk processing is more efficient than atom-by-atom processing. And a plasma centrifuge or something similar would make for a good universal separation step to prepare the feedstocks for the molecular printer arrays. (the concept I have in mind...and remember this is a specific implementation of molecular manufacturing that may or may not be practical is flat, extremely thin plates that are arrays of trillions of individual 'print head' subunits, with rotary mechanisms for positioning groups of a few hundred heads and internal pathways for transporting feedstocks to the needed locations)
Each plate would of course be able to print a clone of itself. Eventually, low probability events (damage from radiation, random chance, etc) would jam pieces of the machinery and a plate would wear out to the point that it would need replacement.
One key limitation is this technology would be extremely good for making huge amounts of very thin objects with molecular precision but the 'print times' for a thick object would become extremely long. Eric Drexler estimates that it would take a year in the best case to produce a meter - thick object.
So you'd have to build smaller, very thin, subunits in mass and assemble big things out of little pieces - or be prepared to wait a long time.
I don't think you understand the power of exponential growth.
Bacteria already do exactly what I am describing. The simplest bacteria use about 2,800 mechanical parts, some of which are fiendishly complex. Those bacteriums can self replicate in about 40 minutes if supplied with ONLY : 1. Sugar. (energy source and feedstock) 2. Water and C02 and oxygen (feedstocks and solvent) and 3. trace amounts of other elements they need (iron, magnesium, etc)
In a week, given unlimited resources, the bacteria would out-mass the planet.
For various technical reasons, the machinery of life are not as fast nor as efficient as they theoretically could be. If we build machinery with internal parts made of graphene or diamond, each part rationally optimized for maximal performance, we could in theory make bacteria seem slow.
So yes, the first molecular printer would be pathetically slow. But if it could print itself in even a week we'd have enough of them to cover entire planets in a year.
I'm not really interested in reading your diaries given your obvious ignorance.
Problem : to manufacture everything we use right now requires a vast and complex infrastructure. To produce any high tech good requires a complex process with thousands of steps with many steps unique to that particular good.
Solution : molecular manufacturing. Living creatures are far more complex internally than the most complicated machine humans have built to date. Yet, it is possible for living creatures to produce every single component they use, atom by atom, when supplied with nothing more than energy and the raw elements they need in a suitable feedstock. (the fact that larger creatures such as mammals lose the ability to be self sufficient like that is a limitation of evolution as a design algorithm and not a technical limit)
Molecular manufacturing is a technology that, through some technical means (there are hundreds of possible solutions) allows us to produce products with control over the bonding of all individual atoms in the product. By definition, since the machinery that performs molecular manufacturing is made of individual atoms bonded a particular way, said machinery will be able to manufacture a new copy of itself, leading to exponential growth.
The most common concept for this technology would be a 3d printer that is extremely small and can place individual atoms. It is accepted by all credible scientists that biochemistry already does this, so said printers are at least theoretically possible. (some have argued that we can't make a printer like this work without a suitable solvent)
Anyways, once we have this technology, space colonization would become quite simple. We would not even attempt to terraform planets - doing so would deprive us of the valuable elements located inside the celestial bodies. We would instead land a few probes equipped with high speed molecular manufacturing equipment, some ore gathering robots, and a power source. Those probes would produce more copies of the robots, the power plants, and the molecular printers which would in turn...
Once you have an entire surface covered with machinery, you order it to produce something to put cargo into orbit (a linear electromagnetic accelerator, or a laser launch array, or a few million space elevator tethers, or whatever is appropriate) and you throw your waste and your finished products into space. In space, you use the same technology to produce space habitats for humans.
Assuming you still have human beings, that is. This same technology would make many other things possible.