As for utility : making technology more useful than it already is is really really hard. There's a lot of obstacles in the way, chief among them limitations on human beings. A lot of the current things that make you bang your head against the wall are pretty much solely caused by humans, not the tech itself. For instance, the biggest problem with automobiles is traffic, which is caused by the human drivers. And our legal liability system nixes self driving cars.
The only way to make most things better is to make them more complicated.
A solar panel pumped electric grid will be much more complex than the current system, since now you have to have variable pricing dependent on sunshine, storage reservoirs, inputs from solar arrays located all over a city, etc.
A fusion reactor, if we ever make one work, will be a heck of a lot more complicated than a fission plant. The magnetic fields or lasers and the control systems are stupendously complex compared to a lead lined box with uranium in it.
A flying car will need to be able to basically fly and maintain itself without human labor. Lot more complex than a helicopter.
Nanotechnology to kill a metastatic cancer has to be vastly more sophisticated than current chemotherapy drugs, because it's a complex task to differentiate between healthy and malignant cells.
The internals of the computer you read this on are vastly more complex than the one you could have bought 20 years ago. All the complexity has been hidden and packed down into OS modules and into the integrated circuits, so there may actually be fewer visible chips if you open up the case. But a diagram that shows every transistor would be thousands of times larger than a picture made of a computer in 1989.
A better yardstick for technological progress is not the utility of technology, but the internal complexity of the technology. A Mercedes today may still be an internal combustion engine automobile - but far more engineering has gone into the design of the auto than into a mercedes of 1959. There's far more sophisticated embedded systems inside it, from electronic keys to a sophisticated crash mitigation system. Aerodynamics and reliability and numerous other factors have had countless iterations of engineering put into them.
Yet, of course, the actual improvement in your life if you owned either car is small. You're more likely to survive a crash in the newer automobile - but crashes don't happen every day, and people drive more dangerously today, so the death rate is comparable. Either car can go 70 mph on the interstate.
All the rest of technology today is similar. A lot of things don't seem to have improved much - but the complexity of the internals have increased. Doctors and hospitals today have a much longer list of things they worry about when they treat for a disease - although outcomes are only slightly better.
He is right about one thing. For the nanotechnology and flying cars and other wonders of the "singularity", the internal complexity of that technology will dwarf anything we have today. Human beings, even working as large teams, don't really have the brain power to create technology this complicated within a reasonable investment timespan. That's why the first stage of the singularity is information technology : we first have to augment our ability to handle complexity (whether through AI or cyborgs or whatnot). The flying cars and the immortality granting nanotechnology come later.
When I wore the Army helmets in basic training, I noticed this design flaw. Basically, there's no padding or shock absorbing foam in the helmet. (there's a redesigned helmet that is in use now that has a little bit of padding but not like a bike helmet). Sometimes, soldiers would playfully rifle butt each other in the head. I noticed whenever this happened to me that the rigid helmet would let most of the force of the blow right through, and it would make a loud ringing sound. Evidently, that's pretty bad when an IED goes off.
Of course, the ultimate solution is to put infantry drones in the blast zone of IED, not human beings. Unfortunately for all the soldiers who have died, the tech won't be ready for another 20-30 years.
The antennas would be phase array - they would be electronically steered, with a mechanical cone to block off axis signals.
Only command signals from the control center would have to be one time pad encrypted. Controls signals (a joystick update a hundred times a second, button presses, coordinates, etc) would use up minimal amounts of keyspace. And you could use a terrabyte flash drive to hold the key file. The video feeds from the drones could use weaker methods.
As an engineer, you must fully understand the cost differences I'm talking about. Making a machine that only has to work in a few battles (since realistically, if you send air superiority fighters into a shooting war, you're going to lose a bunch of them ANYWAY) and is acceptable to lose sometimes is a lot easier than building an aircraft to current standards. As for peripheral visions...that's just insulting. How many air battles happen at visual range? Can you not engineer a camera to give better peripheral vision than actually being there?
Hence the word long term. I'm not disputing that it will be a while before the drone air interceptors are a mature technology. Furthermore, countries that are 20 years behind the United States in technology will take at least 20 years more before they are flying similar technology.
Intelligence isn't just about being right. It's also about having the mental flexibility to try new ideas out, not just the party line. Maybe I'm wrong...but I'm at least posing my ideas in a rational framework and seeking feedback. You have no ideas of your own : what you are probably going to respond is the same argument given in 1960 why drones won't work. Thing is, it isn't 1960 any more, and the electronics grow ever more sophisticated while human bodies remain the same.
I'd like to add to that - you don't have to change the pilots out when they get old. The reason we don't have 55 year olds as fighter pilots is because their bodies aren't up to the physical rigor as much. (there's other reasons, but that's beside the point). You could have drone pilots with decades more experience.
Why do you say that? Directional antenna would mean that the drone would only "listen" for communications coming from the positions of other nodes in the mesh network. Unless the enemy can put jammers all around, including in the air and above the drone aircraft, at least some comm would get through. One time pad encryption means the enemy can't hijack the drone. (since the only copies of the encryption key would be one flash drive aboard the aircraft, and it's pair inside the command module for the drone. A hardware random number generator would be used)
You can't use missiles to kill the jammer because a jammer could be made for less money than a missile.
So you think that manned aircraft, with squishy humans aboard, that require you to pay for maintainence and fuel while the aircraft is flown thousands of hours in order to give the pilot practice is cost effective? Cost effectiveness matters because, again, quantity has a quality all on its own. I suspect that 10 drone aircraft against one piloted one, with all the drones packing the same missiles just might be more cost effective. Heck, you could sacrifice a drone to cause the manned aircraft to give away his position by launching a missile or closing in with guns, letting the operators of the other drones know where to shoot.
First of all, these drones are drones. There's still a pilot - who isn't in the plane.
Second, you can get MORE piloting skill using drones. AND you can push the aircraft much harder. The reason is obvious. Your ace pilots won't get killed. Morover, even in a hot shooting war, a fighter pilot won't be in an actual dogfight more than a few minutes of a mission (most of the time is taken up getting to the combat zone, finding a target, etc). So, you could have your weaker, less talented pilots handle flying the drone fighters to the battle and have your ace pilots take over when the aircraft is in range of an enemy fighter.
Finally, the cost difference
Imagine a piloted aircraft up against 5 or 10 to one odds (because the country that pays for drones and doesn't have to pay for all those costs I mentioned in the post above can spend that money buying more drones). Every one of those drone aircraft has a pilot at the stick just as good as he is, or better. The drones can pull as many Gs as their airframe can take.
Long term, are manned aircraft going to be still used for air superiority?
Cost effectiveness might be a key factor. Drone aircraft don't need to be manufactured to fly for years and thousands of missions. They could be made just good enough to survive 10 to 100 or so sorties, with a 10% failure rate considered acceptable for the first mission. Drone operators could train using simulators and a small number of better quality drone aircraft. For the missions needing drones to loiter over an area for a prolonged period, a different model of drone would be used - you don't need high speed jet interceptors if the enemy has no aircraft left. Also, drones wouldn't need to have the dogfighting performance of an F-35. They could be slower and less maneueverable - but packed with missiles and with a radar system capable of defeating stealth aircraft.
Drone aircraft wouldn't need to be "recalled" or inspected. If a fault is found that might cause a crash, no point in fixing it unless the problem is severe. You could manufacture thousands of them and leave them stored in special packing canisters. Unpack a few every few years and use them testing them to get empirical measurements of average 'shelf life'.
I think that with these and other cost saving measures, you could probably manufacture 3 to 5 drone aircraft for the cost of one manned aircraft with similar capabilities. The MQ-9 Reaper is about 1/3 the cost of the Apache helicopter it supplants. As long as you could guarantee that the drones would always work despite enemy jamming (possible with mesh networking, phase array communication antenna and one time pad encryption, I think) then they would be the only game in town.
You must have flunked biochemistry. Shipping a food long distance doesn't significantly affect nutritional value.
Do YOU want to pick crops by hand? I don't, and I don't think anyone should have to. No mechanical harvesting is a non starter.
Local solar makes sense, because rooftops are land that would otherwise go to waste. But not, say, local windmills (because the efficiency of a wind turbine dramatically increases the bigger you build it)
You're confusing the issue. The "eco-hippies" that want less pollution also ask for simpler lifestyles, "organic" food, and other impractical ideas. The science and engineering people know that new technology will let us accomplish the same goals with less pollution and lower costs, and want to develop the technology further.
Only one of those groups is going to get what they want. Organic food off family farms and "buying local" is a non starter. So is "local generation." Electric cars, giant solar farms run by power companies, and increasing consolidation and specialization through globalism is going to happen.
The "in plain sight" doctrine came about as a result of an old Supreme Court case. What it boils down to is, if the cops execute a search warrant or other lawful search, and they happen to spot evidence of another crime "in plain sight", they can use that evidence to arrest and charge someone. Say the cops are checking your motel room for an escaped prisoner. They can't go rifling through your bag looking for drugs once they've searched the room. But, if you have a meth lab set up in the room, they can get you for that.
The same thing with this database search. Databases can be any arbitrary size : a database could have records on every citizen in the United States. If the cops were given a warrant to check on the records of a specific citizen, the rest of the database should be off limits. Otherwise, there's no real limit to the games the cops could play, and they would effectively have the power to investigate every citizen in the United Stats for a crime at all times. What if the "database" contained the banking records of every citizen in the U.S.?
Before they can think about a moon base, maybe they should fix the problem of getting into orbit in the first place. Right now, the current implementation is not a solution. $10,000 or so a kilogram is stupidly expensive. It costs many millions of dollars to blast just one astronaut into space.
The Amiga could have 112 megabytes of memory? How did you even get that much RAM? I remember upgrading my 486 running Windows 3.1 from 8 to 16 megs of RAM, and thinking that was a huge boost. It was expensive, too...
Crysis and other big games eat over a gig when you're playing them.
Supreme Commander will crash in big battles because it runs out of memory. It hits the 3 gigabyte limit. Part of that is the game's extremely inefficient architecture : I'm sure there must be memory leaks as well, but in any case, in a big battle with thousands of simulated troops it will run out of memory and crash.
Video editing software eats memory like candy.
All the desktop processors released for the past several years have 64-bit support, I thought even the atom does.
Like some of the other slashdotters here, I think they should just stop selling 32-bit OSes. It's inevitable that PCs will routinely have more than 4 gigs of RAM, so we might as well get the transition over with now, so that those of us with decent PCs and 64-bit OSes can enjoy the undivided attention of driver and application developers.
My current machine is almost 3 years old, and it has 4 gigs of RAM. My i7 box, when I build it, will most likely be loaded with 12 gigs.
Actually I use x64 XP. But, certain applications and drivers won't work. My smartpen, an old scanner, and so forth. Most annoyingly is I can't have separate color profiles for my dual displays in 64 bit XP - the program that lets you do that is 32 bit only.
Slashdot Mirror
As for utility : making technology more useful than it already is is really really hard. There's a lot of obstacles in the way, chief among them limitations on human beings. A lot of the current things that make you bang your head against the wall are pretty much solely caused by humans, not the tech itself. For instance, the biggest problem with automobiles is traffic, which is caused by the human drivers. And our legal liability system nixes self driving cars.
The only way to make most things better is to make them more complicated.
A solar panel pumped electric grid will be much more complex than the current system, since now you have to have variable pricing dependent on sunshine, storage reservoirs, inputs from solar arrays located all over a city, etc.
A fusion reactor, if we ever make one work, will be a heck of a lot more complicated than a fission plant. The magnetic fields or lasers and the control systems are stupendously complex compared to a lead lined box with uranium in it.
A flying car will need to be able to basically fly and maintain itself without human labor. Lot more complex than a helicopter.
Nanotechnology to kill a metastatic cancer has to be vastly more sophisticated than current chemotherapy drugs, because it's a complex task to differentiate between healthy and malignant cells.
The internals of the computer you read this on are vastly more complex than the one you could have bought 20 years ago. All the complexity has been hidden and packed down into OS modules and into the integrated circuits, so there may actually be fewer visible chips if you open up the case. But a diagram that shows every transistor would be thousands of times larger than a picture made of a computer in 1989.
A better yardstick for technological progress is not the utility of technology, but the internal complexity of the technology. A Mercedes today may still be an internal combustion engine automobile - but far more engineering has gone into the design of the auto than into a mercedes of 1959. There's far more sophisticated embedded systems inside it, from electronic keys to a sophisticated crash mitigation system. Aerodynamics and reliability and numerous other factors have had countless iterations of engineering put into them.
Yet, of course, the actual improvement in your life if you owned either car is small. You're more likely to survive a crash in the newer automobile - but crashes don't happen every day, and people drive more dangerously today, so the death rate is comparable. Either car can go 70 mph on the interstate.
All the rest of technology today is similar. A lot of things don't seem to have improved much - but the complexity of the internals have increased. Doctors and hospitals today have a much longer list of things they worry about when they treat for a disease - although outcomes are only slightly better.
He is right about one thing. For the nanotechnology and flying cars and other wonders of the "singularity", the internal complexity of that technology will dwarf anything we have today. Human beings, even working as large teams, don't really have the brain power to create technology this complicated within a reasonable investment timespan. That's why the first stage of the singularity is information technology : we first have to augment our ability to handle complexity (whether through AI or cyborgs or whatnot). The flying cars and the immortality granting nanotechnology come later.
That isn't how it seems from the perspective of an actual soldier.
When I wore the Army helmets in basic training, I noticed this design flaw. Basically, there's no padding or shock absorbing foam in the helmet. (there's a redesigned helmet that is in use now that has a little bit of padding but not like a bike helmet). Sometimes, soldiers would playfully rifle butt each other in the head. I noticed whenever this happened to me that the rigid helmet would let most of the force of the blow right through, and it would make a loud ringing sound. Evidently, that's pretty bad when an IED goes off.
Of course, the ultimate solution is to put infantry drones in the blast zone of IED, not human beings. Unfortunately for all the soldiers who have died, the tech won't be ready for another 20-30 years.
Current RPVs are not designed with the airframes for air to air combat.
The antennas would be phase array - they would be electronically steered, with a mechanical cone to block off axis signals.
Only command signals from the control center would have to be one time pad encrypted. Controls signals (a joystick update a hundred times a second, button presses, coordinates, etc) would use up minimal amounts of keyspace. And you could use a terrabyte flash drive to hold the key file. The video feeds from the drones could use weaker methods.
As an engineer, you must fully understand the cost differences I'm talking about. Making a machine that only has to work in a few battles (since realistically, if you send air superiority fighters into a shooting war, you're going to lose a bunch of them ANYWAY) and is acceptable to lose sometimes is a lot easier than building an aircraft to current standards. As for peripheral visions...that's just insulting. How many air battles happen at visual range? Can you not engineer a camera to give better peripheral vision than actually being there?
Hence the word long term. I'm not disputing that it will be a while before the drone air interceptors are a mature technology. Furthermore, countries that are 20 years behind the United States in technology will take at least 20 years more before they are flying similar technology.
Intelligence isn't just about being right. It's also about having the mental flexibility to try new ideas out, not just the party line. Maybe I'm wrong...but I'm at least posing my ideas in a rational framework and seeking feedback. You have no ideas of your own : what you are probably going to respond is the same argument given in 1960 why drones won't work. Thing is, it isn't 1960 any more, and the electronics grow ever more sophisticated while human bodies remain the same.
I'd like to add to that - you don't have to change the pilots out when they get old. The reason we don't have 55 year olds as fighter pilots is because their bodies aren't up to the physical rigor as much. (there's other reasons, but that's beside the point). You could have drone pilots with decades more experience.
There's a reason you got modded troll
Why do you say that? Directional antenna would mean that the drone would only "listen" for communications coming from the positions of other nodes in the mesh network. Unless the enemy can put jammers all around, including in the air and above the drone aircraft, at least some comm would get through. One time pad encryption means the enemy can't hijack the drone. (since the only copies of the encryption key would be one flash drive aboard the aircraft, and it's pair inside the command module for the drone. A hardware random number generator would be used)
You can't use missiles to kill the jammer because a jammer could be made for less money than a missile.
So you think that manned aircraft, with squishy humans aboard, that require you to pay for maintainence and fuel while the aircraft is flown thousands of hours in order to give the pilot practice is cost effective? Cost effectiveness matters because, again, quantity has a quality all on its own. I suspect that 10 drone aircraft against one piloted one, with all the drones packing the same missiles just might be more cost effective. Heck, you could sacrifice a drone to cause the manned aircraft to give away his position by launching a missile or closing in with guns, letting the operators of the other drones know where to shoot.
First of all, these drones are drones. There's still a pilot - who isn't in the plane.
Second, you can get MORE piloting skill using drones. AND you can push the aircraft much harder. The reason is obvious. Your ace pilots won't get killed. Morover, even in a hot shooting war, a fighter pilot won't be in an actual dogfight more than a few minutes of a mission (most of the time is taken up getting to the combat zone, finding a target, etc). So, you could have your weaker, less talented pilots handle flying the drone fighters to the battle and have your ace pilots take over when the aircraft is in range of an enemy fighter.
Finally, the cost difference
Imagine a piloted aircraft up against 5 or 10 to one odds (because the country that pays for drones and doesn't have to pay for all those costs I mentioned in the post above can spend that money buying more drones). Every one of those drone aircraft has a pilot at the stick just as good as he is, or better. The drones can pull as many Gs as their airframe can take.
Outcome is obvious.
Long term, are manned aircraft going to be still used for air superiority?
Cost effectiveness might be a key factor. Drone aircraft don't need to be manufactured to fly for years and thousands of missions. They could be made just good enough to survive 10 to 100 or so sorties, with a 10% failure rate considered acceptable for the first mission. Drone operators could train using simulators and a small number of better quality drone aircraft. For the missions needing drones to loiter over an area for a prolonged period, a different model of drone would be used - you don't need high speed jet interceptors if the enemy has no aircraft left. Also, drones wouldn't need to have the dogfighting performance of an F-35. They could be slower and less maneueverable - but packed with missiles and with a radar system capable of defeating stealth aircraft.
Drone aircraft wouldn't need to be "recalled" or inspected. If a fault is found that might cause a crash, no point in fixing it unless the problem is severe. You could manufacture thousands of them and leave them stored in special packing canisters. Unpack a few every few years and use them testing them to get empirical measurements of average 'shelf life'.
I think that with these and other cost saving measures, you could probably manufacture 3 to 5 drone aircraft for the cost of one manned aircraft with similar capabilities. The MQ-9 Reaper is about 1/3 the cost of the Apache helicopter it supplants. As long as you could guarantee that the drones would always work despite enemy jamming (possible with mesh networking, phase array communication antenna and one time pad encryption, I think) then they would be the only game in town.
You must have flunked biochemistry. Shipping a food long distance doesn't significantly affect nutritional value.
Do YOU want to pick crops by hand? I don't, and I don't think anyone should have to. No mechanical harvesting is a non starter.
Local solar makes sense, because rooftops are land that would otherwise go to waste. But not, say, local windmills (because the efficiency of a wind turbine dramatically increases the bigger you build it)
You're confusing the issue. The "eco-hippies" that want less pollution also ask for simpler lifestyles, "organic" food, and other impractical ideas. The science and engineering people know that new technology will let us accomplish the same goals with less pollution and lower costs, and want to develop the technology further.
Only one of those groups is going to get what they want. Organic food off family farms and "buying local" is a non starter. So is "local generation." Electric cars, giant solar farms run by power companies, and increasing consolidation and specialization through globalism is going to happen.
Right. In 1993 I recall RAM was $50/megabyte.
What era are these RAM expander cards from? I'm just trying to put it into context, because DRAM was obscenely expensive during the 80s and early 90s.
The "in plain sight" doctrine came about as a result of an old Supreme Court case. What it boils down to is, if the cops execute a search warrant or other lawful search, and they happen to spot evidence of another crime "in plain sight", they can use that evidence to arrest and charge someone. Say the cops are checking your motel room for an escaped prisoner. They can't go rifling through your bag looking for drugs once they've searched the room. But, if you have a meth lab set up in the room, they can get you for that.
The same thing with this database search. Databases can be any arbitrary size : a database could have records on every citizen in the United States. If the cops were given a warrant to check on the records of a specific citizen, the rest of the database should be off limits. Otherwise, there's no real limit to the games the cops could play, and they would effectively have the power to investigate every citizen in the United Stats for a crime at all times. What if the "database" contained the banking records of every citizen in the U.S.?
What are they thinking?
Before they can think about a moon base, maybe they should fix the problem of getting into orbit in the first place. Right now, the current implementation is not a solution. $10,000 or so a kilogram is stupidly expensive. It costs many millions of dollars to blast just one astronaut into space.
The Amiga could have 112 megabytes of memory? How did you even get that much RAM? I remember upgrading my 486 running Windows 3.1 from 8 to 16 megs of RAM, and thinking that was a huge boost. It was expensive, too...
Crysis and other big games eat over a gig when you're playing them.
Supreme Commander will crash in big battles because it runs out of memory. It hits the 3 gigabyte limit. Part of that is the game's extremely inefficient architecture : I'm sure there must be memory leaks as well, but in any case, in a big battle with thousands of simulated troops it will run out of memory and crash.
Video editing software eats memory like candy.
All the desktop processors released for the past several years have 64-bit support, I thought even the atom does.
Like some of the other slashdotters here, I think they should just stop selling 32-bit OSes. It's inevitable that PCs will routinely have more than 4 gigs of RAM, so we might as well get the transition over with now, so that those of us with decent PCs and 64-bit OSes can enjoy the undivided attention of driver and application developers.
My current machine is almost 3 years old, and it has 4 gigs of RAM. My i7 box, when I build it, will most likely be loaded with 12 gigs.
Actually I use x64 XP. But, certain applications and drivers won't work. My smartpen, an old scanner, and so forth. Most annoyingly is I can't have separate color profiles for my dual displays in 64 bit XP - the program that lets you do that is 32 bit only.