Seriously -- No "breakthrough" required -- just steady progress for a decade or three. Two things though:
General to this class of devices: You probably have to land them near a power socket capable of delivering a fair amount of electricity at specific voltages, currents, and phases through a plug that matches the socket on the vehicle.
Specific to this particular vehicle: It looks to about as safe as Lawnchair Larry's Weather-balloon hoisted Sears lawnchair. darwinawards.com/stupid/stupid1998-11.html I don't think many insurance companies are going to underwrite a policy for either the driver or for bystanders.
Water channels probably would work. However, there may be a bit of a problem. The Sphinx, pyramids, etc at Giza are built on top of a limestone plateau. It looks like the Giza Plateau is at least 30 meters (100feet) above the peak level of the Nile back in pre-Aswan Dam times. I would think that any system of engineering works capable of lifting boats, innumerable BIG rocks, and prodigious amounts of water up to the top of the plateau would have left some pretty obvious traces.
And every time the Luddites have claimed that there will be no more jobs
If you do some research, you'll discover that the Luddites far from being wrong -- were dead right. Their comfortable cottage industry went away and the factory jobs that replaced it featured lower pay, longer hours, and unsafe working conditions. Consumers, and society as a whole, were better off of course. I doubt that was all that much comfort to those displaced.
But our current circumstance is something new. I really believe it's going to catch society completely flat footed. Hardly anybody is preparing for this. AI coupled with automation is going to eliminate entire fields of work. There will be nowhere to transition to.
I fear that you may be correct. 100 coal miners replaced by three coal mining machine repair guys and a handful of unskilled workers doing general cleanup. 200 factory workers replaced by a few techs.
And if you think that's a problem for the G7 which has substantially deindustrialized in the past several decades, think about the impact in China and other countries that the manufacturing jobs migrated to.
Those with memories that go back further than a few days will recall that economists across the political spectrum told us that unrestricted free trade was a great thing for everybody and that the manufacturing jobs displaced from developed countries would be replaced by zillions of great jobs in new industries. Those like Bernie Sanders and H Ross Perot who questioned that assumption were dismissed as flakes. Turns out that a lot of new jobs actually were created. But nowhere near as many as were lost. And not in the same places. And not for the same sort of people.
Likewise, President Obama's millions of green jobs, that his climate policies would create. Mostly not happening and a lot of the jobs that have been created are installing Chinese made solar panels that will never pay for themselves in inappropriate places. (BTW, I'm NOT against solar when the engineering is done properly. It's a good solution to some problems. There's no reason not use it in those cases. But too often engineering isn't done at all. Promises from ill-informed and not overly honest salesmen are used instead of engineering).
Anyway, I agree with the OP. Kurzwell is probably overly optimistic. And if he is, that will be a problem. And we should be worrying about what to do about it rather worrying about pressing issues like correct demeanor of football players during the playing of the national anthem.
The medical profession can look after itself thank you very much. If current diseases and conditions decline in importance/frequency, you'll probably find that you are suffering from BTS (Brittle Toenail Syndrome) and/or antibiotic resistant DGB (Depraved Gut Bacteria). Both will require immediate action. By specialists. Expensive specialists.
"Indeed. People were stupid back then and wanted cheap slaves, they are still stupid today and still want those cheap slaves."
Who wants a slave? You gotta feed them and house them. They're harder to train than Chihuahuas. One infected whip cut and the first thing you know, you're burying a capital asset. Slavery is over-rated.
Humans are terrible at guessing how long technological progress is going to take
Yes, we are terrible at guessing. But it's hard to see why. In the relatively uncommon situations where progress can be quantified, progress seems usually to be exponential. Example -- Moore's Law. Of course, the exponent is often pretty low. Batteries may be improving exponentially, but the progress seems painfully slow.
Is there any reason to think that unquantifiable change isn't exponential as well?
The one case I tracked at one time where exponential progress failed was CPU speed for microprocessors. It increased in a an orderly fashion from a few MHz to the 2-3 GHz range. Then progress stalled. Why? Because heat dissipation wasn't much of an issue for early CPUs. The 8088/8086 pulled about half a watt on a large chip in a package 5cm long that didn't need a heat sink. So initially, CPU speed was dominated by CPU feature size and switching speed. And they decreased/increased at a steady rate. But after a while, getting heat out of the device became a factor. And eventually heat generation/removal became the overriding issue.
I don't know about other folks, but I'd prefer that self driving car algorithms were:
1) determinate
2) testable
3) tested thoroughly.
I think that there are lot's of applications where learning algorithms should be perfectly OK. Controlling potentially lethal hardware is not one of those applications and, I would think, isn't likely to be for a very long time.
And THAT'S the problem. Nothing wrong with AI really. The stuff Amazon pushes at me while I try to shop is far closer to my interests and needs than the stuff advertisers not assisted by AI tout to me on TV. Used intelligently, AI is possibly useful at times and hopefully harmless.
The problems will come when people with a limited contact area with reality (pretty much everyone I fear) start confusing the educated guesses from AI agents with facts.
"Land surveyors would also like the higher precision to accurately map borders."
I've seen claims that with a half hour's worth of GPS data, surveyors can get cm level lat/lon resolution. That should be good enough for almost all surveying applications I would think. But it's hard to see how they can do that.
Yep. The system is called QZSS. But it gets its better than most GPS performance by using an unusual set of orbit configurations that puts a satellite well above the horizon in Japan at most times. That cuts back on multipath signal problems a lot and also reduces dead spots in cities and mountain areas. The system includes a correction signal that sounds sort of like a hybrid of WAAS and Differential GPS. It may also get some performance enhancement by using a unique ground controlled clock technology. As usual, there are drawbacks. A very large number of satellites would be needed to QZSS the whole earth. I'd assume the correction signal is optimized for Japan, And it sounds like the clock system requires a ground station be in contact with the satellite.
Good logic, but not quire correct. Yes, horizontal resolution is normally pretty much equal in any direction. Vertical resolution, however, is hampered by a relative dearth of altitude information at low satellite elevation angles and the fact that most satellite observations at any given time will come from satellites at low elevation angles. The latter is a geometry thing. If you're good at (3D) geometry, you can think it through and convince yourself.
If geometry makes your head hurt, you need probably to find some satellite acquisition tables and observe how little, if any, of the pass time (for almost all cases) has elevation angles above 45 degrees. I looked on line for an example, but couldn't find one in a few minutes of looking. Maybe better search terms...
the drones the government is developing for covert use are much, much smaller in size and carry a relatively small weapons payload with a limited effective radius. FaceID technology on iPhones would even allow for pre-strike confirmation of target identity & proximity to the device
So if anyone tells you that you look like _______ (Fill in the name of the current chief enemy of the US state and civilization as we know it - Kim Jong Un, Hassan Rouhani, Colin Kaepernick...etc, etc, etc) you should consider immediate plastic surgery. Or, at the very least, LOTS of makeup.
"Isn't "millions of objects" a simple scaling problem"
No. It's a somewhat complicated scaling problem because the objects are interacting and the number of interactions varies non-linearly with the density of the objects.and the diversity of their vectors
That said, it's probably a solvable problem IF everyone knows where everyone else is or there are 100% reliable communications with a central controller that knows where everyone is, can talk to everyone, and can issue commands that will be followed without question. (But count on the first attempts to build said central computer to come in years late, way over budget, and not work very well.)
This is America and suing folks for ludicrous amounts of money is -- along with professional football -- our national sport. So, sure -- $2.6B.
I do wonder where they expect Uber -- which looks to be headed for Chapter 11(?) bankruptcy at flank speed -- would come up with $2.60. Much less $2.6B. But maybe that's the point. Raise Uber to the ground and sow the soil with salt. That'll teach the bastards not to mess with Waymo's IP.
"The quality of Google searches has been going down for at least a decade"
Perhaps. But it doesn't seem so to me. I try DuckDuckGo every now and then, and it seems to me that Google does a significantly better job of finding the (often pretty obscure) stuff that I am interested in. I haven't tried to analyze why/how.
Since I have pretty much all advertising blocked in my hosts file, I don't really care all that much about the quite unpleasant associates Google is shopping my personal information to. Perhaps I should care. But I don't. If I cared, I'd probably try to set up a false online persona. But that's a hell of a lot of work.
Google's business is selling exposure to advertisers. Advertisers are, basically, nuts. Nonetheless, advertisers are nutcases with money burning holes in their pockets. (Why would anyone give real money to marketing people?) It doesn't matter if Google's vast trove of data has any real value in matching advertisers to potential buyers. (I'm guessing that it mostly doesn't).
As long as the advertisers believe it is effective, it works.
"Much more than that doesn't seem too helpful though, three months is a whole lot of searches, and should give plenty of information about what I'm searching for right now."
I should have thought that three DAYS would be sufficient. But what do I know?
"I think I'd opt to live in a cave if that were to occur."
Won't be completely safe there either. But it's probably safer than being out in the open. I think the next cave over has a for sale sigh. I'd go out and look, but it's rush hour. Any thoughts about what to do about the humidity in these caves? Or on solving the orc problem?
By the time we have the digital technology to reliably control millions of flying cars, we might actually have usable hydrogen fusion. Or solar panels paving the Sahara, Gobi, and everything from Barstow to Wichita Falls. And in BeauHD's universe making hydrocarbon fuels from CO2, water, and energy is something any schoolchild can tell you how to do.
"If they aren't limited by having humans controlling them, then it won't matter if they add a 3rd dimension."
I don't entirely disagree. But controlling stuff in three dimensions is not as easy as it sounds for all sorts of reasons. And there are a bunch of privacy, noise, and security issues. Plus a lot of problems like overhead wires. And wind. And handling poor visibility. And coordinating thousands of those suckers including the one right around the corner of the building on your left that you can't see and that can't see you and will be wanting to occupy the same space you plan to be occupying in three... two... one... seconds.
I also agree with the comment that it may be more than 15 years before we have truly safe fully autonomous cars. Not that there won't be lots of driverless vehicles in specific applications such as long distance expressway transport before then. But a driverless car you can safely send your ten year old off to school in is a long way off.
Slashdot Mirror
Seriously -- No "breakthrough" required -- just steady progress for a decade or three. Two things though:
General to this class of devices: You probably have to land them near a power socket capable of delivering a fair amount of electricity at specific voltages, currents, and phases through a plug that matches the socket on the vehicle.
Specific to this particular vehicle: It looks to about as safe as Lawnchair Larry's Weather-balloon hoisted Sears lawnchair. darwinawards.com/stupid/stupid1998-11.html I don't think many insurance companies are going to underwrite a policy for either the driver or for bystanders.
Trained Pilot?
Hogwash!!!
What can be so hard about flying this eggbeater?
HEY EVERYONE, WATCH THIS!!!
Water channels probably would work. However, there may be a bit of a problem. The Sphinx, pyramids, etc at Giza are built on top of a limestone plateau. It looks like the Giza Plateau is at least 30 meters (100feet) above the peak level of the Nile back in pre-Aswan Dam times. I would think that any system of engineering works capable of lifting boats, innumerable BIG rocks, and prodigious amounts of water up to the top of the plateau would have left some pretty obvious traces.
Six times faster. But who's counting?
And every time the Luddites have claimed that there will be no more jobs
If you do some research, you'll discover that the Luddites far from being wrong -- were dead right. Their comfortable cottage industry went away and the factory jobs that replaced it featured lower pay, longer hours, and unsafe working conditions. Consumers, and society as a whole, were better off of course. I doubt that was all that much comfort to those displaced.
But our current circumstance is something new. I really believe it's going to catch society completely flat footed. Hardly anybody is preparing for this. AI coupled with automation is going to eliminate entire fields of work. There will be nowhere to transition to.
I fear that you may be correct. 100 coal miners replaced by three coal mining machine repair guys and a handful of unskilled workers doing general cleanup. 200 factory workers replaced by a few techs.
And if you think that's a problem for the G7 which has substantially deindustrialized in the past several decades, think about the impact in China and other countries that the manufacturing jobs migrated to.
Those with memories that go back further than a few days will recall that economists across the political spectrum told us that unrestricted free trade was a great thing for everybody and that the manufacturing jobs displaced from developed countries would be replaced by zillions of great jobs in new industries. Those like Bernie Sanders and H Ross Perot who questioned that assumption were dismissed as flakes. Turns out that a lot of new jobs actually were created. But nowhere near as many as were lost. And not in the same places. And not for the same sort of people.
Likewise, President Obama's millions of green jobs, that his climate policies would create. Mostly not happening and a lot of the jobs that have been created are installing Chinese made solar panels that will never pay for themselves in inappropriate places. (BTW, I'm NOT against solar when the engineering is done properly. It's a good solution to some problems. There's no reason not use it in those cases. But too often engineering isn't done at all. Promises from ill-informed and not overly honest salesmen are used instead of engineering).
Anyway, I agree with the OP. Kurzwell is probably overly optimistic. And if he is, that will be a problem. And we should be worrying about what to do about it rather worrying about pressing issues like correct demeanor of football players during the playing of the national anthem.
The medical profession can look after itself thank you very much. If current diseases and conditions decline in importance/frequency, you'll probably find that you are suffering from BTS (Brittle Toenail Syndrome) and/or antibiotic resistant DGB (Depraved Gut Bacteria). Both will require immediate action. By specialists. Expensive specialists.
"Indeed. People were stupid back then and wanted cheap slaves, they are still stupid today and still want those cheap slaves."
Who wants a slave? You gotta feed them and house them. They're harder to train than Chihuahuas. One infected whip cut and the first thing you know, you're burying a capital asset. Slavery is over-rated.
Humans are terrible at guessing how long technological progress is going to take
Yes, we are terrible at guessing. But it's hard to see why. In the relatively uncommon situations where progress can be quantified, progress seems usually to be exponential. Example -- Moore's Law. Of course, the exponent is often pretty low. Batteries may be improving exponentially, but the progress seems painfully slow.
Is there any reason to think that unquantifiable change isn't exponential as well?
The one case I tracked at one time where exponential progress failed was CPU speed for microprocessors. It increased in a an orderly fashion from a few MHz to the 2-3 GHz range. Then progress stalled. Why? Because heat dissipation wasn't much of an issue for early CPUs. The 8088/8086 pulled about half a watt on a large chip in a package 5cm long that didn't need a heat sink. So initially, CPU speed was dominated by CPU feature size and switching speed. And they decreased/increased at a steady rate. But after a while, getting heat out of the device became a factor. And eventually heat generation/removal became the overriding issue.
I don't know about other folks, but I'd prefer that self driving car algorithms were:
1) determinate
2) testable
3) tested thoroughly.
I think that there are lot's of applications where learning algorithms should be perfectly OK. Controlling potentially lethal hardware is not one of those applications and, I would think, isn't likely to be for a very long time.
bean counters and financial officers ...
And THAT'S the problem. Nothing wrong with AI really. The stuff Amazon pushes at me while I try to shop is far closer to my interests and needs than the stuff advertisers not assisted by AI tout to me on TV. Used intelligently, AI is possibly useful at times and hopefully harmless.
The problems will come when people with a limited contact area with reality (pretty much everyone I fear) start confusing the educated guesses from AI agents with facts.
"Land surveyors would also like the higher precision to accurately map borders."
I've seen claims that with a half hour's worth of GPS data, surveyors can get cm level lat/lon resolution. That should be good enough for almost all surveying applications I would think. But it's hard to see how they can do that.
Yep. The system is called QZSS. But it gets its better than most GPS performance by using an unusual set of orbit configurations that puts a satellite well above the horizon in Japan at most times. That cuts back on multipath signal problems a lot and also reduces dead spots in cities and mountain areas. The system includes a correction signal that sounds sort of like a hybrid of WAAS and Differential GPS. It may also get some performance enhancement by using a unique ground controlled clock technology. As usual, there are drawbacks. A very large number of satellites would be needed to QZSS the whole earth. I'd assume the correction signal is optimized for Japan, And it sounds like the clock system requires a ground station be in contact with the satellite.
This is +- 3m in any direction
Good logic, but not quire correct. Yes, horizontal resolution is normally pretty much equal in any direction. Vertical resolution, however, is hampered by a relative dearth of altitude information at low satellite elevation angles and the fact that most satellite observations at any given time will come from satellites at low elevation angles. The latter is a geometry thing. If you're good at (3D) geometry, you can think it through and convince yourself.
If geometry makes your head hurt, you need probably to find some satellite acquisition tables and observe how little, if any, of the pass time (for almost all cases) has elevation angles above 45 degrees. I looked on line for an example, but couldn't find one in a few minutes of looking. Maybe better search terms ...
the drones the government is developing for covert use are much, much smaller in size and carry a relatively small weapons payload with a limited effective radius. FaceID technology on iPhones would even allow for pre-strike confirmation of target identity & proximity to the device
So if anyone tells you that you look like _______ (Fill in the name of the current chief enemy of the US state and civilization as we know it - Kim Jong Un, Hassan Rouhani, Colin Kaepernick ...etc, etc, etc) you should consider immediate plastic surgery. Or, at the very least, LOTS of makeup.
"Isn't "millions of objects" a simple scaling problem"
No. It's a somewhat complicated scaling problem because the objects are interacting and the number of interactions varies non-linearly with the density of the objects.and the diversity of their vectors
That said, it's probably a solvable problem IF everyone knows where everyone else is or there are 100% reliable communications with a central controller that knows where everyone is, can talk to everyone, and can issue commands that will be followed without question. (But count on the first attempts to build said central computer to come in years late, way over budget, and not work very well.)
This is America and suing folks for ludicrous amounts of money is -- along with professional football -- our national sport. So, sure -- $2.6B.
I do wonder where they expect Uber -- which looks to be headed for Chapter 11(?) bankruptcy at flank speed -- would come up with $2.60. Much less $2.6B. But maybe that's the point. Raise Uber to the ground and sow the soil with salt. That'll teach the bastards not to mess with Waymo's IP.
"The quality of Google searches has been going down for at least a decade"
Perhaps. But it doesn't seem so to me. I try DuckDuckGo every now and then, and it seems to me that Google does a significantly better job of finding the (often pretty obscure) stuff that I am interested in. I haven't tried to analyze why/how.
Since I have pretty much all advertising blocked in my hosts file, I don't really care all that much about the quite unpleasant associates Google is shopping my personal information to. Perhaps I should care. But I don't. If I cared, I'd probably try to set up a false online persona. But that's a hell of a lot of work.
Google's business is selling exposure to advertisers. Advertisers are, basically, nuts. Nonetheless, advertisers are nutcases with money burning holes in their pockets. (Why would anyone give real money to marketing people?) It doesn't matter if Google's vast trove of data has any real value in matching advertisers to potential buyers. (I'm guessing that it mostly doesn't).
As long as the advertisers believe it is effective, it works.
"Much more than that doesn't seem too helpful though, three months is a whole lot of searches, and should give plenty of information about what I'm searching for right now."
I should have thought that three DAYS would be sufficient. But what do I know?
"I think I'd opt to live in a cave if that were to occur."
Won't be completely safe there either. But it's probably safer than being out in the open. I think the next cave over has a for sale sigh. I'd go out and look, but it's rush hour. Any thoughts about what to do about the humidity in these caves? Or on solving the orc problem?
By the time we have the digital technology to reliably control millions of flying cars, we might actually have usable hydrogen fusion. Or solar panels paving the Sahara, Gobi, and everything from Barstow to Wichita Falls. And in BeauHD's universe making hydrocarbon fuels from CO2, water, and energy is something any schoolchild can tell you how to do.
Broken cars (mostly) STOP. Broken aircraft DROP.
"If they aren't limited by having humans controlling them, then it won't matter if they add a 3rd dimension."
I don't entirely disagree. But controlling stuff in three dimensions is not as easy as it sounds for all sorts of reasons. And there are a bunch of privacy, noise, and security issues. Plus a lot of problems like overhead wires. And wind. And handling poor visibility. And coordinating thousands of those suckers including the one right around the corner of the building on your left that you can't see and that can't see you and will be wanting to occupy the same space you plan to be occupying in three ... two ... one ... seconds.
I also agree with the comment that it may be more than 15 years before we have truly safe fully autonomous cars. Not that there won't be lots of driverless vehicles in specific applications such as long distance expressway transport before then. But a driverless car you can safely send your ten year old off to school in is a long way off.