Ah, I was thinking more in terms of cars that lacked manual controls. I don't think that the accident rate would really drop until you ban manual driving. That would probably also greatly simplify the design of automated cars, as they don't need to guess what the other cars are going to do.
Plus, if you get rid of manual driving you can take advantage of all kinds of efficiencies (no lines painted on roads, no designated lanes (cars use however many lane make sense at the moment), no traffic lights, etc). None of that works with humans at the wheel. In my thinking the ideal car would have no directional controls at all - just a way to tell it where to go.
That, and they don't like a world where random youtube video is listed on the channel guide next to CBS. Their whole business model is based on it being hard to broadcast TV, and so they are the gateways that decide what is good enough to show, and they get paid handsomely to do it. If any studio could just put their shows on TV sets, why would they give a chunk of the revenue to some network?
Apple TV probably doesn't bother them so much, since Apple is all about getting people to pay for an experience - one where they are the gatekeepers. That is basically just another form of what the networks are used to, and they're pretty good at competing in this world (you can argue how good, but the fact is that they've done well in this kind of market historically despite options like PPV and video rental).
The idea that the TV set is just a gateway into an unfiltered world of video, and that anybody with an Amazon AWS account or whatever can create a "network" isn't good for them.
But perhaps it's time TV stop spending 1-2 million dollars per episode?
Well, if they did that then ANYBODY could create a TV show. And, since Google TV would make random-webpage-RSS-feed as easy to tune into as CBS, there goes the oligopoly.
Right now the major networks control the distribution of television, and the barrier to entry is extremely high (what does it cost to actually get something broadcast to the majority of TV sets in the US?).
If you make TV nothing more than an RSS feed, then all you need is a pipe big enough to serve it, and of course that is scalable (buy a pipe for 10 viewers, and then keep upgrading it with increasing ad revenue as it grows).
Internet TV is a recipe for the end of the major networks, and they know it.
Sure, you can see people WORKING on them. You won't find them on the road at all until somebody solves liability.
I'm not saying that this is impossible or anything - it is just the law. However, it will take a lot to reign in the lawyer. There is also the legitimate issue of how do you ensure that there is an appropriate level of liability when manufacturers are negligent.
Those planes aren't computer-piloted. They're human-piloted with a computer that can assist. Ultimately the human is responsible for flying the plane, and the human is blamed when things go wrong. The autopilot is never blamed for a problem, because the human is supposed to switch it off if something goes wrong with it.
Having the human in charge saves the manufacturer from liability - from the autopilot at least.
Give me a call when we have planes flying without any human pilots at all. They're capable of doing so right now, but liability would never let it happen.
My point is that we still have pilots. The computer is there to assist in flying the plane. But, we make sure there is a pilot to blame when things go wrong. Works better for everybody that way (except the passengers, since quite often the humans mess up in ways the computer wouldn't).
I'm quite aware that plans are able to land themselves. My point was that we will always have pilots, even if they cause more harm than good, until we fix liability.
While you'd like to hope that, I doubt this would be the case.
Imagine if somebody invented a computer-driven car. It takes off and everybody is driving them. The annual death rate drops from 10k to year to 500 people per year.
The end result? The car gets banned and the company is sued out of existence for bad engineering. The 10k people who used to die each year were victims of misfortune. The 500 people who die now are victims of the company.
That is why we don't have computer-piloted cars/planes/etc. Our assignment of liability is way off. The first thing I thought of when I read this article was that this guy would going to get the book thrown at him. Sure, he did the right thing, but that isn't what counts in court. Fortunately everybody else seems to be doing the right thing as well, which is a rarity.
The speed of light is the speed limit of anything that has mass in the universe. Space itself does not have mass.
It is also a relative thing. Any two nearby points in the universe are moving apart very slowly. However, over large distances that expansion accumulates until you reach points that are expanding at C, or above.
That galaxy is moving at a fairly ordinary pace compared to anything near to it. That entire region of the universe is moving away quickly.
My workaround is to middle-click on the comment headline to expand it AND open it in another tab. That doesn't collapse the rest of the thread in the current window. Then every minute or two I hit Ctrl-W 50 times to clean up.
Why do you think Bank of America halted all foreclosures?
It wasn't because they didn't know who they had loaned money to. The problem was that they couldn't prove that they had loaned money to them. Rather than arguing to the court that they had digitized records and can only point to database entries and not documents, they forged documents (or rather, they paid people to do this and didn't monitor their activities).
I suspect that in the end some kind of compromise will be reached, because the alternatives aren't acceptable:
1. It won't be acceptable for poor people to lose their homes. 2. it won't be acceptable for homes that are repossessed to be unsellable. 3. It won't be acceptable for yuppies who borrowed $1M to be off the hook without declaring bankruptcy. 4. It won't be acceptable for the banks to collapse.
I'm guessing that the government will step in, give the bank practices a clean bill of health, declare one way or another that the banks have clean title, and then carve out some kind of deal or bailout for poor people. So, the bottom line is that people without means will keep their homes, banks will keep most of their money, and people with means will have to go through bankruptcy/foreclosure/etc. Now, that could happen a lot of ways, but I think this will be the end result (maybe poor people get foreclosed, but get handed assistance to buy a new home, or maybe to buy back their same home at market rates, etc - maybe banks are punished on paper, but bailed out in practice, etc).
The goal of the government will be to maintain the status quo - nobody loses big. This can't be done with a bailout since even the treasury can't write checks this big. However, things can be done to try to restore a semblance of order, which is often all you need for the short term.
I think the more fundamental question is whether the US reforms the market in the long term, or if this will lead to an even bigger crisis in the future.
Well, the Chinese will need those satellites if it wants to fight a war abroad - at least an efficient one. They only don't need them now because they don't go anywhere that ground radio won't effectively service. They've also never really demonstrated their ability to project power in actual warfare.
The ability of the US to invade China is going to be limited. However, the ability of China to even bombard the US with anything short of ICBMs is even more limited. China would have the disadvantages and advantages of fighting on its own territory.
You also miss one of the valuable properties of subs - hunting other subs. The US seems to still be the leader at that game.
I think the bottom line is that neither country is going to pursue war - the consequences are far too great for either side. Then again, maybe the present situation will give somebody inspiration to write another "Red Storm Rising" which would be a pleasant change from most of the books currently available in this genre.
Sure, but not just rare-earths - how about technology manufacture in general?
The US won WWII because it could turn out ships faster than the Japanese and Germans could sink them. I'd question whether that would be the case today. In many ways the US is in the position that the Japanese and Germans started WWII in - far ahead of the competition in fielded forces, but without sufficient industrial base to sustain a war after the complete loss of all existing equipment.
If you want to know who will win a six-month war you count how many planes each side has. If you want to know who will win a 5-year war you ignore that entirely and look at how many factories each side can operate.
As an American I couldn't agree more - we need to get out of the farm-subsidy business - or at least greatly curtail it. These days it is just a big payoff to agribusiness.
That doesn't mean that China gets a free pass though.
The market is big. You can make money being Nordstrom or by being Walmart. Either way nobody is guaranteed a place at the top - constant innovation is required. The nature of that innovation will of course be different depending on where in the market you want to be.
Think about trucking companies that use satellite tracking and routing. Think about traffic/weather data reporting. There are lots of benefits from networking vehicles beyond browsing facebook at 55mph. The companies that deploy these kinds of solutions are VERY aware of the dangers of driver distraction and info overload.
Yup, imagine a cell tower that broadcasts direction signals at each of 50 different phones that have open channels. That tower would have much larger range, and the phones could use much lower power. This would probably work well for any time-division protocol - during each time slot you point the antenna at only the phones that can talk during that time slot, greatly increasing your gain.
You can also stick one of these on the roof of your truck or whatever and now you can have continuous high-bandwidth satellite communication while driving, since you could track a satellite in realtime as you hit bumps/etc. That's probably a harder trick, but as long as you're fast enough it should be possible.
I'd also think that phased array would be easier to mount - just lay it flat on your roof or whatever.
I'd be curious about other applications for this. If you got cheap phased array you could make radar applications a lot more practical. If you can cheaply deploy phased arrays all over the place that also could have signals intelligence applications, or stealth detection applications, or it might work well for things like cell towers, etc.
Sure, but if Apple wants to have more than 20% market share they're going to have to figure out how to compete against that. You could argue that it is easy for Walmart to be the most popular store in the USA when their prices are 30% less than everybody else's, but after all, that's their business model.
The original Mac got clobbered because it was Apple vs the world, and the world won (Apple's ability to even stick around is an accomplishment when you think about it). This is just more of the same. I don't think the iPhone is going to die, but they'll end up with market shares similar to what Macs face on the desktop.
One concern I have with the gravity tractor idea is that it involves putting a huge mass into almost exactly the same orbit as the satellite to start.
Unless this is done VERY carefully, if there is a failure early in the process you end up with yet another big heavy thing that will hit the Earth.
It might be possible to plan the maneuvers such that at no point does the interceptor have a trajectory that will impact Earth. Essentially you'd have to aim "to the side" of the target, not so much literally as from a solutions-space standpoint.
Basically that is how anybody who sells anything works.
Walmart buys cheap toys for 97 cents, and sells them for 99 cents. They take advantage of the fact that people don't like to go to factories and buy lots of 100,000 pencils.
Ditto for market makers. You want to buy 100 shares of a stock at 9AM, somebody else wants to sell 100 shares at 11AM. You could wait around for your optimum price, or you can buy from a market maker instantly, maybe paying a penny extra. Likewise if you want to sell 100k shares of something you can wait for individuals to buy them 10 shares at a time from you, or dump them on a market maker in blocks of 10k each or whatever.
Stocks also have book value - if the company tanks or is bought out you'd expect to get that as well.
Of course, no company whose time has come ever announces "hey, we're doing great, but our business model is over so rather than waste all our cash we'd like to do an immediate and orderly shutdown and return value to our investors." Instead they go into a death spiral until they hit a point where it pretty-much costs the value of the assets just to divest them.
Stocks really do have legitimate intrinsic value - that isn't the problem. The problem is that due to people desperate to make a return better than treasuries they are bid up to a point where they are valued FAR above the intrinsic value.
Either replace the controller with one that reports to the OS what you want it to report, or replace the flash with a circuit that ACTS like flash ram but returns the error profile you want to the controller.
I own a G1, and I think this is actually one of the big problems with the android platform in general. The first phone is only two years old, and it already feels very obsolete. I'm running Eclair on it, but those stuck with vendor updates only are on 1.6, and that is where they will stay.
The decision-makers around android have way too much of a buy-a-new-phone mentality. Phones should be supported for at least two years, which is the shortest time you could get a subsidized replacement. If the phones cost $30 we could treat them as disposable, but for $450 for most of these phones most consumers who aren't enthusiasts aren't going to be replacing them annually (or even every 2 years).
If the firmware doesn't follow the spec, then that is the firmware's problem. Tell the users to complain to whoever sold them the phone. This sounds like IE6 all over again...
Slashdot Mirror
Ah, I was thinking more in terms of cars that lacked manual controls. I don't think that the accident rate would really drop until you ban manual driving. That would probably also greatly simplify the design of automated cars, as they don't need to guess what the other cars are going to do.
Plus, if you get rid of manual driving you can take advantage of all kinds of efficiencies (no lines painted on roads, no designated lanes (cars use however many lane make sense at the moment), no traffic lights, etc). None of that works with humans at the wheel. In my thinking the ideal car would have no directional controls at all - just a way to tell it where to go.
That, and they don't like a world where random youtube video is listed on the channel guide next to CBS. Their whole business model is based on it being hard to broadcast TV, and so they are the gateways that decide what is good enough to show, and they get paid handsomely to do it. If any studio could just put their shows on TV sets, why would they give a chunk of the revenue to some network?
Apple TV probably doesn't bother them so much, since Apple is all about getting people to pay for an experience - one where they are the gatekeepers. That is basically just another form of what the networks are used to, and they're pretty good at competing in this world (you can argue how good, but the fact is that they've done well in this kind of market historically despite options like PPV and video rental).
The idea that the TV set is just a gateway into an unfiltered world of video, and that anybody with an Amazon AWS account or whatever can create a "network" isn't good for them.
But perhaps it's time TV stop spending 1-2 million dollars per episode?
Well, if they did that then ANYBODY could create a TV show. And, since Google TV would make random-webpage-RSS-feed as easy to tune into as CBS, there goes the oligopoly.
Right now the major networks control the distribution of television, and the barrier to entry is extremely high (what does it cost to actually get something broadcast to the majority of TV sets in the US?).
If you make TV nothing more than an RSS feed, then all you need is a pipe big enough to serve it, and of course that is scalable (buy a pipe for 10 viewers, and then keep upgrading it with increasing ad revenue as it grows).
Internet TV is a recipe for the end of the major networks, and they know it.
Sure, you can see people WORKING on them. You won't find them on the road at all until somebody solves liability.
I'm not saying that this is impossible or anything - it is just the law. However, it will take a lot to reign in the lawyer. There is also the legitimate issue of how do you ensure that there is an appropriate level of liability when manufacturers are negligent.
Agreed, but get rid of the "fail-safe" pilots and see what happens to your liability. Even if the accident rate decreased you'd have tons of lawsuits.
Those planes aren't computer-piloted. They're human-piloted with a computer that can assist. Ultimately the human is responsible for flying the plane, and the human is blamed when things go wrong. The autopilot is never blamed for a problem, because the human is supposed to switch it off if something goes wrong with it.
Having the human in charge saves the manufacturer from liability - from the autopilot at least.
Give me a call when we have planes flying without any human pilots at all. They're capable of doing so right now, but liability would never let it happen.
My point is that we still have pilots. The computer is there to assist in flying the plane. But, we make sure there is a pilot to blame when things go wrong. Works better for everybody that way (except the passengers, since quite often the humans mess up in ways the computer wouldn't).
I'm quite aware that plans are able to land themselves. My point was that we will always have pilots, even if they cause more harm than good, until we fix liability.
While you'd like to hope that, I doubt this would be the case.
Imagine if somebody invented a computer-driven car. It takes off and everybody is driving them. The annual death rate drops from 10k to year to 500 people per year.
The end result? The car gets banned and the company is sued out of existence for bad engineering. The 10k people who used to die each year were victims of misfortune. The 500 people who die now are victims of the company.
That is why we don't have computer-piloted cars/planes/etc. Our assignment of liability is way off. The first thing I thought of when I read this article was that this guy would going to get the book thrown at him. Sure, he did the right thing, but that isn't what counts in court. Fortunately everybody else seems to be doing the right thing as well, which is a rarity.
The speed of light is the speed limit of anything that has mass in the universe. Space itself does not have mass.
It is also a relative thing. Any two nearby points in the universe are moving apart very slowly. However, over large distances that expansion accumulates until you reach points that are expanding at C, or above.
That galaxy is moving at a fairly ordinary pace compared to anything near to it. That entire region of the universe is moving away quickly.
This is where the red shift comes from.
My workaround is to middle-click on the comment headline to expand it AND open it in another tab. That doesn't collapse the rest of the thread in the current window. Then every minute or two I hit Ctrl-W 50 times to clean up.
Why do you think Bank of America halted all foreclosures?
It wasn't because they didn't know who they had loaned money to. The problem was that they couldn't prove that they had loaned money to them. Rather than arguing to the court that they had digitized records and can only point to database entries and not documents, they forged documents (or rather, they paid people to do this and didn't monitor their activities).
I suspect that in the end some kind of compromise will be reached, because the alternatives aren't acceptable:
1. It won't be acceptable for poor people to lose their homes.
2. it won't be acceptable for homes that are repossessed to be unsellable.
3. It won't be acceptable for yuppies who borrowed $1M to be off the hook without declaring bankruptcy.
4. It won't be acceptable for the banks to collapse.
I'm guessing that the government will step in, give the bank practices a clean bill of health, declare one way or another that the banks have clean title, and then carve out some kind of deal or bailout for poor people. So, the bottom line is that people without means will keep their homes, banks will keep most of their money, and people with means will have to go through bankruptcy/foreclosure/etc. Now, that could happen a lot of ways, but I think this will be the end result (maybe poor people get foreclosed, but get handed assistance to buy a new home, or maybe to buy back their same home at market rates, etc - maybe banks are punished on paper, but bailed out in practice, etc).
The goal of the government will be to maintain the status quo - nobody loses big. This can't be done with a bailout since even the treasury can't write checks this big. However, things can be done to try to restore a semblance of order, which is often all you need for the short term.
I think the more fundamental question is whether the US reforms the market in the long term, or if this will lead to an even bigger crisis in the future.
Well, the Chinese will need those satellites if it wants to fight a war abroad - at least an efficient one. They only don't need them now because they don't go anywhere that ground radio won't effectively service. They've also never really demonstrated their ability to project power in actual warfare.
The ability of the US to invade China is going to be limited. However, the ability of China to even bombard the US with anything short of ICBMs is even more limited. China would have the disadvantages and advantages of fighting on its own territory.
You also miss one of the valuable properties of subs - hunting other subs. The US seems to still be the leader at that game.
I think the bottom line is that neither country is going to pursue war - the consequences are far too great for either side. Then again, maybe the present situation will give somebody inspiration to write another "Red Storm Rising" which would be a pleasant change from most of the books currently available in this genre.
Sure, but not just rare-earths - how about technology manufacture in general?
The US won WWII because it could turn out ships faster than the Japanese and Germans could sink them. I'd question whether that would be the case today. In many ways the US is in the position that the Japanese and Germans started WWII in - far ahead of the competition in fielded forces, but without sufficient industrial base to sustain a war after the complete loss of all existing equipment.
If you want to know who will win a six-month war you count how many planes each side has. If you want to know who will win a 5-year war you ignore that entirely and look at how many factories each side can operate.
As an American I couldn't agree more - we need to get out of the farm-subsidy business - or at least greatly curtail it. These days it is just a big payoff to agribusiness.
That doesn't mean that China gets a free pass though.
The market is big. You can make money being Nordstrom or by being Walmart. Either way nobody is guaranteed a place at the top - constant innovation is required. The nature of that innovation will of course be different depending on where in the market you want to be.
Think about trucking companies that use satellite tracking and routing. Think about traffic/weather data reporting. There are lots of benefits from networking vehicles beyond browsing facebook at 55mph. The companies that deploy these kinds of solutions are VERY aware of the dangers of driver distraction and info overload.
Yup, imagine a cell tower that broadcasts direction signals at each of 50 different phones that have open channels. That tower would have much larger range, and the phones could use much lower power. This would probably work well for any time-division protocol - during each time slot you point the antenna at only the phones that can talk during that time slot, greatly increasing your gain.
Phased array gives you a lot of options...
You can also stick one of these on the roof of your truck or whatever and now you can have continuous high-bandwidth satellite communication while driving, since you could track a satellite in realtime as you hit bumps/etc. That's probably a harder trick, but as long as you're fast enough it should be possible.
I'd also think that phased array would be easier to mount - just lay it flat on your roof or whatever.
I'd be curious about other applications for this. If you got cheap phased array you could make radar applications a lot more practical. If you can cheaply deploy phased arrays all over the place that also could have signals intelligence applications, or stealth detection applications, or it might work well for things like cell towers, etc.
Sure, but if Apple wants to have more than 20% market share they're going to have to figure out how to compete against that. You could argue that it is easy for Walmart to be the most popular store in the USA when their prices are 30% less than everybody else's, but after all, that's their business model.
The original Mac got clobbered because it was Apple vs the world, and the world won (Apple's ability to even stick around is an accomplishment when you think about it). This is just more of the same. I don't think the iPhone is going to die, but they'll end up with market shares similar to what Macs face on the desktop.
One concern I have with the gravity tractor idea is that it involves putting a huge mass into almost exactly the same orbit as the satellite to start.
Unless this is done VERY carefully, if there is a failure early in the process you end up with yet another big heavy thing that will hit the Earth.
It might be possible to plan the maneuvers such that at no point does the interceptor have a trajectory that will impact Earth. Essentially you'd have to aim "to the side" of the target, not so much literally as from a solutions-space standpoint.
Basically that is how anybody who sells anything works.
Walmart buys cheap toys for 97 cents, and sells them for 99 cents. They take advantage of the fact that people don't like to go to factories and buy lots of 100,000 pencils.
Ditto for market makers. You want to buy 100 shares of a stock at 9AM, somebody else wants to sell 100 shares at 11AM. You could wait around for your optimum price, or you can buy from a market maker instantly, maybe paying a penny extra. Likewise if you want to sell 100k shares of something you can wait for individuals to buy them 10 shares at a time from you, or dump them on a market maker in blocks of 10k each or whatever.
Stocks also have book value - if the company tanks or is bought out you'd expect to get that as well.
Of course, no company whose time has come ever announces "hey, we're doing great, but our business model is over so rather than waste all our cash we'd like to do an immediate and orderly shutdown and return value to our investors." Instead they go into a death spiral until they hit a point where it pretty-much costs the value of the assets just to divest them.
Stocks really do have legitimate intrinsic value - that isn't the problem. The problem is that due to people desperate to make a return better than treasuries they are bid up to a point where they are valued FAR above the intrinsic value.
Or, you can just offer substitute hardware.
Either replace the controller with one that reports to the OS what you want it to report, or replace the flash with a circuit that ACTS like flash ram but returns the error profile you want to the controller.
I own a G1, and I think this is actually one of the big problems with the android platform in general. The first phone is only two years old, and it already feels very obsolete. I'm running Eclair on it, but those stuck with vendor updates only are on 1.6, and that is where they will stay.
The decision-makers around android have way too much of a buy-a-new-phone mentality. Phones should be supported for at least two years, which is the shortest time you could get a subsidized replacement. If the phones cost $30 we could treat them as disposable, but for $450 for most of these phones most consumers who aren't enthusiasts aren't going to be replacing them annually (or even every 2 years).
So, my question would be what does the spec say?
If the firmware doesn't follow the spec, then that is the firmware's problem. Tell the users to complain to whoever sold them the phone. This sounds like IE6 all over again...