Completely incorrect. A device infringes a patent if it fulfills the claims of that patent, regardless of whether it's been further improved. You can file a separate patent for the improvements, and in that case the device will be covered by *both* patents.
That's "dynamic" (i.e. fake) contrast. A display with dynamic contrast can turn down its backlight when displaying a black screen, which artificially increases the ratio between the brightness of a white screen and a black screen. However, that trick can't be used when displaying an image that's half white and half black, so the "real" contrast ratio you see most of the time is much, much lower.
Backlight brightness adjustment is a good feature but it doesn't compare to real high dynamic range. It's easy to see that the "dynamic" contrast ratio is a meaningless measurement: all you have to do is completely turn off the backlight when displaying a black screen and your dynamic contrast ratio is infinite! A real high dynamic range display could display an image of the sun as seen from space where the sun was so bright you wouldn't want to look directly at it, but space itself would be so dark that in a dark room you wouldn't be able to perceive the edges of the display.
I really don't think the hardware is good enough yet. To run smoothly and efficiently robots will need joint motors that are springy and compliant just like human muscles. All of the robot limbs I've ever seen are far too stiff (with the possible exception of BigDog's legs). Just look at this guy's head and arms shake while he's running; there are huge shock forces being transmitted from the feet directly up to the torso through all those stiff joints. Not only is that likely bad for the robot, it means that tons of energy is being wasted. For example, instead of letting the knee swing forward naturally during a step this robot has to run its servos to force the knee to rotate forward.
According to the contest rules the winning algorithm will not be exclusive to Netflix but will be published to the public, so we all win.
The reason BellKor is still first is that the published scores are irrelevant. The scores that matter for the prize are based on an unpublished data set known only to Netflix (to prevent people submitting answers that are optimized for the challenge data and work poorly on everything else). On this secret data set, BellKor's algorithm apparently performs better than The Ensemble's.
Wait, photosynthesis is only 15% efficient? You mean photovoltaic panels are already more efficient than plants? I had always assumed that nature had done much better than that.
Don't worry, DoD will be exempt, along with elected officials (for "security"). We couldn't have mere citizens finding out what their government is doing, now, could we?
You would put a diffuser in front of the laser to cover the whole scene. It might also be possible to use an LED. I don't know which Natal is using.
The Natal tech isn't a simple infrared camera viewing a projected grid, it's actually even cooler than that. Natal uses an infrared laser that fires a picosecond-long pulse once per frame. The camera includes infrared-sensing pixels alongside the normal RGB pixels, but the infrared ones are different: instead of measuring the intensity of the infrared light, they measure how long it takes the infrared light from the laser to reflect off of the scene and arrive at the camera, to a precision of tens of picoseconds. Since we know the speed of light we can calculate the distance from the camera to each object in the scene from this time measurement, to about 1cm accuracy.
The whole thing is similar to radar; in fact one name for this technology is LIDAR. Laser rangefinders work like this too; Natal basically has a laser rangefinder for each pixel of its camera. The coolest part of the whole thing is that none of this tech is expensive. Natal and devices like it shouldn't be any more expensive than your bog-standard webcam once they are manufactured in volume.
This technology is going to change the world in a few years. Not only will it enable cool games and Minority Report-style human-computer interfaces, but it will finally solve the computer vision problems of robotics, enabling autonomous pathfinding robots to navigate the real world successfully for the first time.
If the ISPs had any sense they'd implement their own open payment system where third parties could charge consumers via the ISP and have the charges appear on the consumer's bill, bypassing the Visa/MasterCard payment service cartel. That way the ISPs could still get their cut (it would be a gold mine for them), but consumers would actually have choice in their service providers.
This is much more than an ordinary lithium battery, and the headline is quite appropriate. Internal combustion engines are in a very real sense "fueled by air", as are our own bodies, and using the same principle to extend the life of batteries without increasing weight or volume is a very good idea. It's not totally unprecedented, either, as zinc-air batteries do this; the innovation is making them rechargeable.
Personally, I knew JJ Abrams was no Michael Bay when the orbital drill, after being destroyed, fell into San Francisco Bay *right next* to the Golden Gate bridge, but somehow missed subjecting us to a gratuitous and cliched effects sequence of the destruction of San Francisco's most famous landmark, preferring instead to get on with the story.
Google Code search supports regular expressions, so it's possible with a smaller index at least.
When I said robot cars would be more efficient I meant in total energy use of the system. The peak carrying capacity of a train is obviously going to be higher, but the train can only run at full capacity a small fraction of the time, on a small fraction of the total track length of the system. Outside of a city's downtown area, and even in downtown areas during non-peak times, light rail trains run at far below peak capacity which drags down the efficiency of the whole system. If you try to increase efficiency by reducing service then you make the system less useful and ridership goes down.
Due to these realities of rail service, battery powered cars with regenerative braking will use less energy overall to transport the same people. As for gridlock problems, robot cars should be at least somewhat better at handling them than human-driven cars. They will respond more quickly allowing them to pack more closely together. Packs of them will accelerate/decelerate as a unit almost like a train. They won't commit minor traffic violations and they will get in fewer accidents than human drivers. They will have perfect knowledge of traffic conditions via wireless Internet, and so will be able to route around disruptions and collaboratively load-balance different routes. They will be able to travel on their own to find parking, so cities will remove street parking in favor of centralized garages and use the space for more lanes. Lanes may also be made narrower.
When you start thinking about all the consequences self-driving robotic cars could have, it becomes apparent that as soon as they are available they are going to displace nearly all other forms of transporation.
The dinosaur is not the automobile, it's the internal combustion engine. Battery-powered cars are the future. Furthermore, within 20 years we'll have the technology to make self-driving robotic electric cars, which will be both more convenient and more efficient than mass transit for short to medium length trips. Existing mass transit systems will become obsolete.
The number of UNIX installations has grown to 10, with more expected. -- The Unix Programmer's Manual, 2nd Edition, June 1972
