They almost certainly do *not* mean a metal glass, but rather an actual metal alloy with a low melting point. There are numerous examples of alloys with melting points near room temperature, several with melting points below room temperature.

Metallic mercury is not extremely toxic, it's compounds of mercury that cause trouble. However, several of the alloys mentioned above are entirely non-toxic, and likely to be used despite their higher price for ease of disposal. (And a mercury spill inside a computer with various high-current power supplies could produce some rather more dangerous mercury vapor, and worse.)

Finally, the electromagnetic pump does not imply a ferromagnetic liquid, only a conductive one...a current passes through the liquid perpendicular to the direction of flow and to an external magnetic field, and Lorentz forces push the fluid forward. They do not mean ferrofluid, they mean exactly what they say, liquid metal, and one that is not at all ferromagnetic.

Hope you're joking...on the chance you're not:

This doesn't consume water. It does in fact run on salt water...the water from your tap is not pure H20. The statements about it requiring energy to refine copper and zinc, or whatever the electrodes are, are correct, but there's not even a remotely significant water conservation issue here.

And on the copper and zinc...zinc is cheap and widely available, it's commonplace to plate steel with it and current US pennies are just copper-clad zinc. In any case, it's better than most battery chemistries out there, and the requirements are tiny. The amount of power this sort of cell could provide is extremely tiny, which means the electronics have to be incredibly efficient...likely operating in the nanowatt to microwatt range. This *is* definitely something to approve of.

And guys, neither electrode gets plated during the discharge of the cell. If the copper got plated with the zinc, you would end up with a zinc electrode and a copper-cored zinc electrode, and no power. What happens is that the zinc electrode gets oxidized by oxygen in the water, the resulting hydrogen ions being neutralized by electrons provided at the copper electrode...if an electrical circuit between the electrodes is provided, electrons flow from the zinc electrode to the copper one. The zinc is consumed, turning to zinc oxide and potentially reacting further with whatever salts might be in the water. The copper is unchanged.

Electrochemical plating occurs in rechargeable batteries, but it's typically not material from one plate being plated on the other, it's material from the electrolyte. Some chemistries involve more complex processes...lead acid batteries for example...but none of that is involved here.

It'd have effectively zero effect on global warming. The area of the silhouette of the power satellite's arrays would be tiny in comparison to the silhouette of Earth itself. Even if it allowed massive increases in energy usage by humanity, the proportional increase in energy added to Earth's surface would be tiny. It would also reduce or eliminate the use of fossil fuel plants.

Putting the collector arrays in orbit makes sense. The rectenna arrays used to collect the power on the ground would be open mesh, practically transparent to visible light, and could be erected over fields where crops are grown, placed close to the areas where the power is needed. No huge solar farms disrupting desert ecosystems and fewer big, long runs of power lines, in addition to more efficient use of the solar panels themselves (meaning lower environmental impact from their manufacture, since fewer of them are needed). A solar thermal approach would be even better in terms of environmental impact, though...you still need a lot of solar cell area, and they don't last forever, so they'd need to be replaced occasionally.

A RTG capable of continuously running the robot might be too heavy. A lighter one could charge a bank of batteries/supercapacitors and allow periods of activity in darkness, with rest periods required to recharge. Solar cells could allow continuous activity or a significant drop in recharge time, with the RTG primarily being there to allow operations in shadow.

Tiny sensors are generally made for tiny cameras, with tiny lenses, and are not made to be high quality sensors. There is nothing inherent in small size that means low quality, however, and a CMOS sensor might be quite a bit smaller than a CMOS sensor while retaining the same quality. I mentioned diffraction as an ultimate limit in sensor size...this sensor comes nowhere near it, as pixel size is a few micrometers, larger than the longest wavelengths of visible light. Noise and such are more a result of operating on very low amounts of light...due to the small size of the lens, not the sensor itself. A photodiode based sensor like this might be inherently less noisy...

Kev50027: I'm sorry, but no, you are incorrect. Sensor size has absolutely nothing to do with light gathering power in a camera. The light that hits the sensor is determined entirely by the amount of light that can enter the camera through its lens. If the sensor is halved in size and the lens modified to focus the image onto a sensor of that size, the sensor will receive exactly the same amount of light and record the same image. If "each CCD is seeing pretty much the same thing", then your lens just sucks. I will admit that smaller sensors require better lenses, but sensor size means nothing with respect to light gathering ability.

Sensor size makes no difference in image quality until you reach the diffraction limit. It makes no difference whatsoever in light gathering capability, that is determined entirely by the lens system.

Also, CCD sensors are widely used because they're cheap and easy to make, and more development has been done with them. A CMOS sensor such as this may have numerous advantages.

It was not a total success, but it was an extremely successful flight. This is their second launch ever, and the first stage went off perfectly, parachuting to a safe ocean landing to be recovered and reused...a first in the industry, to my knowledge. It delivered the second stage to the proper trajectory and separated, then the second stage ignited and performed as expected...at this point, they had demonstrated success at all the most difficult parts of launching a payload into orbit, and had a rocket in the air with sufficient fuel and a properly working engine capable of reaching orbit.

There was a control issue after the second stage had burned for some time, but it was due to some simple problem like fuel swirling around the tank...the vehicle performed very well until the fuel tanks started to empty. It's a matter of something like increasing the size of the anti-slosh baffles in the LOX tank or tweaking the software to recognize the effect and compensate, not a rework of any major part of the design. They have effectively proven that they've built a rocket capable of reaching orbit.

They also demonstrated (though they did not intend to) very short recycle times on the pad, aborting once and trying again 24 hours later, then aborting again after actual main engine ignition, then refueling and launching an hour later. They were attempting a great many things, and achieved nearly all of them...saying they failed because they didn't reach orbit ignores the great number of successes in this launch.

Most of the above comments are way off base. Quantum entanglement does *not* allow FTL communication. Setting the state of one particle of an entangled pair to some known state breaks the entanglement instead of changing the state of the distant particle...it can not transmit information. What entanglement means is that, if nothing breaks the entanglement, the relevant parts of the quantum state of a distant particle (established when the particles were entangled) can be inferred with a measurement of the local entangled particle.

In application to radar, one of each entangled photon pair could be directed into a delay line, the other photons being emitted as with normal radar, and the delayed photons could then be correlated with incoming photons to find the signature of the entangled partners that are reflected back. What you gain is better discrimination of your own radar signal from the other radar signals and intentional jamming, not FTL radar.

The solar wind certainly has more mass than sunlight, since light is massless. However, sunlight actually carries more momentum than the solar wind. (yes, photons have momentum even though they don't have mass) Solar sails operate by reflecting sunlight, the solar wind does not come into play. (Though there is another possible method of propulsion, "mini-magnetospheric plasma propulsion", which uses a large magnetic field to catch the solar wind, hopefully with less mass than a solar sail that gives similar thrust.)

This is indeed a photon drive, using the momentum of photons to provide thrust. Exactly the same principle as a solar sail, except that the light comes from a focused laser and can thus be pointed in whatever direction that is desired, as well as being far more intense than sunlight, allowing much smaller reflectors to be used. Both reflectors are pushed away from each other, but one of them could be a launch station with heavy, high-capacity power sources and lasers placed in a convenient orbit and kept supplied with stationkeeping fuel, launching packages which themselves do not need to carry big thrusters or the power sources to run them, and can thus be made much lighter and cheaper.

Josh Warner is correct about the performance of photon thrusters. The "delta-v" of a spacecraft is how much it can change its velocity with its engine and the fuel it can carry. Throwing lighter particles of fuel out the back at higher velocities takes more energy, but provides a higher delta-v for the same spacecraft mass. The limit of this is shining a beam of photons out the back, which takes a lot of energy to produce any useful thrust, but also expends no massive fuel. This laser thruster is even better, as the photons don't have to be generated on board the spacecraft...a refinement that makes the photon drive much more practical. Look around for information on "specific impulse" for more information.