Wikipedia claims that VP9 does/will support alpha channel

http://en.wikipedia.org/wiki/Deep_learning http://en.wikipedia.org/wiki/Restricted_Boltzmann_machine http://en.wikipedia.org/wiki/Long_short_term_memory

But the SRBs have about 5 times the total thrust of the main engines, so the main engines become really helpful only after the SRBs give out. Therefore, they're more like a second stage, and like the second stages of the Saturn V or Atlas V (Centaur) should probably use LOX/LH2 to reduce mass and increase the mass ratio of the first stage.

Chrome's port of WebKit can only work inside chrome; it's designed to be connected up to a shim which talks using IPC to a browser process, which handles a lot of the work. Chrome and its browser process are not designed to be embedded, so using that in LightDM won't work. LightDM would probably have to use the Gtk+ port, which shares a lot of code, but has to be built separately due to all the ifdefs.

All x86_64 compilers ouput SSE, as x87 isn't supported in 64-bit mode.

Ken Jennings is getting $150000 (plus $150000 to his charity), while the producers of Jeopardy are getting a lot more viewers (and therefore a lot more advertising money).

And in bad times, all those loans are called in and no one can get new loans. Sounds like a foolproof plan.

Make cinnamon into a highly toxic chemical.

Thanks for giving us the formal analysis. It also states the point that has been mentioned: "The Control shaft (and associated mechanical elements) should be sized to this torque requirement accordingly - the Input and Control should be considered as parallel power paths rather than as 'power' and a 'control' elements respectively."

There's a reason that that document includes a CVT in the design to power the "control", because the power that goes through the control is the same as the power through the input, so unless you have a convenient high-torque high-power variable-speed electric motor you might as well use the main power source. Of course, if you use an electric motor then it''s quite similar to a Prius, and if you use a CVT then it's pretty similar to the Torotrak, or some other IVTs.

However, this device does have a new feature. There are two different control inputs with different properties, so by switching which one you're braking and which one you're controlling with the CVT you can either go in reverse or at top speed, but still with a CVT's output speed always being positive and less then the input speed. However, that makes it complex for no gain (you could just directly gear up the output of the CVT to do that), so it seems like the things that are new aren't good, and things that are good aren't new.

I'm sure they're just dying to get into that huge market that OQO failed to make any money in.

Many theories posit that the universe is grainy, but the graininess there is so small that we can't measure it. The holographic universe theory has, as a possible consequence, that the graininess is much larger than otherwise, so that it can be (and has been) measured by experiment.

My father went to state school in RI, and was recruited by Raytheon before he'd even graduated. He was working alongside graduates from all the Ivy Leagues, getting paid the same. It doesn't matter what the name on the diploma is, what matters is the effort you put in and the skills you provide for your employer. Save your money, avoid crippling student loan debt, and get those smaller class sizes anyways.

The graduates working at Raytheon were the same, but, as you said, they were also being paid the same. Couldn't the smarter students get higher-paying/better jobs elsewhere?