He didn't limit that to just the Amazon.

Not being readable even when someone has the device in hand is exactly what these secure FPGAs are meant to protect against!

It's not a non-issue. It's a complete failure of a product to provide any advantages over non-secure equivalents.

It's not a matter of experiencing more gravity though. The sun has more pull on the moon than the earth. That doesn't mean the moon gets pulled into the sun. The hydrogen on the outer layers of this sun would still have more or less the same orbital velocity as the rest of the sun. There has to be something disrupting the orbit.

Good maths and all but there's one thing you need to consider- If you're in stable orbit you don't actually fall inwards.

The sun for example has twice the pull on the moon as the earth (do the maths and see for yourself). It doesn't fall into the sun because it's in a stable orbit.

Likewise in this example. It's not a case of the black hole pulling more than the sun at a given distance. It can, but it's not all that relevant, plenty of orbiting bodies have more gravity pull from a nearby larger mass than they exert themselves but that's not what determines whether or not something gets pulled into the larger body.

What does determine whether or not something gets pulled into the larger body is if something disrupts the orbit. In this case the most likely culprit is charged particles from the event horizon stripping the sun of its outer layers.

Gravity alone doesn't selectively pick out hydrogen and leave helium behind. I'm guessing that's more the usual atmospheric escape when an object gets too close to a mass of charged, high velocity particles. Earth can't hold onto it's hydrogen against the suns solar wind. In this case a sun can't hold out against a black holes radiation.

A computerised system as ruler is a great idea. But to be a true democracy it has to be open source with democratically voted in patches to the source code.

Say the Ruler starts developments in public parklands that the majority doesn't want. Just submit a patch "-if (isParkland() ) develop(); +if (isParkland() ) protect();" Have patches voted on at each election cycle. The patches that get a majority go into the codebase and the Ruler program then runs this new patched code that people have elected it to run.

A few hundred years of evolution of such a program should produce a great Ruler. One that isn't susceptible to corruption or greed but instead rules exactly to a prescribed program that has been refined to perfection.

Yes I really do look forward to our computer overlords.

My Nokia N900 does run Linux actually.

That shouldn't be modded funny. I remember the Wollembi Pine retailing for $100 a sapling here in Australia a few years ago. Each sapling was numbered. There was a lot of novelty in having the Nth Wollembi Pine in the world.

There's a hell of a lot of money to be made here.

Can't find exact install base, but The Legend of Zelda: Majoras Mask required the Expansion Pak and it alone sold 3million copies. So 3 million at an absolute minimum to get a ballpark figure going. Plenty of other games highly encouraged people to get it too.

The N64 memory upgrade would be an actual example of a successful console upgrade. Plenty of people bought that and it was well supported.

Devices using the public spectrum should be forced to detect other devices in range and share the spectrum evenly with those other devices.

Any devices not sharing the spectrum evenly would be banned.

Just record all the transmitted data and you can decrypt in half an hour. The cluster will just let you listen sooner but it's unnecessary.

(i am assuming it doesn't do frequency hopping since it's working in a narrow satellite band).

Maybe no good in the traditional sniper role but if this weapon ever gets to the point where it has an effective range that's much greater than any small arms fire then it will have a new role all of its own. That role will be called "shoot the Taliban and laugh at the counter fire".

Removing the instruction decoder is a bad idea. The microcode is different for every CPU out there and it wouldn't be as fast without an instruction decoder in any case. CISC tends to be memory efficient while RISC tends to be processing efficient. Thanks to the instruction decoder, the x86 currently gets to store it's instructions CISC style while running them RISC style. If you had to write your code using the microcode directly you'd end up with ~100bytes of code for something as simple as the equivalent of the FSQRT instruction. That means cache misses and increased memory use.

The instruction decoder doesn't even use much power in any case and is necessary for many modern CPU features. In fact many ARM CPUs have instruction decoders that support more opcodes than the x86 decoder (standard ARM instructions + THUMB + Jazelle, etc. tends to add up to more than what the x86 has to deal with).

An argument could certainly be made that Intel could support a new instruction set on top of its current CPUs that's more memory efficient than x86 and i'd agree with that (although it'd require yet another run-mode and i'd hate that). But bare metal programming isn't really realistic or even any faster.

They tried that. They are currently trying to sue Apple and RIM for over one beeeellion dollars. Sun also paid them some money to get the Kodak lawsuit company off their backs after Kodak claimed to own a patent covering a program that gets help from another program.