quarterwave

This is in fact a profound question.

The physics of electricity propagation in a powerline circuit is fundamentally the same as the propagation of FM radio waves, or even the beam from a flashlight. All of these examples involve electromagnetic energy propagating at the speed of light. So why do we need wires for powerlines, but not for propagating radio signals or light beams?

The principle is that light (in a vacuum at least) travels at a constant but non-infinite velocity, c. Hence, the electrical wiggle received at an observer's location now i.e; (x,t), has been caused by some earlier wiggle conducted by the source (x',t') such that the t-t'=(x-x')/c. We should expect that the effect of this 'time delay' is more profound if the source is wiggling faster in time.

A more specific way to state this is that the electromagnetic power radiated into free space by a dipole increases as the square of the dipole moment and the fourth power of the frequency. Now compare a powerline (60 cycles/sec), with FM radio (100 Million cycles/sec), and with the light from a flashlight (500 Trillion cycles/sec). That's why even atomic dipoles can produce intense visible light, while it would take a very very large dipole to radiate a similar intensity at 60 cycles.

Feynman Lectures Vol.II is absolutely the best reference to learn this stuff.

That is not to say that a bird needs to make contact with two wires on a utility line, in order to suffer harm. If the line voltage is high enough (e.g; 110,000 Volts, as in high-tension power transmission), an electrical corona would form around the line from electrostatic effects. The corona is actually ionized air indicating the high electrostatic field strengths in the region, it can emit a bluish glow and growl at 60 cycles. Birds can no doubt sense this corona & stay away.

Erlang = Owls (message passing)

Crystalline III-V channels can be grown epitaxially on silicon substrates, with buffer layers to grade the strain due to crystal lattice mismatch. With silicon wafers heading to 450mm diameter the economics would argue against native III-V substrates.

One advantage of native III-V substrates is they are semi-insulating (very high resistivity) so there is no need for transistor isolation wells. However, insulated substrates could be obtained on silicon by means of wafer bonding with an intermediate dielectric layer.

Thanks, good point about telcos.

I have home back-up power based on an inverter charging a lead-acid battery (located in a sheltered area outside the home), which costs about $100/kWh. Usage is about 1-2 hours discharge per day. No matter how well serviced, I've found these batteries don't last beyond four years. Hence I'd pay even $400/kWh for a well-engineered deep-cycle battery that is safe, maintenance-free, and will last at least 10 years. Excluding balance of system, even.

This is an excellent 'road map' to the two key theorems of information theory.

The focus on decoding complexity in the noisy coding theorem is particularly welcome. A separate article amplifying just this aspect (error exponent, Pareto complexity, etc) would be welcome.

Thanks for this reference.

Thanks for the detailed perspective.

My limited & roseate view of a 21st century Lisp machine is based on an old theme - a massively parallel computing system using bespoke silicon logic blocks.

As you have noted below, not only are the cache sizes in a modern CPU monstrous, there's also the compilers optimized for these caches, instructions, branch prediction units, etc. No point in ending up with a chip that is much slower than an equivalent one running on a specially-designed virtual machine, which is itself much slower than MPI.

Dreaming on, such a Lisp machine would need a vast collaborative academic effort with substantially new IP design, in say the 32nm silicon process node. That's the most advanced node where lithography is still (somewhat) manageable for custom IP design.

Any chance of a resurgence in Lisp machines? Especially in view of the changes in CPU architecture due to semiconductor scaling challenges.

i used to be a college teacher for several years & my sympathies are with the author.

Advice for those considering this path in science/tech - learn to write code on the side. Pick something mainstream that will be around for a while, and which you can tap for a sideline. Develop deep expertise, spend as much time continually educating yourself as you do for others.

That's correct, thanks for pointing it out.

Is there an Erlang shell to evaluate expressions?

There seems to be a sub-window to the right of the module editor, but I'm unable to place the cursor there.

Tried in both Firefox 35 and Chrome 39.

While the Jeeves stories are good, they can't hold a candle to the time when lawlessness raised its head at Blandings Castle, or to the effect of Mulliner's Buck-u-Uppo tonic on the Bishop and the Vicar (quoting from memory, e.&o.e):

"Tell him we're a couple of cats"

"We're a couple of cats"

"Oh, that's all right then" said Mulliner as he stood aside to let them in. The Bishop, being an artist at heart, mewed as he climbed in, to lend verisimilitude to the deception.

A perfect storm of wonderful English prose with a boundless absurdity of form and circumstance.