Why does someone have to be a "chump" to pay more for it? If the company's earnings have grown 10x then the company ought to be worth about 10x. Those earnings... do you realize they belong to the shareholders even if they just accumulate in the company's bank account? Dividends are a "feel good" disbursement for companies that aren't supposed to grow. If you want to invest in dividend paying companies go for it. Or you could buy apple and any time your holding is worth more than your basis, just sell some. You'd have done quite nicely even with that conservative approach, and yet still enjoy the feel of 0% capital appreciation that you crave.

Is "peak" the new "-gate"?

If you follow peak oil theory, yes. Basically it implies peak-everything. All resources become more scarce as the cost of energy rises. Not only the cost of production and distribution but even the raw feedstock for many products is fossil based, eg fertilizers, plastics, asphalt.

It's less about dealing with locally cloudy days than producing for 24 hr baseloads and being able to respond quickly to demand variance. Most importantly you want this plant to be able to pick up the slack for _other_ renewable energy generators when they get cloudy days or unpredictable winds. The only large-scale technology we have that can do that now is hydro.

it doesn't hold water in the case of a gyroscope.

I was speaking to professionally designed products as opposed to just a demonstration of the gyroscope principle. You sound like you have some EE experience but I wonder if you are seriously suggesting that someone would attempt an all-analog control system for such a thing. Even little remote controlled toy helicopters now use DSP for that purpose, and their part count is extremely low.

Sure you _can_ do a gyroscope with a few op amps, but if you're actually going to make a segway-style product then you also need steering, low battery safety, tip-over safety (ie jog it back upright if you're going too fast), and so on. It also needs to be reliable and manufacturable, and with a micorcontroller-based design that is a lot easier to do because you can have self tests (factory and power-on) that sanity-check your inputs and such. And you can do all of that very easily in a $2 micro.

I'm curious what actual products you can think of where such low-speed analog control systems are preferable to a nearly-free microcontroller. I have the habit of reverse engineering nearly every new electronic device I buy and have not seen that kind of implementation for a loooooong time.

The computing proof-of-work is a way to make the money non-free for the purpose of initial allocation _without any central authority_. Once a unit of currency exists, that processing is over and done with. And eventually the whole space is allocated.

Do you have a better way to do it? I'm not following how your proposal works.

Troll? Wow. I was just explaining for the benefit of non-EEs why it isn't usually done this way since the advent of inexpensive microcontrollers. This is hardly controversial.

It's not amazing but it's certainly harder. Probably higher part count, more electrical noise to deal with, harder to debug, harder to implement delays and state machines, more wiring etc. It's just impractical for most purposes.

How can anyone have access to a company's share price in the future??? Do you have even the slightest understanding of what you're talking about? Longer latency to exchange, for real trader vs HF trader. Seriously, that gives HF a few milliseconds view into the "future" because he knows what price is going to be offered immediately after his trade.

They might also be trying to break the internet just ahead of the next Wikileaks release.

If their TCP implementation kept a cache of recent final congestion window sizes by IP address, they could legitimately start off the next connection with the value from the last one.

Wouldn't it also be necessary to cache the _rate_ of transmission so you don't overflow some intermediate queue? Eg imagine your sever is on gigE, feeding into a 1 Mbps uplink, and then a loooong pipe to the client who is on the other side of the world. In this case you might want to have an initial cwnd of a few dozen packets, but if you were to fire them all out immediately at 1gbps you would lost most of them at the first hop even though it's less than the available bandwidth*delay.

So as I understand it this problem is more than just choosing the initial congestion window, it is also a matter of how fast you fill it. Normally that timing is driven by the acks coming back, but in the absence of that the sender needs to originate the timing.

electricity demand could increase significantly.

If it's night-time load that's not a bad thing per se - it means we're making better use of our capital.

Don't the Chinese do a spectacular amount of "grunt" lab work such as gene sequencing? I agree there could be a notable disparity in lab hours per dollar but it still fails to measure the amount of innovation and/or value that is produced. They may well be far head of us, but this sort of thing defies such a ham fisted analysis.

Judging scientific productivity in terms of patents filed is like measuring software value in lines of code. I realize that's not the only metric here but the fact that they're even looking at it this way is ridiculous.

Watts are a _rate_ of consumption, thus Watt-hours is a _volume_ of energy transferred. For example if a 100-watt bulb is turned on for two hours, then you have consumed 200 watt-hours.

Whether to use hours/minutes/seconds, and kilo/mega/giga, is just a matter of convenience - that's chosen based on whatever multiple people are most comfortable working with. For home usage "kilowatts" and "hours" are pretty sensible, don't you think?

kWh (dumbest unit ever...)

What's wrong with kWh? Would you prefer coulomb-volts, therms, equivalent-snickers-bars-worth-of-calories, or what?