Cars have about a 12 year half life. The composition of the cars on the road will be subject to that. So even if 100% of the cars being made suddenly were autonomous it would be a long time before most of the cars on the road were autonomous. The insurance industry will have a lot longer to see the changes coming than implied here.

The long life of cars is such that I can't imagine the security mess that will come with owning a 10-20 year old autonomous car. People gripe now about XP not getting security updated, but imagine still having Win95 based vehicles still on the roads that are 10 years past EOL support. People will be irate when their Chevy HAL gets an over the air update that disables the autonomous function after the company lawyers freakout over a hard to patch vulnerability.

Somehow people think autonomous cars will only leak pixie dust, and that people will happily throw away their drivers license and just take Johny Cabs everywhere.

Or you could just rent a gasser for the few times a year you take a road trip.

Seriously, we drive our crappy little Leaf more than our other two cars combined, and by a large margin. We take the crappy Ford Escape for longer local trips, and the Tundra truck for hauling our camping kit, or a load of crap from Home Despot. It is not a stretch to expect that we could readily get to 20-25% electric market share even without road trip capability, but the range does need to be more like 150 real EPA miles minimum for folks stop feeling like they are taking a big risk.

"Driverless cars are the revolution. Electric makes existing car use patterns better. Driverless makes an entirely new paradigm. It may eliminate mass car ownership. It might eliminate parking lots. It might eliminate light rail in suburban areas. Taxis. Deliveries. Shipping. Police reponses."

Huh?

Autonomous cars won't need to park? Do they warp to a distant land when you don't need them? Even Taxis have parking lots, though usually private ones, for when they are not operating. Similarly you see a lot of taxis sitting on the side of the road waiting to be assigned, the same would be necessary for a Johny Cab fleet. The fleet has to be big enough to handle peak usage, which is likely >10x what you need on the road at 3 AM.

Why would light rail go away in suburban areas? Often these are people taking the train because it is less of a headache than driving into the city center during rush hour (right, your idea is that autonomous cars somehow vanish when not in use...).

Deliveries and shipping will disappear?! How would that work? I buy my widget and it never comes? Thanks a lot for nothing.

Police responses go away? Anarchy here I come, HAL has eliminated the police! Do you expect machine gun turrets on top of autonomous cars, maybe automated stop and frisk bots too?

We need a Godwin's law for people who use "paradigm" in a sentence in place of "Effing miracle happens here".

Take my money and give me my receipt. If you put gimmicks around that process to up sell me, you are pretty far down a slippery slope.

GaAs and GaN have some real advantages over silicon, but price and density has not been their strong suit. Silicon is cheap per unit of area, and is compatible with copper metallization. GaAs in particular has always been sensitive to copper contamination, hence the use of gold for most of the interconnects. GaN has very fast switching speeds while handling an order of magnitude more voltage than GaAs (~10x the breakdown for the same Ft). So for a ~100 GHz Ft silicon can handle just a volt or so (think 65 nm nodes and smaller to get those speeds), GaAs can handle about ~5V, and GaN is more like 40V.

The real story here is the ability to run GaN with existing silicon equipment with silicon type costs, rather than the relatively low density fab equipment GaAs and GaN are usually processed on (usually the repurposed cast-offs from old silicon fabs).

GaN by its nature operates at high voltages, like 15V minimum, so don't expect a GaN based processor any time in the near future. Instead it sounds like they are claiming a 100x improvement in the Ron*Coff figure of merit for power switching devices. The payoff would be smaller and more efficient power electronics, such as the mentioned laptop supply brick, electric car power electronics, etc. Don't expect this has any bearing on the plateau of Moore's Law.

So IE is old and badly maintained, so just "trust us" that our new Edge browser will get better maintenance and support after it replaces IE. Never mind that it will be the same pile of schlubs maintaining the thing as were maintaining IE. It will be different this time, just "trust us".

If MS ever officially apologizes for the Ribbon interface and throws it into the same burn pile as Clippy, then I might start trusting them a little more. So far they have a nasty habit of shoving garbage down their customers throat, being horrified by the public outcry, temporarily mending their ways, then reverting to old behaviors on the following product cycle. But "trust us", this time is different, we promise to force updates that won't wreck your machine...

Was the drone black?

I'm fine raising the price of gas to a natural level (no subsidies). Oil companies don't need the welfare anymore. Subsidies in general bug me. Once they are in place they never go away. Modest tariffs to add a little friction at the borders is also a good thing if you ask me, it make it harder for big companies to arbitrage the labor supply.

I'd rather see funding in general come from taxing the top 5-10% more, especially the top 0.1%. Even independent of adding solar panels I would like to see the top income brackets taxed much more, with the funding going to infrastructure and education. I'd also like to see military funding rolled back, which is more practical if we manage to become less dependent on unstable countries for our energy needs.

What was your point again?

So if we had 100,000 workers installing panels, they would only have 7 hours per panel to install it?

Horrors. We might need to get unions involved to keep people from exceeding that rate and put hard working solar installers out of work.

Wrong. Most people forget that there is a time dependence on the supply demand curve, and often an NRE portion. Supply/demand works as you state as a way to describe what will happen in the short term.

If you sell 1,000 cars you have to charge more than if you know you are selling a million of them. The miracle of mass production shows up when you can ammortize the development costs over many units. The more you sell, the more you can spread that cost around and more reasonable spend extra development money to lower the production costs (make a custom battery plant, buy robots to replace workers, etc).

So over time you can get 1990's Ferrari performance for Chevy prices. Many of the major safety and convenience features in a Chevy started out in luxury cars and trickled down as the volumes went up and the prices came down. As volumes go up you find that there si higher integration and a drop in the per unit price, which a textbook beginner supply/demand curve is inadequate to describe.

Who pays for your current power plant when it needs maintenance, repair, or replacement? We have these things called "utility companies" that have the charter to keep the lights on (some do a better job than others). Similarly if utility companies are driven to put in energy storage by laws or incentives, they would pool user fees (you do have an electric bill, right?) and pay for ongoing maintenance and repair of panels and storage systems.

Today we have subsidies, the argument being made is to adjust those to favor solar instead of coal, oil, and natural gas. Utility companies are largely driven by rules to best serve their customers. If subsidies make solar the cheapest option, the change will happen as old power plants wear out and get replaced.

But surely you are not as dumb as your comments appear to be?

So a 14x14 mile square? Put a few of those in the Mojave desert. Put many more out in Eastern Oregon, Texas, New Mexico, etc. We have many large and mostly empty places in the USA.

Let's get that started, and also start funding/subsidizing the heck out of energy storage projects to help get solar to fill in more than just daylight hours.

That has been proposed for a while. However, it takes huge amounts of energy to get that many panels into orbit, and space is still a very harsh place do construction work.

Geosynchronous would allow satellites to hover over their receiving stations, but the distances are quite large, making it very hard to make a narrow enough beam to power a city a couple hundred miles away without also cooking it.

Low earth orbit is better, but now your giant death ray is moving, and must track one of many receiving stations as it orbits. No easy task either.

Ground based solar is vastly more serviceable, but needs an upgraded power grid to push around power from where we have it to where we need it. Storage is still needed if we want to go beyond the ~20% point, but generally solar can fill in the peak usages caused by summertime AC use. There still is a lot of low hanging fruit for solar to help with, and the limitations of storage shouldn't be used to distract from us from using solar up to that point.

He's a loonie, and so far the only candidate I actually want to vote for.

I don't think that is a statement on how good of a candidate he is, but rather just how awful the other choices are.