that is exactly what the government does with the CAFE standards.

No, it isn't. The CAFE standards traditionally used the weighted harmonic mean of the mpg values, which gives exactly the same result as the weighted arithmetic mean of the economy expressed in gallons per mile. There are some other quirks- dual fuel vehicles are treated much more favorably than they probably ought to be, for instance- and the standards were recently changed to give bigger vehicles a break. But the larger point is that the EPA isn't completely stupid and does realize that the arithmetic mean is not the correct way of calculating average fuel economy.

One more case where SI units are easier to use. 1 liter/kilometer is 1 square milimeter. Isn't that so much simpler?

For what it's worth, the physical interpretation of this would be that a car with a fuel economy of a given area would be able to drive without needing on-board fuel storage if it were following a trail of fuel with that cross sectional area.

Employers want more to drive prices down, workers want fewer to reduce competition. Employers have more money and a better lobbying arm, so their opinion is the one we tend to hear.

It depends on the start mechanism of the ballast. Rapid start and programmed start ballasts give very good electrode life, but at the cost of reduced efficiency. Instant start is most efficient, but it substantially reduces electrode life. Given that lamps are generally rated for substantially longer life with programmed start than instant start, electrode life must be the limiting factor in at least some cases.

Who thinks it's okay to sit on their phone?

You can flip this around and ask what company bases their product on theoretical ideas about how people ought to use it rather than watching the way people actually do? I don't think it's sensible to drop a phone in water, but that hasn't stopped companies from making phones that are water and drop resistant. People in the real world also tend to put their phones in their back pockets, especially bigger ones that may not fit comfortably in a front pocket, and that inevitably means they get sat on. A company that makes a phone that's likely to be sat on needs to make it durable enough to hold up when that happens, or they'll be rightly criticized for failing to produce a quality product.

Yes, that's a premium price for a 28" 4K display. Dell is currently selling theirs for $430. Note, though, that the relatively inexpensive 28" 4K displays are using TN rather than IPS, which is a big reason they're relatively cheap.

Increased efficiency could actually help with cost, even if it makes the actual LEDs more expensive. First of all, improved efficiency would reduce the number of individual LEDs needed for a given amount of light, which would counteract some of the increased cost. Second, the LEDs are only a small part of the package, and improving their efficiency would make everything else easier. It would mean cheaper power electronics, which reduces cost. It would also mean less waste heat, which would mean a smaller heat sink, which is the single biggest thing in most LED lights.

And those electrodes are probably a big reason many people have much lower than expected lifespan for their CFLs. The electrodes undergo a lot of wear during the initial power surge when the light is turned on, so ones that are turned on and off many times per day will die long before their rated lifespan.

It doesn't decrease the incentive to produce software nearly as much as the threat of being sued for violating patents that never should have been granted. There's plenty of software out there that attracts customers by being good and doesn't need the threat of patents to succeed.

The contract terms will only work against actual customers, though. They won't do a thing to stop an enemy or prankster who hasn't actually bought the product or service, and consequently hasn't entered into the contract. All it will do is prevent people who are actually well informed from commenting.

How they've managed to make so many people believe that's the way coffee is supposed to taste is something I'll never know.

The best explanation I've heard is that darker roasts stand up better when you add stuff to the coffee. If you drink your coffee black, you probably want a mild roast or it will be too bitter to drink. If you dilute it with a bunch of milk, flavored syrup, and maybe drink it cold, you won't be able to taste the coffee unless it's roasted almost black.

That simply isn't true. The 4th Amendment says that:

no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.

IOW, they can ask for a warrant when they have a strong reason to believe that something contains evidence; they don't have to be absolutely certain. That's what "probable cause" means: enough evidence to convince a skeptical individual that something is probably true. It's a fairly strong standard- the person asking for a warrant needs to present some kind of evidence rather than just a hunch- but it doesn't demand certainty. That's why people who ask for warrants are not routinely punished when the warrants don't pan out; they only get in trouble if it can be shown that they materially misrepresented facts they used to support their warrant request.

You're thinking about the 4th Amendment right to avoid unreasonable searches and seizures, but cyborg implants potentially invoke the 5th Amendment right against self-incrimination. If the implant is actually a part of the person, some lawyer will argue that forcing the person to divulge the information on it is forcing them to testify against themself. When does the information in the cyborg implant stop being like information on a device like a phone and start being like information in your brain?

Nobody knows for sure. There was a recent analysis that purported to show that the King's gambit is a loser for white, but even that wasn't a completely exhaustive analysis. Instead, the analysts decided to prune any line that resulted in a sufficiently lopsided position as presumptively winnable, which reduced the analysis to something tractable. But even that was for just one possible line of play, and one that was considered relatively easy to analyze. Nobody has come anywhere close to solving the whole game.

You must not have looked very far, then, because checkers- also on the list- has no random element, at least when played from the standard starting position. In some tournament variations, the starting position is chosen randomly from a few positions with the first few moves already made, but beyond that it has no random element.

In any case, it's not clear that inclusion of a random element is a bad thing. One of the drawbacks of chess is that the lack of a random element allows it to be analyzed in depth in advance. That places a huge emphasis on memorizing standard opening libraries, which seems counter to the point of individual strategic skill. In contrast, games with a random element can't be analyzed to the same depth in advance. That forces players to adjust their strategy on the fly rather than relying on somebody else's analysis.