No question about that. On the flip side, I'd argue that those power projects are corporate welfare, making the entire country pay for power generation that is used by only a small percentage of the country, for the primary benefit of a few power companies that happen to get the grants. I'm not sure that's really a good use of government resources. Power companies should pay for their own construction, or else they should have to pay back the money to the people with interest.

One of the biggest fiscal mistakes in our country's history was spending so much money to build private power and communication infrastructure with public funds. If the government pays for it, the government should own it and lease it out for public benefit without taking a profit. When our government has done it this way — various municipal fiber projects, TVA, etc. — the results have been high levels of efficiency at a low cost. When our government has done it the other way, the results have been monopolies that have to be broken up.

Cancelling projects is frequently stupid because of sunk costs, and I would bet good money that the current administration did not do adequate analysis to determine whether this is the case, because they have a long history of failing to do so, but that doesn't mean that they aren't right to question that spending.

What we need is a few dozen clones of TVA in various regions of the country, operating in a not-for-profit fashion as a government-owned corporation to build and maintain power infrastructure. Federalize as much of the infrastructure as possible, make all future construction paid for by the government be done through one of those companies so that private companies don't solely reap the benefits, etc.

The real problem is that Republicans scream "Socialism", so Democrats try to work around it, and the result is corporate welfare, where everything is as inefficient as possible.

I'm laughing at the failure to recognize that COVID was the driver of those deaths, not the vaccine. That's why the overall death rate in the U.S. actually dropped by about 5% in 2022, making the increase predicted by that website rather laughably wrong.

Before the bots, it was troll farms. Now the troll farmers have lost their jobs, replaced by software. Plus ça change, plus c'est le même chose.

Easier is relative, though. All the nuclear weapons on Earth would still be two orders of magnitude too little to get an adequate atmosphere. As I understand, you'd need several thousand gigatons to get a low single-digit percent of Earth's atmospheric pressure.

And for humans to survive for more than about a minute even with external oxygen (the Armstrong limit), you'd need to reach about 40% of Earth's atmospheric pressure. There's probably not enough CO2 ice on all of Mars to pull that off. Best guess is that you'd need four or five times as much just to reach that limit, though the best-case estimates would result in exceeding that limit by a factor of two, so there's a lot of uncertainty here.

Whether releasing a lot of that CO2 would cause enough of a greenhouse effect to melt more polar ice is unclear, but one would assume that if this were possible, the planet would not have cooled, so that seems unlikely. Chances are, you would have to melt *all* the ice and periodically add energy from some external source to re-melt it as it forms, or else built planet-sized mirrors in Mars L4 and L5 to increase how much sunlight hits Mars.

Surviving Martian soil is the easy part. Being able to survive at near zero air pressure is the hard part.

That's not what a cancellation fee does, though. By definition, when you pay a cancellation fee, they are no longer providing service.

Because that $2,000 is consideration for the other party providing something. If the penalty clause is the entire remainder of the contract fee, then the other party should also be compelled to provide service for the remainder of the contract term, or some equivalent consideration. Otherwise, it isn't really much of a contract.

If they get out of providing service, then you should get out of paying, except for some penalty to make up for sunk costs, e.g. the prorated cost of provisioning initial service, the prorated cost of a phone that was free with contract, etc., plus some *reasonable* amount to discourage people from pulling out of the contract on a whim.

Also, understand that the company providing the service had way more power over the contract than you. You were almost certainly told "take it or leave it" when presented with the contract. That's why putting limits on what contracts of adhesion can do is generally considered to be a critical function of government.

The part you're missing is that contracts like this are contracts of adhesion, and there may or may not even be an option to sign up one month at a time. And even if there is, having a penalty clause for canceling a contract is reasonable, but having a penalty clause that massively exceeds any plausible damages isn't, particularly when one of the parties in that contract has dramatically more power than the other, and that party is the one writing the contract and demanding the penalty clause. That's why it is reasonable for governments to limit the amount of those damages through statutes. It is just compensating for that inherent power imbalance.

Also, real-world companies aren't typically selling bonds against their subscription revenue, and unless this is a very small business and the contracts are among equals (which a customer relationship almost never is), a bank isn't going to care about the difference between 1,000 subscriptions and 1,001, nor do stockholders. They care about the difference between 1,000 and 100,000. Orders of magnitude matter. A few cancellations around the margins are noise. So although you might be correct in theory, in practice, single cancellations don't matter, and if the cancellation numbers are high enough to matter, there's something much more seriously wrong with the company, and locking consumers in to a long-term contract likely serves no one's best interests, including the company's, because that just reduces the pressure on the company to fix those structural problems.

The thing is, if you stop using the service after two months, they aren't providing you a benefit, and it isn't reasonable for them to keep collecting money. And charging exorbitant fees has the net effect of forcing people to continue paying a month at a time because they can't afford the cancellation fee all at once. That's what makes it predatory.

If we were talking about a small company, where someone canceling service (e.g. a maid service) would mean that they have to go seek out other clients to stay in business, then charging such a fee makes sense. For a big company, it is really rather hard to justify.

This is doubly true if the company either does not offer a month-to-month plan or charges only slightly less for it. At most, you have cost the company the difference between the yearly contract and month-to-month price, and if the penalty is greater than that difference, that's really not right.

They will if they want viewers over age 30. I like shows where I can just keep watching episodes one after the next for a month, so nearly everything I watch is a decade old or more. It's not worth my time and effort to figure out whether I might like to watch a show if I'm going to run out of episodes in a day of viewing.

I can only think of one show that I've watched when it had fewer than 20 episodes, and I regret not waiting longer, because it would have been much more enjoyable watching three seasons instead of two a year from now. Modern shows require too much effort for too little payoff. The threshold where I feel like it is actually worth my time is about 40 episodes. And most new shows will be canceled long before they reach that threshold, which means most new shows aren't worth my time.

Yup. I never said it made sense. They're several years too late, and should have shut down that purchase before it happened. But I'm thinking about some quote along the lines of "The best time to figure it out was years ago, but the second best time to figure it out is today," or something like that. :-)

It's a Dutch fab company that got gobbled up by a Chinese company. It was originally part of Philips. Foreign companies didn't create jobs in this case, unless the fab grew significantly in the last six years. If anything, they've been selling off some of their existing fabs.

From all indications, the main purpose of the takeover was to prevent sending technology secrets to China, and possibly to prevent the illegal sale of their chips to Russia.

Nothing to see here. Move along.

This. The dot-com bubble was still a bubble, but there were real companies producing real websites, some of which were even useful.

It's the same thing with AI. The fact that the technology exists and sometimes is used for things that are useful doesn't change the fact that there's a *huge* hype bubble around AI, and everybody and their mother is dumping piles of money into AI, hoping that they'll get lucky and back one of the winning horses. All the while, companies with no real business plan other than "AI" or "AI first" or whatever are hoping the investors won't notice that the companies really don't know what to do with AI, and they're just hoping that if they build it better, first, they'll "win" or whatever.

It's a bubble. It's a huge bubble. I have no idea when it will burst, but it will. They always do.

Did Bose sell non-cloud speakers? Yes? Then they paid a premium because of those features. The fact that streaming to multiple speakers requires the cloud is, by itself, a major bricking of basic functionality that probably destroys the entire utility of these speakers for a large percentage of their users.

Yes. If the company is competent, yes. This is all just trivial server logic. You maintain a fixed frontend/server endpoint library that parses the inputs and ties it to your backend systems. If you rewrite a backend (which should be rare), you update all of the frontend code to translate the data as needed. If you want to change the way your data is passed between the device and the server, you fork a new endpoint library.

The ongoing cost to support older devices, then, is an hour or two of maintenance work if you do some major backend rewrite, or pretty much zero otherwise, because you're continuing to keep the servers alive to provide services for your current devices.

So when a company says that it is turning down support for older devices because it can't afford to support them, what it really means is either A. they have saturated the market, and the only way to get people to buy the old devices is through planned obsolescence of the old devices, B. they have found some critical security bug in the old firmware and nobody knows how to build new firmware for the hardware anymore because the toolchain won't run on their current operating system, or C. they're incompetent and don't know how to design servers properly.

Either way, it's a strong reason to never buy their products again.

I mean, yes ostensibly, but the reality is that cloud services shouldn't be *allowed* to disappear unless the company is either going out of business or is shuttering an entire business unit. As long as Bose still makes any cloud-based speakers at all, that's really not okay.