...I can see this could be useful for preventing reviews that claim I can't spell or do grammar.

It needs to be opt-in, not opt-out - not only for fear of butchering an author's beautiful sentences, but also because some authors who self-publish in English are successful enough that publishers of books in other languages sometimes pay for the right to publish a translation. Those contracts are usually exclusive, meaning that the author agrees not to let anyone else publish a translation of that book in that language. If Amazon's AI generates one without the author's knowledge, the author could get hit with a breach of contract lawsuit.

I imagine the term was invented in some meeting where a super pedantic engineer was dismissing all other concerns because they were not on his list of "functions", and in desperation the rest of the people said, "geez, ok, look, these are non-functional requirements..."

You could well be right :-) I've noticed that names tend to stick, even as the thing that the name refers to changes, so that the name doesn't really fit any more. One example that comes to mind is RAM versus ROM. RAM is random-access memory, memory that you can access in any order, instead of having to access in sequence, like a tape. ROM is read-only memory, but that's random access too. If you say "RAM", people assume it's writeable, so why isn't RAM called RWM? Or why isn't ROM called RORAM?

(My guess is it's because early computers often didn't have any ROM, and to boot them, you had to enter a simple program into the RAM using switches on the front panel. That program would then load a more sophisticated program from tape or punched cards. When computers gained some storage to hold that simple program, the thing that distinguished it from other types of storage was that you couldn't alter it.)

The way someone explained it to me was that "functional" refers to the reason(s) why the system exists. What result is the user or customer trying to achieve? A non-functional requirement is something that the system needs to make it possible or practical to meet a functional requirement, but that would be of no use on its own. A system for administering loans needs an audit trail to comply with the law and to detect and prevent fraud, but there would be no point in having an audit trail on its own.

On the other hand, from the point of view of someone in the audit team, being able to audit the accounts is a functional requirement. Maybe everything is a functional requirement to somebody.

True, but Perl isn't used for many new projects these days. Python developers are much easier to find than Perl developers, and probably cheaper, which is what this exercise is really about.

The problem was made much worse by the fact that the UK Post Office is allowed to bring its own criminal prosecutions. If they had to do what everyone else does - ask the police to investigate, and the police then have to convince the Crown Prosecution Service that there's a reasonable chance of getting a conviction - many of these cases would've been thrown out for lack of evidence.

I heard about it when it launched, but it doesn't seem to have made much of an impact. I think it was meant to compete with dedicated websites for serialised fiction, like WebNovel and Dreame, and maybe even Wattpad. But you had to pay for episodes with some Amazon-only currency that you couldn't use for anything else, not even other stuff that Amazon sells.

Amazon didn't make it very appealing to writers, either. I write fiction (as a hobby-that-sometimes-pays-for-dinner-or-beers). Amazon might've relaxed the rules since I looked at Vella, but I think I would've had to write something specifically for Vella. They wouldn't accept anything that had been published anywhere else, not even on their other exclusive platform, Kindle Unlimited. (That one's been a runaway success. Maybe they thought it would be easy to replicate with a different reading model.) The Amazon-only money made it hard to work out how much I'd get paid for each chapter or story that a reader read. I didn't bother looking into whether I'd be allowed to publish a Vella-specific story somewhere else after some time. The idea of having to write exclusively for an as-yet-unproven market was enough of a deal breaker on its own.

More telling, perhaps - I'm in various writers' groups on Facebook, with many members who are much more successful than me, and I don't think I've ever seen anyone mention Vella as a viable marketplace for stories.

The announcement of what they did in July says that NLNet has given (or will give) them €26,500, or about US$29,000, which suggests it's rather more than a hobby project.

The photo shows it with shoulder-length hair, which in microgravity is a hazard for getting tangled in things like air vents or the robot's own hands. It could just as easily have been made with short hair, or no hair at all. Or is the newsreader look just for the publicity photos on Earth, and they'll swap the hair for something more practical before launch?

The janitor has the capacity to enter into contracts, because he's an adult human and (presumably) isn't mentally impaired in any way, but he doesn't have the authority to enter into contracts on behalf of his employer.

I don't know exactly how it works with online shopping, but I assume that when you place an order, certain employees or managers have the authority to cancel it within a short time, and if they don't, a contract is formed by default between you and the company. There were a few lawsuits about this in the early days of e-commerce, when database glitches or bad programming caused expensive items to be listed at very low prices, customers ordered them, and the company refused to honour the deal. The company's lawyers argued that the contract didn't come into existence when the user pressed the "buy" button, but rather when the company billed the customer's card, or when the company sent an email to say the order had been accepted.

If a court ever rules that an AI can enter into contracts, then companies that use them will update their rules to make it clear that their AIs don't have the authority to do that.

Amazon are presumably confident that there's no easy way to trick their website into selling you a TV for a dollar, so they just wave everything through. If a company wants to use a chatbot in its sales process, they'll just have to post a disclaimer that either it's not allowed to negotiate on price, or that any discount you persuade it to give you has to be approved by a human. This isn't really any different from bricks-and-mortar shops, where the janitor isn't allowed to sell you anything, or the cashier can help you to decide what to buy, but isn't allowed to sell it at a discount just because you asked nicely.

In many countries, though not in the USA, publishing a work under someone else's name without their permission is an infringement of their moral rights. ("Moral rights" in those countries' IP laws are a specific concept relating to respecting or preserving the wishes or reputation of the creator of a work, not a general theory of how people should behave.)

Friedman might have been able to sue Amazon and/or the author of the fake books for libel, but that would have been time-consuming and expensive, assuming the author even lives in a country that pays any attention to the US courts.

"End-to-end encryption" normally implies that the service provides encryption for its users' messages, but doesn't keep any of the keys on any servers that it controls. Keys are held only on users' devices, and encryption and decryption happens there. So if law enforcement want to read a user's messages, they either have to seize the user's device (and defeat its security), or find a flaw in the implementation of the encryption, or brute-force the encryption.

That's the sort of encryption that this bill is aimed at - the service could have decided that it would be able read its users' messages, or pass them on to law enforcement, but has "deliberately blinded" itself by choosing to make that impossible. If, say, you and I agree that we'll use our own encryption for messages sent over an unencrypted service, then I guess technically that's end-to-end encryption too, but that's not what the term normally means. The service wasn't involved in our decision to use encryption, so it hasn't "deliberately blinded" itself to what we're saying.

No, but those don't (or didn't) have end-to-end encryption built in, so if the authorities decided that you were using them to plan or commit a crime, it was relatively easy for them to find out what you were writing or saying.

Or as Brian Kernighan put it, "Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it."

So ZO'R TV doesn't have to provide evidence that they own the copyright to the material?

In theory, yes, in practice, probably not. If I file a DMCA takedown notice alleging that some film A that you uploaded to YouTube infringes the copyright in some film B, then the part of the notice where I assert that I own the copyright in B (or I'm authorised to act for B's copyright owner) is made under penalty of perjury. So if I lie about that, I could go to jail. But I've never heard of that actually ever happening.

All the other parts of the notice, in particular the part where I allege that A infringes B's copyright, are not made under penalty of perjury. The official reason is that in the general case, the question of whether A actually infringes B's copyright can be answered only by a court, but you can supply your own conspiracy theory.