You might have been right several years ago, but it's now pretty common for appliances to either allow or require internet access.

And this guy isn't most people. He's a software engineer. He should have known better.

He would have realized it had he bothered to download the PDF of the manual and read it. It says a cloud account is required for any feature marked with an asterisk. Basically, he didn't do sufficient research on the product before he bought and installed it.

Before I bought my EV charger, I made sure it could function just fine as a dumb charger even though it offers cloud connectivity.

I always just set an alarm by tapping my finger on controls in the Clock app, never using my voice. Once set (and I can see it's correct), it's never spontaneously changed it to some other time. So why not just use the app directly?

AI is a tantalizing tease. You can see glimpses of what it can potentially do, but then it hallucinates and produces something beyond useless and potentially dangerous.

Well, no. LLMs are what you're describing, not AI in general. Unfortunately, since LLMs are the new hotness, that's co-opted "AI." Perhaps Apple should try non-LLM-based AI.

And you're right about the junior and mid-level developers being let go.

The problem with that is it doesn't afford junior and mid-level developers the opportunity to become senior developers.

It is either 12 noon or 12 midnight.

Or more simply noon or midnight, neither of which ever occurs at any hour other than 12.

There will always be a need for plumbers.

I bought a Lombard specifically to run the Yellow Dog Linux distribution on. (This was before Mac OS X.) It worked really well. Even WiFi via a PCMCIA card was supported.

Breville, an Australian company, makes great toasters and microwaves.

My High School has two computer rooms: one with Commodore PETs and the other with Apple ][s. The head of the math department hand-networked the PETs to a single CBM dual drive depots them not being able to do this. You had to have a trophy on top of your PET to use the drive. If two or more computers tried to use it, it wouldn't be happy. It was clunky, but it worked. It also saved on buying drive for every computer.

There was also a paper teletype in that room connected to a remote system.

The Apple ][s each had their own drive since those were cheaper than the CBM dual drives. I programmed on everything and eventually got my own Apple ][.

The quote you're looking for: "Computer science is no more about computers than astronomy is about telescopes." - Edsger Dijkstra

The knowledge is irrelevant. Whether you learn from it is irrelevant. Whether you remember it is irrelevant. It's only the mere possession of an unauthorized copy to begin with that matters. Copyright is literally only about the copy. It's right there in the name. You're making this a more difficult concept than it actually is.

Assuming the entity you borrowed the book from had an authorized copy to begin with, the First Sale doctrine allows said entity to do anything they want with it: lend it, destroy it, put it under the short leg of a table, whatever. There is only ever one authorized copy in the current scenario.

Humans are allowed to learn from *authorized* copyrighted materials, i.e., you bought a book so you have an authorized copy. Meta is using *unauthorized* copyrighted materials, hence guilty of copyright infringement. But you would also be guilty of copyright infringement if you obtained an unauthorized copy of a copyrighted material.

That can happen regardless of who or what is driving.