AI-friendly: you can use the LLMs however you wish, but you must thoroughly understand every part of submitted code.

Slop-friendly: understanding anything is optional.

That's into the absurdist territory, but it reflects the real pattern of behavior. People do go around committing petty crimes for various reasons.

Well Heaviside was a bit of a kook, especially in his later years, and that claim seems entirely in character :) My contacts with Fraktur are intermittent and brief, so I don't really have an opinion; I mostly encounter it in an old German-English dictionary which I occasionally consult, where the German text is printed in Fraktur. I don't have too many problems reading it.

If that's a reference to Thaler v. Perlmutter, the only thing that's been established is that an LLM can't be considered an author under the Copyright Act, only a human being can. It says nothing about the consequences of a human claiming authorship of LLM-generated code, which would be relevant here.

In other words, it’s not just a tool problem, any more than it’s a human resources problem or a leadership problem. Instead it is a systemic problem [...]

Shades of an LLMism, a bit padded, a quarter of a century ago. These days someone could easily give it a stink-eye. I'm sure that training has ingested this along with countless similar examples.

I for one do. Old habit when typing in longish C constants (ECONNRESET) or shell environment variables. I'm used to typing capitals by holding the Shift with the pinky of the other hand than the one entering the letter, so with long strings of capitals sometimes I'd have to switch for every other letter, which gets old fast. With Caps Lock, I press it, type in the letters, press it again.

Completion mostly works these days if you have it, but you don't always.

I'll defer to your judgement re optical properties, but I want to offer a counter-anecdote about practicality.

I've had myopia and have been wearing negative diopter glasses for over half my life. I've never needed vision correction for reading. This is not an unusual combination, and if you want a celebrity example, watch some old Apple keynote videos with the late Steve Jobs, who would repeatedly lift his glasses to read something on the phone in his hand. This "works", but can be inconvenient in some situations.

A while ago, I started thinking about progressive bifocals. Cursory web searches told me that my particular combo was impractical. My optician didn't see a problem, so I decided to trust him and got a pair made. The TLDR is that they work for me much better than the old ones. There was a period of adaptation. Going downstairs and looking at the floor/ground were a bit disorienting for a while, but i don't notice it any more. Switching between the monitor and the phone or paper in front of me works, which is why I wanted the bifocals in the first place. I only use the old glasses for watching TV, probably meaning that my TV-watching posture sucks, but fortunately I don't watch a lot of TV. I still take off the glasses for sustained book reading.

Build quality deteriorated (from impressive heights) more than 25 years ago, when HP's calculator manufacturing moved to China. Not on account of China itself, but it was definitely a cost-cutting measure, and higher-end calculators were becoming an endangered species even then. For example, keycaps used to be double-shot injection molded, so the legends could never wear out; no more, now they're silkscreened like with everyone else. The new key mechanism could never reach the robustness and reliability of the old one, which is a problem if you're used to every keypress felt in your fingertips being correctly registered.

(Not everything was premium quality. On my late 15C, the faceplate logo wore out and the soft sleeve crumbled to dust after a couple of years. But the machine itself continued to work flawlessly until an unfortunate accident with a space heater.)

Additionally, the new Voyagers (1x series) are not running on the original, custom HP "Nut" CPUs, but on ARM microcontrollers, presumably via firmware emulation. It's impressive that the whole things works so transparently, but as I dimly recall, there were problems with that emulation in the first 15C Collector Edition runs, supposedly fixed now.

So, if you buy a new Voyager these days, you're getting a convincing replica of the originals from the '80s, nothing more. Caveat emptor.

Who, then, understands the code? If the answer is "no one really", entropy will overwhelm your codebase sooner or later. Otherwise, you need to read the code, and for that the knowledge of language is still relevant.

It's a bit more varied, even in the Indo-European family. What does tend to happen is that the words for handedness get positive (right) or negative (left) associations in idioms, but additional meanings are not universal. In French, "droit" additionally means right (as human right), but not "correct" (yes it does have a bunch of adjacent meanings). In German, "recht" gets to mean "law" or "justice", shared by some Slavic languages ("pravo") -- but not all of them, which have the word "desno", without any association with rights/justice/correctness. The Latin "dexter" gave us "dexterity" and "dexterous", but also nothing alluding to justice. Et cetera.

As an aside, "left" originating from "left over" sounds like folk ethymology to me. Dictionaries point to "weak" as the original meaning.

They just need to exclude the shops in a 150 m radius from Terminal D in the Tallinn harbor to get accurate statistics, the Finns rarely go beyond that ;)

Secure as defined by a duo of monopolists. It's a contractual concept and doesn't have a firm relation to security-related characteristics. I'd trust GrapheneOS to be as secure as anything Google is capable of releasing, but that doesn't help them if Google refuses to vouch for a device running their OS. Which is also why your check/credit card analogy falls flat.

What else do you expect, given the economic incentives on one side, and the immaturity of the discipline on the other? Writing robust software requires time, money and competence, in a purely empirical approach, since we have no fundamental theory of software. The pressure is for quantity and features in minimum time. The approaches are incompatible, and economics win every time.