Good and general thoughts.

I've come to mine from the position of having at least two prior careers ended by the march of technology. Not bitter about 'em, but they were kind of a scramble (for the first) and frustrating (for the second).

I used to make beautiful and challenging special effects slide work on what has been called variously a "Rostrum Camera", an "Animation Stand" or an "Optical Printer". I'd have held my skills up to anyone else's in my large metro at the time. Then came "Genigraphics" - GE's computer-generated slides. They looked frankly, like crap. Those in my profession thought "OK, but they'll never be so good they'll displace us". But they turned out to be "good enough" for most business people, much faster than what we were doing. And eventually I saw the initial images coming out of Photoshop and realized I had to scramble for a new career.

That career was rich computer-based multimedia (on CD-ROMs in the marketplace, but not exclusively). I took my creative skills, retooled with software for making images and 3D animations and multimedia programming. I worked happily enough in that for maybe another decade before the Web really started pushing on me. "OK, but the web will never be so good it displaces rich multimedia". But in fact, it was obviously good enough for everyone else. And it eventually became media-rich enough that the distinction just doesn't even apply.

In both examples: Speed in the first, and Connectedness in the second, the advantage I didn't completely understand was what undermined my craft as a career. I was just as good on day N as day N+1. But I came to understand that it's vanity to think I couldn't be replaced by something "lesser", even if it had an advantage I didn't.

"Oh sure, but they won't have..." - yes, it may take a while, but eventually, they will.

Just gotten used to my cheese moving (and happily no longer in the workforce during what's bound to be a turbulent t ime).

Worry for my kids, though.

With great respect for your skills and your long tenure as a coder, a couple other ideas here:

1. We're all (me included) looking at this at a moment in time, at "AI", however it's construed, in mid-2025.

Have LLM's and their recent ingestion of the Internet reached the point of diminishing returns and we've reached a stasis? Or are we soon going to see more punctuated equilibrium and AIs that are advanced enough to do the whole thing buglessly from a product description?

I really don't know, and I've seen positions advocated from all over the spectrum. But I do realize that the main constant in the world is change.

2. Can you call someone a "machinist" if they've never performed a taper by hand on a screw-driven lathe, but only fed stock into and programmed a CNC lathe to do the job?

I think that a machinist today probably has to do little machining by hand, or that that specialty is a boutique trade.

So maybe a machinist is someone who knows something about the domain and can use the tools to do productive work.

The world changes around us, and orphans our crafts all the time.

This. Claude is flat out great at the boilerplate. Certainly better than me: it comes up with more thorough solutions than I would ('cuz I'm lazy), and in instances where I'd have to dig into arcane platform docs (multiplatform stuff), it just knows.

Lets me get to the _real_ work of a project far quicker.

Answering my own question: changing the default admin password apparently IS a mitigation for the issue that firmware update cannot fix.

https://support.brother.com/g/...

If simply changing the admin password (as I have) is enough mitigation.

Is it possible to generate a magic password that opens admin access regardless?

I think one of the issues is that the printer can become a platform for creating DDOS attacks, outbound from your network: part of a botnet.

Placing it on your local network is protection from somebody accessing it directly from the Internet, but not from automated attacks from other affected devices on your network.

Actually, Brother printers have traditionally been one of the best responses to the razors-and-blades scam.

Yes, they sell toner (for mine), but there's no vendor lock-in that I've been able to find (unlike HP, which I'll never buy).

They're solid, great value plays, for not much money.

What, are you just not going to have a printer?

Well...yeah!

Didn't know about this. Very cool.

Well, maybe over the network: random as a service.

You of course lose provability, but if true randomness became necessary for some future phone app, you could just ask for and receive one.

Yeah, I know, but it's bugged me since they changed it that I have to use [option] to fix what they broke.

[option] is to get UI behaviors that don't work in the expected way.

Well, the expected way was to maximize. So [option] should get you the iPad full screen, because the green button has well-defined, regularly-used expected UI behavior.

Pisses me right off. That and backward vertical scrolling. At least you can fix the latter with a setting, but why should I have to fix my OS?

He'd make a great artillery target.

Feature-itis is the death of usable, useful software, commercially or Open Source.

I don't mean you can't SELL such software (*cough*MS-Office*cough*), but the software sucks.

Apple's success is understanding that one simplifies by removing choice, and this helps _most_ people find the software more usable and useful.

(And no, Apple doesn't walk on water, and yes, they make tons of mistakes and bad choices. No fan-boi here.)

There are a number of constituencies who will _hate_ software simplified this way. One of those is the typical developer, who's an "ultimate customizer" and typically wants all the options available and discrete control over them.

This difference between developers and "most people" is one of the reasons so much software has awful usability: Developers build it for someone like themselves. If there's no one with the professional capacity of evaluating a design's usability, or no corporate will to understand and implement the findings of such an evaluation, the software is gonna suck, for most people.

Yes, there are a few unicorns out there: Developers who know that the typical developer isn't like most people, and can empathize with people who aren't customizers. But they're scarce as...unicorns.

And there are a few use cases out there (IDEs, for example) built for these customizer constituencies, which do well by providing all the options for them. It's often hell getting these expert-focused tools past the usability staff, who, obviously, aren't like developers, and can't see their own blind spot here.

Now maybe I can unload all this Flooz!

They don't need a "right-to-repair" clause. Right to repair needs to be a basic part of the law - a right of any owner of any thing, not just on a one-off contractual basis.

The DMCA and WIPO treaty should be fixed or done away with unless we all can repair what we buy.

As of today, there's no reasonable way to use passkeys on your choice of browser on macOS without enabling iCloud keychain, which I am not going to do.

I can use my Yubikey as 2FA, but not as a passkey without involving Apple's cloud.

Nope.

Another way Passkeys fail.