Sunday I was in Austin and saw a Waymo being put on a tow truck at an intersection, and a cop car with flashing lights behind it. I doubt this was a drill, and I had gone through the intersection an hour before, so I guess it actually broke down there. Welcome to the future!

IMHO, Perl's biggest problem was that Larry didn't know how to say "no" to every crazy idea he saw.

My pet misfeature is "long statement that does something UNLESS condition". Yes, let's reverse the flow of cause and effect of the traditional "if" statement, what could possibly go wrong when someone tries to read your code?

Also protip: use HTML <p> to insert paragraph breaks on /.

As crap as AI is shaping up to be, it still seems to be not much worse than a call center in India. This adds a second front in that war.

One time back in the '10s I called Apple for something at about 5PM Austin time. It got answered by someone in Ireland. My mind was blown, both because I think Apple had recently opened a call center in Austin, and because that meant it was almost midnight at the call center.

Anyhow, it's about time they figured out a way to "tariff" off-shored non-manufacturing jobs.

If you want tap water to taste better, get a reverse osmosis filter installed under your kitchen sink. That filters out stuff using the water mains pressure, and ejects water into the drain with the crap that was filtered out (meaning it's now got twice as much crap in it). Then it keeps a gallon or so in a low-pressure storage tank for dispensing later. It has no taste, same as distilled water, and is still drinkable even when 35C warm. Try that with regular tap water.

Also it filters out all the biological stuff, so if you put it into an old half-gallon juice bottle, you don't get nasty mildew stuff growing under the lid.

Which ones? Every one is different. I absolutely hate saccharine, but I love aspartame (which has a before taste, not an after taste), and it's been in common use for three decades now. But I don't like the sweeteners used in "zero" drinks as much (which is aspartame plus other stuff). There were cyclamates which were good, but got unfairly banned in the 1970s and still are. Also, "sugar" is uncommon in the US because it has been tarriffed to hell for decades to prop up both the sugar and corn lobbies in favor of HFCS, which is not good for you.

Treating them as all the same is quite unscientific, you have to know which ones are which instead of just turning your brain off and listening to media narratives intended to scare you for ratings. Even simply assuming that you're getting sucrose sugar in non-diet drinks is wrong.

My parents would drink 1980's Diet Pepsi, which used saccharine. It was absolutely completely awful. Too bad I was just a little too late to experience cyclamates, which were unfairly banned, and still are. But I happen to love the taste of aspartame, so that's why I drink diet sodas, especially when the alternative is HFCS. And it's just two amino acids, so not likely to ever be a real problem, aside from the few people who already had trouble with one of those amino acids.

University of Southampton? Meanwhile, I've seen many pics of UK takeaway pizza boxes full of fried everything, something I've never seen in the US.

As for the temps, it's not global warming that's why every summer here is 35C+, it's because much of Texas is at the same latitude as Cairo.

Yeah, like that.

There is a big difference between a commented disassembly and the original source code with original comments and labels.

At least that one edited the disassembly to show data bytes. The equivalent books for the Model I/III used a brainless disassembler that mowed right through data.

This is exactly the same as the pagetable.com version of ten years ago. The main difference is an official open source license for it.

I certainly hope they still have the source code to newer versions and will release them. There were improved versions released for the 6502 as Microsoft Basic II for the Atari, and the Epson HX-20 BASIC for 6301. And then there were the 68000 versions that are still completely un-leaked.

The 6502 was basically all new code, then it was ported to the 6800, 6800 variants, and the 6809. At least I think the 6502 was first. The 68000 versions probably evolved from that code base.

The x86 version was a port of the 8080 version, and those who know their x86 code say that it clearly was ported by one of those cross assembler translation things that made less than optimal code for memory moves, and it took a few years before someone cleaned those up.

But they did do something cool for the original 8080 version, just more primitive than you were thinking. They wrote an "emulator" for PDP-10 that was done entirely with macros. It would generate PDP-10 code from 8080 mnemonics so that they could test it without having an actual Altair present. This 6502 version also used a hacked-up PDP-10 assembler, as can be seen by its use of the same macro syntax, but I would be surprised if they had the ability to run its code on PDP-10.

OSI was one of the original targets of this. I spent some time looking at the OS65U version just last week. I could definitely see it being based on this code. They kept to a single binary version and monkey-patched it like crazy to add stuff. Or at least they did after they found a bug in the string garbage collection, when they had already ordered thousands of mask ROMs with the buggy code. The keyword table was 255 bytes long because they didn't want to rewrite the code in CRUNCH and LIST, which used absolute,X addressing mode. They had a keyword FLAG which was like a list of custom POKEs.

I had a special interest in this because the first computer job I had in my senior year of high school back in 1982 had a full-rack Challenger III with hard drive and RAM boards for two users. Yes, two users on a single 6502. I was surprised that I could remember its default file passwords.

Actually Bill Gates didn't work on the 6502 versions. He was only involved in the x80 releases. The last version that he did any major work on was supposedly the Tandy 100 portable.

But I started on a TRS-80 Model I Level II, and within a year or two had already begun disassembling it. So I really did learn assembly language from Bill Gates.

There is in fact a 6809 version out there, but it was from the Dragon 64. Which is almost exactly the same as CoCo, and with a math package not much more advanced than this current 6502 release. The code is mostly a direct port between 6502 and 68xx, so with a good disassembler that lets you assign labels it's not hard to match up to other binary versions.

And it's still from the era with only one numeric type. There was a relatively modern (roughly equivalent to CP/M) 6502 version for Atari ("Microsoft Basic II"), but no non-Intel source from that era has escaped yet. The Epson HX-20 (runs on a 6800 variant) is another of these more modern versions, and both of them have blown me away with how much new stuff they have.

What I really want to see is source code to one of their 68000 versions. That's a big hole in what is out there. The old classic MacOS version would be really nice to see, especially since it had both a binary and a BCD math version. The same code base was also used to make an Amiga version.

There was also a 68000 version for the Tandy 6000 that was equivalent to the CP/M versions. But I have only seen a binary math version of it. I think it would be fun to port MS BASIC to Sega Genesis, which is why my special interest in a 68000 version.