Notices by David Nash (dpnash@c.im)

@inthehands I can’t comment directly on 1955, but can on 30 years later, in 1985-ish.

Not a chance of an iPhone (or, at least, a rough 1985 equivalent). At best, I think you *might* get some fairly nice software on a single sort-of widely used platform. But only on one, at a time when there were far more platforms, with numerous differences that mattered a lot, compared to now.

Just thinking of my own early programming experiences: There were something like 6-8 *actively used* flavors of BASIC (still being used occasionally in the early-mid 80s for commercial software, amazing as it might seem now), each one having very different ways to do mundane things like clear the screen or do pixel-by pixel graphics. Porting a graphics-heavy program from, say, Apple II or Atari BASIC to IBM PC was obnoxious at best.

Pascal was more consistent across systems, but I remember some fairly significant differences between the Apple (IIe/old Mac) versions I learned as a kid vs. the VAX/VMS version I saw in intro CS in college.

C was also more consistent, at least on Unix boxes, but there was still an awful lot of shoot-from-the-hip coding there. The version of K&R I had in the mid-1990s (so, 10 years later or so) still had notorious buffer overflow sources like gets() in its sample code, and this wouldn’t change much till the internet and widespread publicly accessible networking raised the danger level on those a lot. An actually good AI *now* would be aware of the scope of possible problems there, but in 1985, I’m much less confident it would have been.

I don’t know nearly as much about the big corporate/research systems that mostly ran FORTRAN or COBOL, and I *suppose* that those environments might have been more consistent and thus a bit better for an AI project like that, but I have my doubts.

@feld @nyrath The sudden shift from what I was discussing (gen AI and LLMs, which are *notoriously* bad at this sort of thing) to generic “computers” for control (which can be, and often are, perfectly fine for it) is quite remarkable indeed.

@nyrath > Can AI Help Manage Nuclear Reactors?

Definitely a case of Betteridge's law of headlines applying, in full force.

I once asked ChatGPT how to change the engine oil in a 2017 Chevrolet Bolt. It gave me tolerably detailed steps for the process to change the engine oil. Unfortunately, this is an electric car, so it has no engine oil to change in the first place. The equivalent level of ignorance in advice to nuclear plant operators is too terrifying to contemplate.

@lanodan @Quenby The more I think about it, the more likely it seems that it was just a plain old TAB keystroke after setting a single tab stop at exact center, rather than a dedicated key. (These were the last generation of boringly mass-produced office electric typewriters, before the short-lived partly computerized ones, and they tended to have the bare minimum to get the job done.) Still, that bit of memory got de-indexed long ago and I don’t really know for sure.

@lanodan @Quenby The first (loud THUNK) sound was hitting either a dedicated “center” key or a tab stop set to the center. (It was over 40 years ago; I’m genuinely uncertain of the exact mechanism, but whatever it was, it THUNKed the typewriter carriage to exact center.) Then, backspacing once per two characters in the title offset the carriage to the correct position to begin typing the title and have it centered (+/- one character, at most).

@Quenby

Centering “This is my title”:

(Loud *shtunk* sound)
(Think “T,h”)
(backspace)
(Think “i,s”)
(backspace)
(Think “Space,i”)
(backspace) …

Yeah, I’m oldish, at least. But could be a lot older. This was in the early 1980s, probably one of the last 4 or 5 typing classes at my middle school that used actual typewriters.

@michael_w_busch @knud @mekkaokereke @lerxst @OldFartPhil @grumpasaurus Couple anecdotes from my life as an aspiring young scientist (late 80s/early 90s, major and Ph. D. in chemistry):

1. The undergraduate school I went to (small STEM-oriented college) had a foreign language requirement for admission. It had no requirements for specific language but said "preferably German".

2. UC Berkeley, one of the candidates I had for grad school in the early 1990s, either required or strongly recommended that its chemistry graduate students take at least introductory German.

In both cases, this was not because current research needed an understanding of the German language. Of course, by the 1990s English had been the dominant language in chemistry research for decades. It's because an absolutely enormous amount of historically relevant chemistry research, particularly in organic chemistry, was done by German researchers in the late 1800s and early 1900s. The Beilstein database, one of the premier collections of data about publications in organic chemistry, began in the late 1800s and did not publish information in English until the 1960s.

https://en.wikipedia.org/wiki/Beilstein_database

@selzero My favorite example of this is still the La Brea tar pits in Los Angeles.

"la brea" = "the tar" in Spanish, so we have the the tar tar pits in LA.

I recently decided to leave #Facebook, and posted a note to everyone to let me know contact info if they wanted to stay in touch.

So far, after just over a day, from people who I didn’t previously have any contact info:

- 3 emails w/updated address + cell/SMS number
- 10 new followers on Bluesky
- 1 new Mastodon/other Fed follower + 2 other people shared Fedi addresses

Not bad for just one day. I don’t expect everyone to give me their info, but it also confirms that the important people will still want to stay in touch.

@fedithom @Infoseepage @hacks4pancakes It's still *possible* to be someone born in 1988 and not be aware of the choice of user name when using (part of) their birth year to do the usual quick-and-dirty username disambiguation.

But a CEO of a company with a popular product used for a long time largely by people who really, really don't want to be associated in any way with Nazi shit should recognize this as a Bad Choice (TM). And if they don't...well, that says something right there.

(Obfuscating 88 with binary is particularly sus. It requires you to actively *think* "I want to put '88' in my user name and *not* just because using my birth year is convenient -- because otherwise I'd just use the decimal version.")

@airshipper @frawst Ad tech. Lots of ad tech. It’s not just “hey, Internet Rando, give us some cash and we’ll throw your ad onto the search results somewhere.” That’s chump change they’re not interested in. Instead a lot of it is a huge underlying bidding/auction setup where advertisers constantly bid (often enormous amounts) for favorable placement. https://support.google.com/google-ads/answer/2472742

Last I saw, about 75% of their revenue was ads, with most of the rest Google Cloud services.

@foone I am currently paid to touch other languages.

They also have their faults, but at least debugging by typing “rm -rf node_modules” (while marveling at just how freaking many GB of disk space this frees up) is not one of them.

@inthehands @nikclayton I'd actually go so far as to say that Reveal Codes back in the 80s made HTML almost instantly comprehensible, even with emerging complexities like (then-larval) style sheets, in the 90s. And *that* in turn made it possible for me to make the jump to web development as a career when a previous one started to head south -- more so in many ways than having a reasonably decent background in "what the annoying pedants call programming" programming as well.

@inthehands @nikclayton This all reminds me of one of the first actual "debugging" tools I ever used (this was before I had done enough "what-people-who-think-HTML-isn't-programming" programming to use or need a traditional debugger for it).

"Reveal Codes."

This was the magic WordPerfect command from the 80s/90s that showed you why, for example, the fricking italics weren't working or that one section on that one page was just a little bit off. "Reveal Codes" popped up a little text area (not even a window, this was DOS, after all) showing a section of your text and all the associated "codes", which were tags indicating the structure or appearance of whatever they contained.

A *lot* like HTML.

And using Reveal Codes to figure out where your document was going wrong, and why, and how to fix it, was a lot like wrangling HTML to get your web page to have the right structure and appearance...and, as noted, also like trying to figure out how to get a cranky API to get the kind of output that you need it to deliver.

@inthehands This reminds me of the (fortunately) brief time during the cryptocurrency bubble when a few techbros thought that “code as law” would be a workable concept. Even without the instances of “code as law” losing badly to “existing law as law”, there were way too many cases of the code doing exactly what code will do, and screwing its own designers over in ways that were completely predictable to non-techbros.

@inthehands @exchgr The Republican Party, now desperate for someone to run at all, drafts their nominee from 2012, Mitt Romney, plus some luckless Senator or governor as VP. The non-Democratic vote is gloriously split, and Harris wins all the states won by either Obama or Biden at some point, with the possible exception of Indiana. A few other states fall in her column that wouldn’t have otherwise. Romney wins Utah and maybe Idaho and one or two other Western states. Trump, whose loss is beyond the wildest ability of corrupt election officials to undo, disappears into a black hole of self-pity before any of his remaining trials.

@FibroJedi @inthehands This precise set of shenanigans is the main reason I left Facebook almost entirely about 5 years ago. I still have some family members who are there all the time, so I drop by occasionally to look at some of what they're up to, but it's been a *long* time since it was even sort of useful for its ostensible purpose of staying connected to any significant number of important people in my life.

Apparently the change from "mostly reverse chronological, plus ads and similar junk" to "engagement, engagement, engagement all the time!" meant that "I *just* saw this but didn't engage with it right away" was considered a negative factor for "I may, just may, want to see it again in the future".

@JessTheUnstill @inthehands After seeing this dynamic play out multiple times with words or expressions that seem neutral to positive to me*, my increasingly common addition to "fsck racism and white supremacy" is "in addition to all the other awful shit they do, they eat otherwise useful words and phrases and turn them into racist garbage."

* one of the more recent ones, and a clear example: discontinuing the former standard default name for git repository branches. That was an easy change for me to make, but I've run into too many white folks who have trouble understanding why it's important.

@inthehands That workflow only works with pastry tubes that have the Mibbler interface for stapler/JavaScript integration. Note that Mibbler version 14.67.blobcat or higher is recommended for best performance. If you haven't mibbled your pastry tubes yet, the next best bet is a NoSQL plugin that routes the JSON into the cloud with the LightningWeasel protocol. With that particular setup, do *NOT* plant the high bits in any type of soil. They need to be artisanally curated in hydroponic pods for 6 weeks first.

After a period of relatively restrained handling of "AI" topics, my division at work decided that all the developers, designers, engineers, whatever, ... need to "use AI more in our everyday work". (Oh, joy.) This included a series of workshops designed to introduce everybody to some representative examples.

One workshop involved Github Copilot, and the following things happened to one development team, all senior developers:
- Copilot generated a unit test case that was hard to get to pass.
- When asked to generate empty test cases, Copilot generated the same (irrelevant) code over and over again.
- Copilot stopped giving suggestions to one developer after a while.
- Getting useful information out of Copilot frequently required a lot of fussy or non-obvious prompt editing and tweaking.

I won't supply direct quotes without the explicit consent of the people involved, but there was a very clear general sense that Copilot was not fit for purpose -- even when it did produce something not totally wrong, it was not a useful timesaver for the types of work this team was doing.

It wasn't just Copilot that seemed half baked. The workshop's guidelines (which are themselves part of a fairly polished Github repo) were poorly proofread. One example had a prominent typo in some HTML you were supposed to generate: '<button class=""btn" ...>' (note the extra double-quote). A newbie to web development would very likely add the spurious double quote mark to otherwise ok Copilot output to make sure it matched the instructions.

Finally, our IT department disallows results from Copilot that come from training on "public" code, for what should be fairly obvious legal concerns regarding copyright and similar issues. For one developer, Copilot repeatedly started to generate a result but then stopped, with an alert that the result appears to match known "public" code.

If it wasn't clear before that Copilot's basic mode (no "private code" option) is a copyright-laundering and license-laundering tool, it's really obvious now.