¹: Often correlating with a marked decline in the quality of their technical writing. I don't know which direction the arrow of causality goes, but I have observed LLM use inducing madness under reasonably-controlled circumstances – different phenomenon, similar principle. I suspect the use of AI codegen systems is causing the reduction in discernment ability, rather than a sudden drop in discernment ability causing increased use of AI codegen systems.

> “In January 1956, we assembled my wife and three children together with some graduate students. To each member of the group, we gave one of the cards, so that each one became, in effect, a component of the computer program”

> In the summer of 1956, John McCarthy, Marvin Minsky, Claude Shannon and Nathan Rochester organized a conference on the subject of what they called "artificial intelligence" (a term coined by McCarthy for the occasion). Newell and Simon proudly presented the group with the Logic Theorist. It was met with a lukewarm reception.

> “They didn't want to hear from us, and we sure didn't want to hear from them: we had something to show them! [...] In a way it was ironic because we already had done the first example of what they were after; and second, they didn't pay much attention to it.”

> Logic Theorist soon proved 38 of the first 52 theorems in chapter 2 of the Principia Mathematica. The proof of theorem 2.85 was actually more elegant than the proof produced laboriously by hand by Russell and Whitehead. Simon was able to show the new proof to Russell himself who "responded with delight".

---

Tools that could cheat:

> Around 1983, Eurisko, an early attempt at evolving general heuristics, unexpectedly assigned the highest possible fitness level to a parasitic mutated heuristic, H59, whose only activity was to artificially maximize its own fitness level by taking unearned partial credit for the accomplishments of other heuristics.

     Mel finally gave in and wrote the code,
     but he got the test backwards,
     and, when the sense switch was turned on,
     the program would cheat, winning every time.

Tools that could communicate:

> A significant advance in modems was the Hayes Smartmodem, introduced in 1981. The Smartmodem was an otherwise standard 103A 300 bit/s direct-connect modem, but it introduced a command language which allowed the computer to make control requests, such as commands to dial or answer calls, over the same RS-232 interface used for the data connection. In data mode, all data forwarded from the computer was modulated and sent over the connected telephone line as it was with any other modem. In command mode, data forwarded from the computer would be interpreted as commands. In this way, the modem could be instructed by the computer to perform various operations, such as hang up the phone or dial a number. The modem would normally start up in command mode. The command set used by this device became a de facto standard, the Hayes command set, which was integrated into devices from many other manufacturers.

> The experimental challenge consists in showing that a distributed set of agents can develop from scratch a vocabulary to identify each other through names and spatial descriptions. […] When there is already a sufficiently shared language, the first part (initiation) may be absent. In that case only linguistic means are used to identify the object. […] When more agents use the same word for the same meaning, communicative success increases and therefore the word-meaning association becomes more stable. It has been shown in an earlier paper that coherence emerges (see Figure 1) [8]. In the following subsections the mechanism is defined more formally and a concrete example of language formation is given. […] A mechanism has been presented in which a group of distributed agents develops a vocabulary to name themselves and to identify each other using spatial relations. It was shown that a vocabulary indeed emerges in a group of agents through a series of conversations. The mechanism also copes with the entry of new agents or new meanings.

> With its origin in the Georgetown machine translation effort, SYSTRAN was one of the few machine translation systems to survive the major decrease of funding after the ALPAC Report of the mid-1960s. The company was established to work on translation of Russian to English text for the United States Air Force during the Cold War. The quality of the translations, although only approximate, was usually adequate for understanding content. By 1998, "for as little as $29.95" one could "buy a program for translating in one direction between English and a major European language of your choice" to run on a PC.

and, of course, every interactive program ever made.

---

Comment adapted from: https://en.wikipedia.org/w/index.php?title=Logic_Theorist&ol... https://en.wikipedia.org/w/index.php?title=Reward_hacking&ol... https://users.cs.utah.edu/~elb/folklore/mel.html https://en.wikipedia.org/w/index.php?title=Modem&oldid=13732... https://en.wikipedia.org/w/index.php?title=Hayes_Microcomput... https://doi.org/10.1162/artl.1995.2.3.319 https://en.wikipedia.org/w/index.php?title=SYSTRAN&oldid=136... https://en.wikipedia.org/w/index.php?title=Machine_translati...

> The word “briskly” here means “quickly, so as to get the Baudelaire children to leave the house.”

> “Wipi!” Sunny shrieked, which meant “I’d much prefer gardening to sitting around watching my siblings struggle through law books.”

But they're not going to backdoor an Apple ][e, or a random 80m¢ microcontroller, for basically any value of "they"; so you can just use one of those instead, and save yourself the hassle.

Just write a new OS. It's a weekend project to get enough groundwork that you can bootstrap a clean system from clean source code.

> And wouldn't there be difficulty comparing binaries built from significantly different environments?

Not really. Starting from stage 0, compile the compiler under test (stage 1), then use the compiled compiler to compile the compiler (stage 2), and compare the stage 2 artefacts. Provided that your comparison program is known-good, and the stage 2 build is deterministic (not the case for some real-world programs, but true for things like tcc), this lets you verify that the two compilation procedures work identically.

The authors I'm aware of making >$20k don't do any significant gating. That practice is only really common with people republishing through Amazon (which probably demands it).

Then maybe we should do away with "business". Small operations do not tend to operate that way, and broadly everyone prefers their output (if not price) in the domains in which they operate – with a few notable exceptions. In theory, software should allow small operations to each serve millions of people, such that we (e.g.) only need a few thousand search engine providers to meet the needs of the entire world's population. Not everything has to be enterprise-scale, and indeed many things should not be.

If "business" means treating people badly, then we don't need it.