I had some test functions where minimization could be optionally used, but wanted to do one where minimization was needed, like the Ackermann function. Most of the frontier models struggled with doing this, although I may have been prompting incorrectly. Although - if I had been prompting totally correctly, I probably could have gotten what I got out of a frontier LLM in early 2025 and before.

Incidentally the test function that tells you if a number is prime in Emacs Lisp with primitive recursion is

(defalias 'prime (c (c (c (r 's (c 'z (p 1))) (p 1) 'z) (c (r (p 1) (c 's (p 2))) (c (c (c (r 'z (c (c 's 'z) (p 1))) (p 1) 'z) (c (r (p 1) (c (c (r 'z (p 1)) (p 1) 'z) (p 2))) (p 1) (p 2))) (p 2) (p 1)) (c (c (c (r 'z (c (c 's 'z) (p 1))) (p 1) 'z) (c (r (p 1) (c (c (r 'z (p 1)) (p 1) 'z) (p 2))) (p 2) (p 1))) (p 2) (p 1)))) (c (c (r 'z (c (r (p 1) (c 's (p 2))) (c (c (r 'z (c (r (p 1) (c 's (p 2))) (p 2) (p 3))) (c (c (c (r 's (c 'z (p 1))) (p 1) 'z) (c (r (p 1) (c 's (p 2))) (c (c (c (r 'z (c (c 's 'z) (p 1))) (p 1) 'z) (c (r (p 1) (c (c (r 'z (p 1)) (p 1) 'z) (p 2))) (p 1) (p 2))) (p 2) (p 1)) (c (c (c (r 'z (c (c 's 'z) (p 1))) (p 1) 'z) (c (r (p 1) (c (c (r 'z (p 1)) (p 1) 'z) (p 2))) (p 2) (p 1))) (p 2) (p 1)))) (c (c (r (p 1) (c (c (r 'z (p 1)) (p 1) 'z) (p 2))) (c (r 'z (c (r (p 1) (c 's (p 2))) (p 2) (p 3))) (p 2) (c (r 'z (c (r (p 1) (c 's (p 2))) (p 2) (c (c (r 's (c 'z (p 1))) (p 1) 'z) (c (r 'z (c (r 'z (c (r (p 1) (c 's (p 2))) (p 2) (p 3))) (c 's (p 2)) (c (c (r 's (c 'z (p 1))) (p 1) 'z) (c (c (c (r 's (c 'z (p 1))) (p 1) 'z) (c (r (p 1) (c 's (p 2))) (c (c (c (r 'z (c (c 's 'z) (p 1))) (p 1) 'z) (c (r (p 1) (c (c (r 'z (p 1)) (p 1) 'z) (p 2))) (p 1) (p 2))) (p 2) (p 1)) (c (c (c (r 'z (c (c 's 'z) (p 1))) (p 1) 'z) (c (r (p 1) (c (c (r 'z (p 1)) (p 1) 'z) (p 2))) (p 2) (p 1))) (p 2) (p 1)))) (c 's (p 2)) (p 3))))) (c 's (p 1)) (p 3))))) (p 1) (p 2))) (p 1)) (p 1) (p 2)) (c 'z (p 1))) (c (c (r 'z (c (c 's 'z) (p 1))) (p 1) 'z) (p 1))) (p 3) (c 's (p 1))) (p 2))) (p 1) (p 1)) (p 1)) (c 's (c 's 'z))))

S&P 500 - up 16%

Nvidia - up 25%

ICLN (global clean energy ETF) - up 46%

Micron - up 246%

Seagate - up 270%

Western Digital - up 342%

For the past few days I asked it to built a mu-recursive Ackermann function in Emacs Lisp (built on the primitive-recursive functions/operators, plus an extra operator - minimization). I said that the prime detector function it already built should be able to use the same functions/operators, and to rewrite code if necessary.

So far it has been unable to do this. If I thought it could but was stumbling over Emacs Lisp I might ask it to try in Scheme or Common Lisp or some other language. It's possible I'll get it to work in the time I have allotted from my daily free tier, but I have had no success so far. I am also starting with inputs to the Ackermann function of 0,0 - 0,1 - 1,0 - 1,1 to not overburden the system but it can't even handle 0, 0. Also it tries to redefine the Emacs Lisp keyword "and", which Emacs hiccups on.

A year ago LLMs were stumbling over Leetcode and Project Euler functions I was asking it to make. They seem to have gotten a little better, and I'm impressed Gemini 3 can handle, with help, primitive recursive functions in Emacs Lisp. Doesn't seem to be able to handle mu-recursive functions with minimization yet though. The trivial, toy implementations of these things. Also as I said, it tried to redefine "and" as well, which Emacs Lisp fell over on.

So it's a helpful helper and tool, but definitely not ready to hand things over to. As the saying goes, the first 90% of the code takes 90% of the time, and the last 10% of the code takes the other 90% of the time. Or the other saying - it's harder to find bugs than write code, so if you're coding at peak mental capacity, finding bugs becomes impossible. It does have its uses though, and has been getting better.

After spring 2023, Nvidia stock seems to follow a pattern. It has a run-up prior to earnings, it beats the forecast, with the future forecast replaced with an even more amazing forecast, and then the stock goes down for a bit. It also has runs - it went up in the first half of 2024, as well as from April to now.

Who knows how much longer it can go on, but I remember 1999 and things were crazier then. In some ways things were crazier three years ago with FAANG salaries etc. There is a lot of capital spending, the question is are these LLMs with some tweaking worth the capital spending, and it's too early to tell that fully. Of course a big theoretical breakthrough like the utility of deep learning, or transformers or the like would help, but those only come along every few years (if at all).

The project I worked on was enormously successful in terms of revenue growth. I, and people on my team had a huge nationwide impact, which started when we were about a dozen people (it has grown now to several dozen).

Whereas even a manager of one of the big box stores would only have a limited geographical impact. Whereas my work would always have nation-wide impact (and at some companies programmers would have world-wide impact). I turned on a payment option for my platform one quarter, and very quickly people were using it for one million a month in purchases. Which kept going up.

The book Capitalism without Capital talks about this. Some aspects of it are alluded to in Fred Brooks 1975 book The Mythical Man Month.

To build a car, a lot of effort has to be made in making the car - not just the end result, but the making the glass, tires and so forth. Whereas with programming, I write an app, or a feature for an app, and the end result is duplicated and distributed around the country (or even around the world) for free, or virtually free. I'm not helping make commodities one at a time like someone on an automobile line is. It is something different.

innovator's dilemma

https://qa.debian.org/popcon.php?package=firefox-esr

https://news.ycombinator.com/item?id=4611830

Nvidia, Microsoft, Apple, Alphabet/Google, Amazon, Meta/Facebook, Broadcom, Tesla, Berkshire Hathaway, and Walmart.

A commonality to most of them (and to a lesser extent all of them) is they write software.

If a company not on the list like Ford has an F-150 truck come off the assembly line, some of that $40,000 cost is in the capital expenditure for the plant, any automation it has, the software in the car and so on. But Ford has to pay for the aluminum, steel and glass for each truck. It has to pay for thousands of workers on the assembly line to attach and assemble parts for each truck.

Meanwhile, at Apple a team writes iOS 18, mostly based on iOS 17, and it ships with the devices. Once it is written that's it for what goes off on iPhone 16. There may be some additional tweaks up until iOS 18.6. The relatively small team working on iOS has it going out with tens of millions of units. Their work is not as connected to the process of production as the assembly line people attaching and assembling the F-150 truck. If some inessential feature is not done as a phone is being made, it will be punted to next release. This can't be done with an F-150 truck.

Software properly done is just much more profitable than non-software work. We can see this here. Yes, some of the latest boost is due to AI hype (which may or may not come to fruition in the near future), but these companies got to this position before all of that.

I was watching a speech by Gabe Newell talking about the (smaller) software industry of the 1990s, and the idea back then to outsource and try to save on salary costs. He said he and his partners went the other way and decided to look for the most expensive and best programmers they could find, and Valve has had great success with that. Over the past 2 1/2 years we've seen a lot of outsourcing to cheaper foreign labor, FAANG layoffs (including Microsoft's recent Xbox layoffs), and more recently attempts to lower costs by having software produced by less experienced vibe coders using "AI". I have seen myself at Fortune 100 companies, especially non-tech ones, that the lessons of the late 1960s NATO software engineering conferences, or the lessons learned by Fred Brooks while managing the OS/360 project in the 1960s haven't been learned. Software can be a very, very profitable enterprise, and it is sometimes done right, but companies are still often doing things in the same way they were attempting such projects in the early 1960s. Even attempts to fix things like agile and scrum get twisted around as window dressing to doing things in the old-fashioned corporate way.

People here were pretty skeptical about AlexNet, when it won the ImageNet challenge 13 years ago.

https://news.ycombinator.com/item?id=4611830

The monetary push is very LLM based. One thing being pushed that I am familiar with is LLM assisted programming. LLMs are being pushed to do other things as well. If LLMs don't improve more, or if companies don't see the monetary benefits of using them in the short/medium term, that would drag Nvidia down.

Nvidia has a lot of network effects. Probably only Google has some immunity to that (with its TPUs). I doubt Nvidia will have competition in training LLMs for a while. It is possible a competitor could start taking market share on the low end for inference, but even that would take a while. People have been talking about AMD competition for over two years, and I haven't seen anything that even seems like it might have potential yet, especially on the high end.

A year ago both its trailing and forward P/E were higher. So the stock is relatively a bargain compared to what it was a year ago.

The price implies that revenues and profits are expected to continue to grow.

> My intuition is that the absence of the rapid, generationally transformative, advances in tech and industry that were largely seen in the latter half of the 20th-century (quickly followed with smartphones and social networking), stock market investors seem content to force similar patterns onto any marginally plausible narrative that can provide the same aesthetics of growth

I wouldn't disagree with this.

Also, in some sense there can be some fear 5090s could cannibalize the data center hardware in some aspects - my desktop has a 3060 and I have trained locally, run LLMs locally etc. It doesn't make business sense at this time for Nvidia to meet consumer demand.

One is - Google, Facebook, OpenAI, Anthropic, Deepseek etc. have put a lot of capital expenditure into train frontier large language models, and are continuing to do so. There is a current bet that growing the size of LLMs, with more or maybe even synthetic data, with some minor breakthroughs (nothing as big as the Alexnet deep learning breakthrough, or transformers), will have a payoff for at least the leading frontier model. Similar to Moore's law for ICs, the bet is that more data and more parameters will yield a more powerful LLM - without that much more innovation needed. So the question for this is whether the capital expenditure for this bet will pay off.

Then there's the question of how useful current LLMs are, whether we expect to see breakthroughs at the level of Alexnet or transformers in the coming decades, whether non-LLM neural networks will become useful - text-to-image, image-to-text, text-to-video, video-to-text, image-to-video, text-to-audio and so on.

So there's the business side question, of whether the bet that spending a lot of capital expenditure training a frontier model will be worth it for the winner in the next few years - with the method being an increase in data, perhaps synthetic data, and increasing the parameter numbers - without much major innovation expected. Then there's every other question around this. All questions may seem important but the first one is what seems important to business, and is connected to a lot of the capital spending being done on all of this.