Isamu

Same with essay assignments, you exercise different neural pathways by doing it yourself.

Recently in comments people were claiming that working with LLMs has sharpened their ability to organize thoughts, and that could be a real effect that would be interesting to study. It could be that watching an LLM organize a topic could provide a useful example of how to approach organizing your own thoughts.

But until you do it unassisted you haven’t learned how to do it.

I am very interested in the Community Movement, Wikipedia is lacking references https://en.wikipedia.org/wiki/Community_Movement

And also the Waldensians https://en.wikipedia.org/wiki/Waldensians

Yeah I remember my class was the last in our high school to learn the slide rule, the next year we transitioned to (expensive!) calculators.

People had to be taught not to go wild with the extra precision.

No problem really, I think most people want to bootstrap quickly from the low level into the higher abstractions that they care about, few people want to stay down in the cpu itself.

I think this is a good overview for most people, this is probably what they want.

For me personally I was surprised given the name that very little is about cpus and most of the material is in the operating system.

>The source material was a compact cassette tape from a demo, found in a drawer.

Was this demo his, or someone else’s IP? If he is cleaning up or modifying his own property, not a lot of people have a problem with that.

If it is someone else’s work, then modifying with AI doesn’t change that.

I think they just don’t want AI generated works that only mash up the work of other artists, which is the default of AI generated stuff.

People invoke Marcus Aurelius but are not really engaged in the classical study of Greek stoicism, they read into it what they want to see. It’s a lazy justification for what you already want. Any “epiphany” that comes from that is self serving, cloaked in moralistic terms.

Anger is an energy

>LLMs do not have opinions.

Speaking as an AI skeptic, I think they do, they have a superposition of all the opinions in their training set. They generate a mashup of those opinions that may or may not be coherent. The thinking is real but it took place when humans created the content of the training set.

That’s interesting, they use the Singularity to refer to technological explosion in general rather than specifically AI becoming super intelligent. This seems to predate the AI usage.

>"Singularity is a time in the future. It'll occur when the rate of change of technology is very great--so great that the effort to keep up with the change will overwhelm us. People will face a whole new set of problems that we can't even imagine." A look of great tranquility smoothed the ridges around his eyes. "On the other hand, all our normal, day to day problems fade away. For example, you'll be immortal."

The Bremermann limit is about computational density. It makes sense to talk about computers that are ever more powerful but not necessarily larger.

So we talk about the supercomputers we call cell phones that are orders of magnitude more powerful than the desktop computers I used years ago.

It doesn’t make sense to talk about making super large computers before reaching the density limit, that’s a confusion of concepts.

Making computers faster has involved making them smaller because the speed of signal propagation.

What’s up with the ending that makes no sense?

>Where are the physical limits? According to Bremermann (1982), a computer of 1 kg of mass and 1 liter of volume can execute at most 1051 operations per second on at most 1032 bits. The trend above will hit the Bremermann limit roughly 25 decades after Z3, circa 2200. However, since there are only 2 x 1030 kg of mass in the solar system, the trend is bound to break within a few centuries, since the speed of light will greatly limit the acquisition of additional mass

They shift from talking about the transistor density to somehow considering a supermassive construct. Reminds me of LLM mashups.

So you reach back to 2022 for the problems cited in 2025? I guess nothing else happened this year that could affect the economy. Nope, nothing.

Yeah at work I’m paid to own some components that I didn’t write and don’t entirely understand, so I figure my job is to help discover answers for the questions that arise.

I would not want to be a public maintainer though. I don’t have the patience or motivation to use my spare time for that.

Isn’t the implication that cipher is a loanword? So language relatedness is irrelevant?

We use “arabic” numerals around the world. So use of an Arabic loan word is unsurprising.

The criticism is of the spin in these articles. The experience these kids get is great, it should happen more. The articles always spin to get your attention, and the subject matter is fascinating, but it can be presented with less spin.

And frankly any kid deserves praise for doing the unglamorous work that this takes. Very few can be arsed to put up with the extra work that it takes to do anything worthwhile, we are a nation getting lazier every day.

Making things, I made my daughters wooden practice swords, two-handed claymores. Now for Christmas I always make something. This year was cross-body bags with elaborate patterns. Fun for me, I’m just glad they put up with the eccentric choices.

A great article that starts by not discounting the written accounts that are available.

I am reminded that the Maya language decipherment really moved forward once the written account by Diego De Landa was taken seriously.

Yeah I’m dreading a change in Apple TV in this direction. I avoid the crappiness of smart tvs by not connecting them to wifi and just using Apple TV, but how long before that becomes crappy too?

… because a capitalist would never scam a robot for profit, never! Too scrupulous.

Plus capitalists don’t know the first thing about what communism is and hence couldn’t pull off the scam, they only know that “communism bad”.

Tldr: reconstruction of statue painting is based on residual pigments alone, and tend to look garish because the reconstruction is just an expression of the available data and fully saturated color.

But where we see wall frescoes and the like they are painted with what we would call artistic taste and not like the garish reconstructions.

Westinghouse was acquired as a brand under Tsinghua TongFang.

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

So WebKit is the walled garden? I thought it was the App Store.

My local museum had a room of poorly labeled and unlabeled objects that I loved and so I went down the rabbit hole of identifying and investigating most of them. Then I wrote the missing Wikipedia page for my favorite. If they had provided good labels I would never have gotten so knowledgeable about them.

>Numbers are just a shortcut for dealing with repetitions

An interesting explanation, I think I agree

Someone commented here that hallucination is what LLMs do, it’s the designed mode of selecting statistically relevant model data that was built on the training set and then mashing it up for an output. The outcome is something that statistically resembles a real citation.

Creating a real citation is totally doable by a machine though, it is just selecting relevant text, looking up the title, authors, pages etc and putting that in canonical form. It’s just that LLMs are not currently doing the work we ask for, but instead something similar in form that may be good enough.

I believe this one is a deliberate construction, they knew the evolution of the pieces and gradually put it together.

There’s search programs too, for smaller patterns. This construction is just too big and with such a long period. The search space would be enormous.

I got involved in this stuff years ago when I modified a search program for Life to search any CA rule. That’s how we found the HighLife rule and others like Day and Night.

Yeah me too I was stoked for a moment. But I guess I’m in maker mode right now.

Another way: morality is always in the context of your wants versus society. Society is an objective fact here, as is yourself.

That you want to live your life is a given. In order to avoid bad backlash from society you might objectively choose to adopt some moral behaviors.

Discourse about morality is unnecessarily complicated by historical things that we don’t want to shine a light on, like slavery and the treatment of women. We can’t explain why we have been talking about morality for millennia but haven’t come to the conclusion that our treatment of others matters.

Objection: first observe that much of morality is about “doing the right thing”, where the right thing is hard because it is opposed to selfishness.

Selfishness is the default and the natural impulse. You rarely have to convince people to take the selfish option, unless they are long indoctrinated to be selfless.

The objective reason to act morally is to live in a society where you have a chance to receive civility in return. Common morality, for example “don’t murder”, is about avoiding a continuous escalation of bad actions that are likely to come back to you.

It’s about having a functioning society, rather than a prison where you are constantly looking over your shoulder to avoid being shived by the brother of the guy you murdered.