Univalence says that equality is equivalent to equivalence, ie, formalizing the notion of when we can use equivalence rather than equality as a step in a proof. In practice, we often only care about proofs “up to equivalence”.

A way to think about this:

- equality is an identity map

- equivalence is an isomorphism

For example, 2 in Z and 2 in R do not have an identity map between them — but do have an isomorphism.

I think the key insight of univalence is not collapsing equivalence into equality — but allowing it to remain a second truth relation.

We don’t want 2 in Z to be equal to 2 in R (because we collapse type distinction), but we do want them to be equivalent — so we can do equivalent reasoning about arithmetic in R to reach conclusions about Z.

However, your last question is incorrect: people are working on that, but there haven’t been hugely useful results.

But as an example, I’ve been slowly working on implementing frameworks for theory distillation — eg, take a corpus of science papers and derive a consistent model of the world from them, such as in Lean. (Or more specifically, a sheaf defining what consistent theories are possible.)

You can see the collapse of black marriages started earlier and is steeper than other demographics — though eventually white marriages enter a similar decline, along with Hispanic marriages. (Asians hold roughly equal.)

This strongly suggests the effect is correlated with poverty and not race — especially with the white effect starting during the collapse of US manufacturing.

But if we agree the correlation is with poverty you haven't refuted my point — as welfare is a strong incentive structure on those experiencing poverty. That chart is perfectly consistent with what I said.

So your criticism is wrong at two levels:

- you didn’t even discuss the correct time period

- you showed data consistent with my claim

That it destroyed black families (among others) is merely happenstance rather than intentional racism — stemming from the overall failure of technocratic managerialism.

https://www.youtube.com/shorts/UPxpKe4N8b8

The “Welfare State” is nothing but modern share cropping from the party of Jim Crow.

As such, LLMs can only bullshit.

The amount of text isn’t equal and the benefit is primarily that I can push the LLM into a basin you don’t know the keywords for — something that takes a few minutes to type a couple sentences and me scan the results to make sure they’re roughly on track. And much, much less time than me trying to (re)write a treatise on a topic.

Dropping you an LLM reply is the equivalent of dropping you a link and telling you to learn about the topic — and appropriate in roughly the same circumstances.

I’m only out the cost of the drive, which is like $40 and doesn’t require anybody on the other side cooperate with me.

- - -

More broadly…

You call it unlikely mixes, but we see it all the time:

- people already have a computer for gaming or work

- people (ie, “preppers” like we’re discussing) buy a generator for emergencies

- local emergency response sets up mesh networking during disasters, both official and unofficial

Have you ever tried to use a handbook you’re not intimately familiar with during an emergency? It’s rough.

For personal preparedness, nothing replaces familiarity and practice — eg, weekend survival trips and reading your manual ahead of time.

But for providing information in a random lookup manner to unpracticed people who weren’t prepared? Yes, I think an LLM/chatbot is the practical way to operationalize all that information which you stored (eg, survival guides or machine manuals).

Also, it’s unlikely a general purpose chatbot would be superior at survival advice to one specialized for that purpose — and indeed, is likely to refuse your questions as “unsafe” or “criminal”.

That way your machine that, eg, normally plays video games or does AI work can support relief efforts by supporting emergency response IT. You don’t need to mothball the machine, just have an “emergency” boot USB than can run the services from your home generator.

You don’t even need to bring it with you: turn it on and leave it “best effort” at home, while you continue to use it via WAN.

My point is that if you feel that way, be loud and proud about it: having to deny the humanity of embryos suggests you’re avoiding cognitive dissonance or trying to manipulate others.

They are absolutely a human.

To sibling comment’s point: personhood versus humanity is already the legal distinction - eg, children and invalids are fully human but not fully people, so we don’t grant them full legal rights and instead let others exercise control over them. For children in particular, we grant them more autonomous control of their rights as they mature.

Your path of denying humans their humanity has a dark history, including within medicine.

I don’t support that, even if they’re very tiny and fragile humans early in their development — eg, if we could raise them in artificial wombs, they may have a right to life as a human, despite what anyone else wishes.

Denying their humanity is denying their right to life as a fellow human.

They’re much cheaper to run, eg, Llama 3.3 Instruct 70B is 5-10x cheaper than Sonnet 5.

https://aws.amazon.com/bedrock/pricing/

Say you have 20% of usecases that require the more expensive model — but in 80% you could just use Llama instead of Sonnet (eg, for basic queries of a document). That saves 80% of that 80%, or 65% of your total bill!

That is the kind of “swap” that’s likely to occur in automated tooling as pricing pressure kicks in — “can you save 65% on our AI bill by switching Bedrock over in 80% of uses?”

A quick search suggests that we tried it a bunch on animals, failed on humans, and of the first three human transplants, you had an adult (20s), toddler (2yr), and adult (50s). All of them were due to medical necessity, as it was a high risk procedure.

Is there something you imagine the poster would be shocked by in that history?

Human is a fact of DNA — of which embryos are unmistakably humans. But we don’t ascribe rights based on humanity, but on personhood, which is why embryos have virtually none.

You come across as dishonest and put people off from your perspective when you won’t admit basic facts — eg, this is experimenting on humans.

I think perhaps your aversion to basic facts is that you don’t feel comfortable being honest:

You support experimenting on living humans.

We’ve know that since 1943 when McCulloch-Pitts came up with the first “artificial neuron” definition. And since LLMs are a descendant technology — our assumption should be they’re reasoning in some internal learned logic.

This is what the evidence supports — eg, the “stochastic parrot” crowd never can explain transfer learning. Whereas for the internal reasoning crowd that is easy: removing your top level judgments from a theory still leaves you with useful terms for describing a new theory — eg, removing your judgments about “which animal is this?” but preserving the underlying structure for representing an image in your new judgments, “is this cancer?”

There’s 80 years of reason to think DNNs reason and zero support other than “sTaTs R mAgIc!” to support the stochastic parrot interpretation.

Ignorance isn’t argument.

For sufficiently large values of 2 and small values of 5, the statement holds.

So if you’re a bakery and a customer offers $9000+ for a fresh roll every day forever, you should almost certainly take them up on that offer. A smart baker could probably get that number sub-$5000, but you’ll always come out okay around 25x the yearly cost (in this case, $365).

Similarly, if the amortized yearly cost of a customer is $12 (ie, $1 per month), then a $300 forever price is financially indistinguishable from a permanent subscription. (Actually, better: time value of money, they can’t cancel the annuity you buy, etc.)

So there always is a price where that is financially viable.

I’m missing the irony.