Hacker Newsnew | past | comments | ask | show | jobs | submit | ted_dunning's commentslogin

You're right.

It's the physics of cooling the beasts and the communication delays that make those plans ludicrous.

To turn your assertion on its head, the fact that the supporters don't seem to be able (or willing) to do the math to fact check these proposals is not an indicator that the plans will work.

As a starting point for comparison, the total power budget of the ISS is under 100kW and a single supercomputer rack dissipates about 4x that. What changes to the ISS can be made to get 100x more power and dissipate 100x more heat?


Oh I have done the math. There are multiple ways to get cooling to work in space:

1. There is no real size limit to radiators in space, especially when in solar orbit.

2. High temperature chip architectures can be used, operating at 600K.

3. Heat pipes can bring the temperature even higher, such as to 1200K.

4. Special 3D radiator geometries can be used to optimize heat escape.

5. Metamaterials can be used to optimize photon emission in the best directions.

Together, these will shrink the required radiator area dramatically. Beyond these standard ideas, other exotic approaches exist at the edge of viability.

The ISS in contrast is restricted considering it has to sit in Earth orbit.

You may see this video (AI researched and generated) to grasp some of the points: https://www.youtube.com/watch?v=s7Mv_OcBXI8


Putting your single egg into multiple baskets isn't all that much better. Now you have many points of failure rather than a single point of failure.

The guardrails are channels.

If you have a mutex on a structure, linters such as are packaged into Goland will catch oversights quite effectively.

If you are using fancier concurrency structures, you should consider channels instead.


Channels are not for everything. Plenty of mutex cases cannot be rewritten as channels, or will be very unwieldy so. In fact, every large Go project I have seen uses mutex here or there.

Theoretically you can use channels to simulate a mutex, but I agree with you there are use cases where a mutex makes more sense. They are even used in the standard library, for instance to implement sync.Once.

But generally I would agree that if you need to code parallel execution, channels are a good way to do it, because you can avoid race conditions if you share data only over channels. The biggest problem is that a lot of people don't understand, that channels with a buffer larger than 1 are a sign of problems in the architecture.

There is a type of parallel programming with workers for specific functions, that always leads to performance issues. The problem is you need to right-guess the distribution of work, when you have to define the amount of workers for a specific function. At least one go routine for one request is a much better approach than function-specific workers.


Removing enough allocations to avoid fragmentation can be maddenly difficult/tedious.

You made up a group in the past and you made up things they say and then draw the inference that a different group in the present is somehow morally disadvantaged by obvious inference.

Perhaps your name-calling is not actually as logically grounded as you think. It definitely seems to depend on unfounded leaps.


Does it terrify you to look at children?

Not so many years from now, some of them will surpass you. A few years after that all (that survive to that point) will surpass you.

Does that terrify you just as much?


That’s kind of a strange comparison. It’s the natural order for a population to thrive, reproduce, age, repeat. I’m not taking a side on the original comment, but the idea of human skill being completely supplanted by AI is not the same thing as having children and getting old.

The seeming sincerity of your question in the conext of comparing children to AI is what really terrifies human beings.

AI is not a living or conscience entity, no matter what the hype men are selling society.

A child is a living, breathing, growing, and changing conscious entity. It is the natural order for the young to supplant the old, no matter what the politicians and billionaires desire.

"AI" - terrifies anyone who understands the pact our society rests upon: that labor is valued and can be exchanged for goods and services to survive. Thereby enabling a person to support their families without having to do everything themselves.

If AI replaced a noticeable fraction of society, destroying their capacity for work. That threatens and ultimately blows up this compact between working class and capital class... With it, the foundations of a modern technological society.... It may sound like hyperbole, or some fantastical prediction. But really it is basic economics, like econ 101... And personally the last few years have terrified me, not because of AI directly, but because how ignorantly blind many smart and tech savvy people are... You are marching us to collapse with a smile on your face...


Ice-nine was no fiction.

In Erdös idiosyncratic nomenclature, all the best proofs are "in the book" and it was always a joyful thing to not only find a proof, but to find the proof that is in the book.

Who cares if it is God's book or the machine's Xeroxed copy?


Long before Erdös, we had Plato and Socrates develop the theory of anamnesis, that there is no such thing as learning, but rather, whatever we supposedly learn, we actually remember (we knew it already and had forgotten it). Presumably this should be understood only of universal facts (like mathematics), not contingent facts (like who was the president of the U.S. in 1950).

Remember from ...when?

Before birth. ...Hey, don't point that pitchfork at me, point it at Socrates. In his defense, that kind of does describe when LLMs acquired their knowledge (if we consider "birth" to be the moment when the already-trained weights are sent to the GPU) https://en.wikipedia.org/wiki/Anamnesis_(philosophy)

> Before birth

Any scientific basis for this claim?

Pre-conception is unlikely to be really possible outside some esoteric circles. While in the womb there could be some limited experiences that get ingrained in the mind as memories, but I don't think that's the topic here.


I mean, my reaction to God coming down and saying they were bored of being God and instead they would just sit around and answer all of the mathematician's questions would largely be the same, so yes, who cares if its God's book or the machines Xeroxed copy?

"The Book" is more interesting to me if I am the one coming up with the ideas to fill it in. Maybe this is a bit egotistical, but I'd like to think it is allowed to have a desire that you, personally, are contributing to something in a meaningful way. Like, if you are on a sports team, it'd be more fun to win a game if you were on the field than if you were benched, and I think that's okay. And ultimately I don't find dredging for proofs from an LLM particularly meaningful, nor do I see it as a particularly personal contribution, as anybody else could have done the exact same thing with the same prompt.

This isn't to say I wouldn't love to read the proofs in "The Book" for problems I care about, I just think I'd eventually get bored of only reading. And so its hard to be enthusiastic when this book is being built through an LLM.


If ASI does create an abundant future I think many are going to have that familiar listless feeling of enabling cheats on a computer game and all the mystery and fun is gone.

Technology in general (smartphones, social media, search) even without AI is creating this feeling, as it shrinks the world and makes it less mysterious.

It's worse than boredom it's more like nihilism.

Then when you strip purpose and meaning from a human you get something very bad, despondency being the best case outcome.


Aye, but it’s also possible for people to find their own purpose and meaning. Some find it in religion, some in art, some in love or nature.

It will be a transition, for sure - there would no longer be meaning in “winning the game” in a capitalistic or scientific sense. Anything you want to produce or learn, the AI could already produce or has already learned. Now you have to do it just for the love of the process.

I have a musician friend who likes to say that good artists overwhelmingly make art for their own benefit. Not to advance the world or blow people’s minds, but because something inside of them needs to come out, and art is how they express it. And that part of us isn’t going to go away.


Basically that's Viktor Frankl's insight and it's more important than ever. Combined with the Buddhist precept of non-attachment.

> it'd be more fun to win a game if you were on the field than if you were benched

This is a good analogy for AI work displacement. Probably would resonate with some of the college students who boo'ed Eric Schmidt.


Really?!

Care to cite a reference to that proof?


"The Nature and Meaning of Numbers" (1888) by Dedekind. He proved that any set of mathematical objects conforming to the second order Peano axioms is isomorphic to the natural numbers. Peano axioms basically formalize the notion of well-behaved (that is practically useful) counting numbers.

If you were to dig deeper, you'd get to the murky philosophical depths of foundations of mathematics, but I prefer to not go there. Practically, if you want to reliably count something, you end up with the natural numbers (or, maybe, their subset that ultrafinitists are trying to formalize).


Yeah... the railway that has just had a multi-hour outage because they looked like a spam account to Google Cloud!

If you don't like the backside, check out the inside!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: