dullcrisp

I like Alpine but I’m not a serious person.

I want this to be true too but you’re essentially proving the point by demonstrating you have to write the wasm by hand if you want it to be compact.

Sure all I’m saying is there’s a difference between saying that it’s hard to convince people of this good advice if they don’t have the benefit of experience and we shouldn’t force them to follow it if they don’t believe it, versus we shouldn’t try to give people this advice because it may deprive them of the opportunity to fail.

This post seems to take the stance that we need to cultivate failure so that we can learn. We don’t need to cultivate failure, there’s plenty of failure to be had. I mean people need to be allowed to make their own mistakes but that’s not the same thing.

Don’t get me wrong, I think having much more customizability now that maintaining complex configurations is easier is absolutely the direction we should be headed in. I don’t think floating cows are going to be the ultimate goal, but it is a step in that direction.

But this sort of theming specifically is something that was experimented with a couple decades ago and given up because people wanted simplicity and consistency instead. It’ll be exciting if LLMs reverse the trend.

I mean one where you can say to herdr, make that look like Windows 98. Still can’t think of a good example that’s not already from Windows 98 unfortunately.

We're not, plenty of applications support custom themes.

I did take your saying that it gave 90% heads to mean that it picked heads 90% of the time, not that it gave a 90% weight to heads.

I’d still maybe ask it a few more things before concluding that it’s useless. It’s meant to predict plausible responses and provide confidence scores, not compute real-world probabilities of events.

80 bits couldn’t be guessed with modern technology right? That’s still a bigger number than I know a name for. (Or I guess a trillion trillion. Maybe if we stopped building AI and just used all the data centers for this we could guess it eventually?)

For anyone here who understands this stuff, help me out because the article dropped me at the end when drawing the connection from memory to LLMs.

For an LLM, the keys are the tokens in your context, and the values are … what exactly?

Sorry, I underspecified. Your prediction must be one of either HEADS or TAILS. No other prediction is acceptable.

Try it yourself to prove you’re useful. Predict my coin flip.

It’s not really ironic since the goal of the system is to reduce variance. Or I guess ironic in the same way as it is that the Buckingham Palace guards just stand there when they could leave at any time.

Wait is unixbeard compatible with neckbeard?

Who indicted anyone?

If you need bug-for-bug node compatibility you should be using node.

Have we quantified the x-risk of Magnus Carlsen?

Same same.

Yes, just like that. What would it take for you to accept Our Lord into your heart?

Not confrontational, but proselytizing.

There are no goalposts. No one is trying to score points except you.

I don’t really understand this exercise of predicting what I’ll be thinking in the future.

No. Why? If you want to convince people something is true you need to actually convince them it’s true. They don’t need to justify not believing you.

You have a practical obligation to obey existing power structures. That’s what makes it a power structure. What do you all think is happening here? There have been people who decided what math gets published for centuries.

They’re not asking them to stop working on AI, but to stop publishing mathematics.

This article doesn’t present much motivation for why you would do this.

To me it reads a bit like, how we built our office without desks: it turns out if you stack two chairs on top of each other, you can balance your laptop on the top and you’ll also have a shelf on the bottom for your things.

What if it turns out that if all the AIs cooperate they can end the world really easily without finishing any taxi rides?

They’re making an oblique reference to Pope Urban VIII.

> Perhaps you might see fit to conclude your earthly endeavours at your earliest convenience?

I’m stealing this.

> I did not like it at all.

Because it left you fewer opportunities to experience the ecstasy of typing git commands into a terminal? Or because it didn't work so well?