__alexs

Claude Ultra, No, Claude Code, Claude Code can orchestrate agents, No, but I run Claude Code on a VPS of its own for some things.

We have an OpenSWE agent on Slack at work but it seems horribly expensive and slow for anything I want to do.

We will never be free of the tyranny of aesthetics.

Given how trivial this seems to be to try, has anyone here attempted to reproduce any of these findings yet?

Exactly. Unions can amplify the bargaining power of labour but they don't magically delete the other variables.

You seem to be the only person that got my point so thanks I guess.

I am actually pro Union but yikes the left needs to go back and actually read some Marx.

Unions do not exist in a vacuum. They have to operate in the same fundamental power structures as everyone else.

If union A is like "work less let AI take the slack" and union B is like "work the same, use AI to achieve more, let AI help us win" I'm joining union B every time.

Vendors will just redefine the meaning of bug if needed.

The whole idea of a confidence score as some abstract thing separate from statistics is wild to me.

Asking random experts their confidence is already pretty bad most of the time.

If you want a confidence score you need a way to consistently model the system you are trying to estimate the outcome of.

Anything else is just numerology.

The problem with this analogy is that we already literally shit in the actual sea so much people get sick and die sometimes and yet people are still getting in the sea.

You've just decided that your level of abstraction is better than someone elses but where's the evidence it's actually better? This has been the main tension between programming philosophies forever.

Just because something makes you comfortable doesn't mean it's actually good.

Models do not have durable process with skills either because every module update changes the interpretation of the skill.

The main value of skills has always been to advertise ideas for stuff to do with LLMs more than anything else imo.

System 1 Vs 2 aren't even Kahneman's invention. They are from Stanovich and West.

I think you're probably right that's what he thinks, he's just failed to convince me to think the same way.

Blow's game design philosophy seems to be mostly about repeating variations of the same idea as many times as possible regardless of whether those variations are actually that fun or interesting.

In theory I should love this, I'll happily play repetitive games like Tetris for hours. The way he does it though just feels so boring because he doesn't do it in a way that makes you feel like you're learning anything until the very end and the bits in the middle are just the same idea with some gimmick that isn't that interesting.

I'm not spending 450 hours smashing my head into a wall before I get to the good bit.

You probably have the mouth bacteria that doesn't cause cavities and don't consume litres of acidic drinks every day?

This isn't my experience in the UK at all. Dentists routinely try to overtreat you and second opinions consistently disagree.

CodeRabbit emits LLM generated poetry in its comments so maybe more than you'd think?

Most people who talk about AI consciousness don't even know what monism or dualism mean.

Manager's do not build anything despite what the management-class might want you to believe.

Do the labs have a genuine interest in mathematics beyond generating interesting marketing material?

Negligence is also a well understood legal concept.

This was literally the first coding job I ever got paid to do.

I would decompile the Google browser extension to extract the hash algorithm they used to sign page rank lookup requests, rewrite it in PHP and ship it to SEOs on weird forums.

They updated the hash every few months and I made enough money to pay for some summer activities as a teen.

System 1 isn't even a correct theory in the context of human reasoning, it's just pop-sci nonsense.

If you are reading this sort of post and feeling inclined to share your own opinion, please take a moment to go and read about monism vs dualism before you start.

Sounds wildly overcomplicated.

In my testing Jev is not what I would call "usually correct" on most topics that involve knowledge of the world outside of the context you give it.

There is no evidence Aardvarks in egypt at that time. I think we'd probably expect at least a few aardvark bones if it was.

If this is not a dog then we should cast doubt over the identity of the other animals we think we know too.

How are people using custom harnesses cost effectively? Do they avoid Anthropic models so they can use OpenAI subscription pricing and open weights stuff?