MattGaiser

I end up accepting as it’s far easier to just turn it on low volume and listen for my name.

In general, there’s very little info that costs much to learn nowadays. The human standing in the front is a disciplinarian to force you to learn it.

A lot has to do with the story. Nobody would likely listen to this as pure music.

No easy way to fight against them happening. Can I spend $40 on my productivity? Not easily. Can I spend $1000 on a meeting? Trivially.

Millions of people using Claude code couldn’t write hello world in the language they are using, if they even know what it is.

I am anecdotally aware of at least one project where the author can’t recall off the top of his head what the stack is.

> If so, why is the pace of innovation and updates in their AI development products (Codex, Claude Code, Antigravity) so pedestrian?

Claude seems to be pumping out features fairly rapidly?

AI is apparently cheaper for cleaning my feet and cutting my hair.

A lot of it is simply that they are far more open to the idea of curiousity as having value than most people.

Aren't all of these Accorp/Delve reports?

That is probably the most notable example and in the end, those few still lost.

What has the tech industry ever resisted on moral or reputational grounds?

I do some AI training work. I haven’t seen data analytics tasks in over a year. That seems to be solved as far as the AI people are concerned.

> They were hired on the expectation that they'd do their own work.

Is the company opposed to LLMs? As this is a very specific expectation in this day and age and companies seem to be going the opposite way.

And I would bet he judges your work with AI, assigns you work generated by AI, and perhaps evaluates whether you yourself use enough AI.

> You wouldn’t commit a binary.

If a binary was as expensive to generate as some AI projects, I suspect that we would. We do not commit a binary because a binary is cheap to produce locally.

Maybe we get to the point where a conversation can reasonably be used to reproduce software, but that is way too costly at the moment.

> Students’ reliance on AI “ has paradoxically raised the floor of class discussion to a generally better level in courses with difficult concepts, but has also tended to preclude stranger, more eccentric and original thoughts,”

The challenge is that this is overwhelmingly the reason people go to college in the first place. They want to raise the economic floor of their lives and have greater certainty in the outcome. Precluding strange and eccentric outcomes is certainly acceptable or even desirable with respect to that goal.

> I have no reason to expect this technology can succeed at the same level in law, medicine, or any other highly human, highly subjective occupation.

I mean, if anything, I would expect it to help bring structure to medicine, which is an often sloppy profession killing somewhere between tens of thousands and hundreds of thousands of people a year through mistakes and out of date practices.

As medicine is currently very subjective. As a scientific field in the realm of physical sciences, it shouldn't be.

Even if you don't want to do yolo mode, there are things like Copilot Autopilot or you can make the permissions for Claude so wide that they can work for an hour and let you come back to the artifact after lunch.

I would argue SEO should already be considered dead internet theory. Most of it is not intended to do anything but convince Google.

A dentist buying freelance articles from a guy off Upwork is not intending to communicate anymore than this guy generating articles is.

One of the issues is that the purpose of business internet writing is not to be read, but to be ranked well.

Many organizations are extremely low capacity. I am aware of a company that has merely a few hundred gigabytes of data and is stuck on Oracle.

The difficulty isn’t in shutting the island down. That is undisputed.

It’s what the Iranians will do if you shut the island down.

Maybe they will incinerate less as a result, but they will still shut down the rest of the gulf.

The path of reasoning the agent took that led it to generate the output. The GitHub search bits got posted after my comment, so while it is clearly real, it just seems injected by Raycast.

> So why not do that?

Iran incinerates the other oil infrastructure in the Middle East and the world enters a global recession?

As usual with US military thinking, there is limited thought to the day after taking the action.

It could simply be something in the Raycast integraton?

I actually like that I don't have to leave Github to deal with various feedback, especially if I switched branches already to do other work.

Post the trajectory if this is real.

Well, it never cost much. For the most part you just need to install it into your test framework.

The problem is that the value of doing it is essentially none.

Every company I have worked for has passed on fixing things that just impact Firefox.

In my first job back in 2019, a support ticket came back about a dropdown bug in Firefox. It didn’t even make it to engineering before they told them to switch to Chrome.

You missed the point of my comment.

> It isn’t going to happen. Planes don’t run on solar. Boats don’t run on renewables. The lubricant needed for wind turbines comes out of the earth. Dams need the same lubricant. Building roads, oil. Installing renewable infrastructure, oil. Running combines to harvest vegetation, oil. Building renewables requires massive amounts of oil.

All of this? About 30% of oil usage on the high end. You are listing the small uses for oil.

May some oil always be needed? Yes. But nowhere near as much as we produce today.