sejje

Why is that a reply to my comment?

I have cookie banner blockers, too. We get plenty of those banners in the US.

I will never remember that, but I could instantly remember steam.com

I like this idea.

What's your security/cryptography background?

> I am pretty sure, this thing solves half of the problems the above solve, and it does it half-good.

What makes you pretty sure?

Try AdGuard DNS or an adblocker. I didn't see any ads on the site.

I think it's really worth considering self-hosting these days. Doesn't take much of a machine to run a web app, for instance.

I do a little of that, but for anything public I put it on a VPS.

I use one of the big providers; they're all the same. Empty linux virtual machine with an IP. That's all you need. They start at $4/month or so, maybe cheaper if you shop around.

This whole website is nothing but tweet-sized interactions.

You have big shady trees? That'll make it hard to grow grass.

Some trees put out a chemical barrier to growth, as well. Red Cedar and Black Walnut do it around here.

My uncle had a bunch robbing his stationary (dog) feeder. So he shot one and hung it above the feeder.

It solved the problem.

yeah--the reason is the bloat and overhead electron brings.

nobody cares about the web layer

> gpui.rs

I clicked. Never trust a gui library without screenshots.

I'll never quit building TUIs, though. I'm building more TUIs now than ever, with AI assistance making it easy.

Right, so we should have both options.

We're not on the cusp, we're in the process.

World-changing doesn't happen overnight. The internet didn't, either.

This is like 1991 internet.

> The man was accused of plagiarism, and there was no final verdict on if he did.

Bullshit. It wasn't the first round of plagiarism, or redactions. His whole supposed life story was a fraud. He passed off other researchers' patient interviews as his own.

Because something was wrong in his brain and he couldn't handle the stress of his situation.

Many people have faced similar or worse charges, and did not commit suicide, so we'll have to point at something else.

Fuck them for what they did to him, but we should be rallying to get him set free. We can't start pointing fingers at other people for suicides. We have agency.

JSTOR didn't want the criminal charges

What did you bring over from prime-agent? (I use prime-agent as my daily since it launched)

I primarily like how it manages sessions, and how agents can easily reference other sessions.

Versions of "get better at [thing]" that don't involve practice are largely misled. There are rare times where you can read and gain a new skill, but largely you will have all the available skills, and you need to level them up.

Most people never truly practice. They do things that resemble practice, but doesn't have the same effect.

My example is from pool: many players have played pool league for 20+ years without improving their rank. They "practice" every week by playing with their friends. Nothing about pool is, in my opinion, hereditary / "natural talent"...athleticism doesn't matter at all.

I became a better player than those others in a couple of months by focused, deliberate practice on the things I was worst at. Later, I practiced one skill (a draw shot) for countless hours--I was bartending in a mostly-empty bar. I became the best in town at the skill in a couple of weeks.

In all my time practicing, I never saw another person practice. People want to try all the other shortcuts and tips to avoid "hard work." It sure pays off, though.

It may be ipython making it work well with ds flash, too. I haven't really run many separate experiments, to be honest.

I also like prime-agent's way of handling sessions better than any other harness i've used. You can run multiple agents from one instance, although the scoping could be better.

But they can interact with past sessions, so preserving context isn't as important all the time. I just tell them to search for [thing] in another session.

It seems to have no problem with all the skills and things the other harnesses are using. I use superpowers and ponytail a lot.

It's my daily driver now. I like it better than opencode. But it doesn't ask permission. So I put it in a VM.

I run sol on medium about 90% of the time. Maybe that's worth a try. Especially for a small change.

I think harness/model pairs matter more than your analysis lets on.

I've had great luck with the ds flash v4, paired with prime-agent for the harness--I like the results a lot. And you get to see thinking tokens.

I haven't liked the model as much in opencode.

Sol & luna have been great everywhere. sol plans, luna builds.

Some people have alternative models review each other's output.

The future might be parallel attempts and selecting the best-of-N solutions.

Aaron was a figure in the community for a long time, beginning when he was a child

I'm not a lisper, but it's one of the oft-touted wins of lisp.

This is easily the best time there's ever been to be a human. And while I think many creatures' existence is pleasant, I would never trade the human experience for another.

I think the world is what you make it, in Western countries (standard caveats like disabilities). I empathize with misery experienced in other places, but it has nothing to do with AI. It has to do with humans, bad decisions, power monopolies.

Anyone miserable in America is choosing it, or very unlucky.

I've been using ds v4 flash with prime-agent (a pi derivative), with opencode go, and I really like it.

The idea guys will really think themselves gods, now.

Google search is so bad compared to peak google.

Maybe the web is really that much worse, and the SEO tactics so hostile to genuine content.

But, honestly...just bring back the old google, where I have all kinds of search modifiers to perform exactly the search I want, that just returns all the matching results that have been indexed. Let me sort out the rest. How do you remove the ability to do "exact text search"? It's the most basic of search functions. Remember having "|" modifiers? AND modifiers?

If google launched that again, even as a separate engine, I think they'd have a good product again. Maybe being good doesn't pay the bills for google, though.

I never ask for time estimates, but all the models will sometimes give me them. It'll sometimes estimate 1-2 weeks for something that is ~ one more prompt and 10 minutes of waiting for the model to churn.