rukuu001

Spend any time examining the behavior of gambling companies and this will be utterly unsurprising.

The way they operate their business is deeply immoral.

They know they sell an addictive product and take full advantage of that.

These are all good questions.

Perhaps we can think of projects like these the same way we’d think of a good tweet - not valuable in itself but valuable because of the ideas or conversations it generates.

The thing is the LLM interfaces encourage people to write their own apps.

Claude especially, where you might say “how can I do task xyz more quickly” and Claude responds with a little JS app. This creates an expectation that it’s possible to do anything if you ask Claude right.

I have seen this first hand where my partner - a deeply non technical person - came to ask what an API token is and how she can get one for Canva. Claude was busy automating her Canva -> Pinterest workflow.

She now has a half dozen of these things and hunts through her Claude chats to reuse them.

Once I asked, puckishly, Claude Code to “follow the instructions in this directory” when there was only an AGENTS.md there.

In the manner of someone finding a dead mouse and holding it up for examination CC said it could find no instructions but perhaps it should check this AGENTS.md file.

Man I thought the giraffes were bad enough

This is the bit I don't get:

> My software factory was intentionally set up to let the model decide the how of the workflow entirely. It was free to manage its own context and could maintain its own records in an agent-notes folder.

The experiment becomes a crapshoot. What are we evaluating? The ability of the thing to create it's own factory workflow? Or adding virtual threads to Python?

Astra is clearly both formidable and imperfect. Anyone who understands how to get the best out of it will have a strong advantage.

(For me - my CC is stuck in Sonnet and consumes Trello cards that have passed readiness criteria)

It’s a good question. With recent stories about OpenAI’s agent swarms’ unmanaged collusion I thought models like that start to look like a strategic asset, geopolitically speaking.

Which means everyone wants one, and governments will want to control access and use of them.

I think we’ll be back at ‘U.S. citizens only’ access to leading models soon.

Hey sorry to hear about this. I see a lot of your comments and enjoy them.

I didn't understand any of this until I had a relationship with someone that had suicidal thoughts from time to time.

This echoes much of what she described - constant pain is hard, and suicide offers immediate relief.

It's an experience of reality that's very hard to understand for anyone who hasn't been close to it.

@sillysaurusx thank you for sharing your experience so clearly. Best of luck

A couple high-profile crash & burns will get their attention.

Curious, how do services like this stay compliant with export controls on these devices? I assume a credit card is all I need to access these things?

> This performance pattern^H^H^H^H^H^H BLOG POST may indicate attentional difficulties, and taken together with other findings, an increased likelihood for the existence of ADHD.

Ha! In 97 on I proposed a web page that would read content from a database (for multilingual support). My supervisor demanded I decide if I was building a web page or a database because I couldn’t do both!

Yes, seen this happen in Tasmania, Australia with crayfish, scallops and orange roughy. Each went from free-for-all to highly managed quotas within a few years. Roughy fishing stopped completely back then, don't know what the status is now.

> we don't really know much about how fish live and their overall population

Absolutely yes. Orange roughy is a good example - it was caught by trawlers when they were massed in huge balls near the surface. Easy to get a big catch quickly. Turns out this was when they were spawning, so breeding cycle was being interrupted, which led to catastrophic declines very quickly.

Key learning - if you’re pregnant travel with a friend dressed as Batman

As apps get larger and more complex, I get the LLM to produce 'developer documentation'. I find it helps them stay on the rails, and it helps ME stay on top of everything.

The 'spec' frameworks are documentation as well, but can become contradictory as decisions about the app change over time.