InsideOutSanta

Then everything else is documented inside Bitwarden. It's just a matter of installing the apps and logging in using the credentials stored in Bitwarden and a 2FA token generator.

There's no need for relying on any phone OS features.

That said, I think protecting your privacy is a good idea in general.

I don't think it ever makes sense to transport actual data across borders if you care about governments gaining access to it.

A few years ago, I think there would have been a market for a HyperCard-inspired tool that targeted modern technologies - probably web tech.

Today, I guess vibecoding is what HyperCard was in the 80s and 90s, the tool you can use to quickly and easily build a little program that fills your specific need.

I agree with you that sheer scale is the primary issue, but for a particular reason: these systems are all prime targets for malicious actors intent on manipulating people on a large scale.

I should also think that libertarians should be in support of private people deciding not to support companies or people that support causes they disagree with. That's an excellent example of people's freedom of action.

The ability to generate a lot of code will not solve any of these issues.

"Eager" is good. "Overeager" is definitionally bad.

Based on my entirely subjective experience, the $100 Moonshot plan using only K3 is comparable to the $200 Anthropic deal using the whole Fable allocation and Opus 4.8 for the rest.

Instead, I've settled on a top-down approach: I write an in-depth spec for the whole app first, then have the LLM implement reviewable sections of it.

That's not a meaningful number, and even if it were, quadrupled isn't nearly enough.

> I started with a simple hello world application. Then added in the habit list, the ability to add a habit, edit a habit and so on.

This hands-on, incremental approach works, of course. But I don't think it is the most effective way of using LLMs to write apps.