Reviewing/refining/QAing code for two long range things we've been working on.

You mean reward sats? Mainly, it's legally not your money until you withdraw it. Also, it's our money we're sending you - not money we sent you on someone else's behalf.

If you can going to comply with BSA/MTL, there's not much point in using ecash as it defeats the purpose (privacy).

It is. No one had a full map then though I'd guess.

Crazy to think about. I wonder which graph we'll have in another 1000 years that demonstrates how naive we are.

I was suprised to learn Alby Go also has it builtin. I guess it's a thing.

Would need to be a QR of a bolt12 or lnurl. Good luck normies!

Agreed. I think they're trying to appeal to younger generations - who aren't driving and certainly aren't driving for fun/speed/adrenaline.

For that purpose, the UX of the interior looks pretty great imo.

I agree. While I focus on consumers in my old comments, I should be more upset about how it misleads producers. Internet money doesn't eliminate free riding.

The interior is really well designed from what I could tell!

tldr

1. meaningful improvement over rug pull incentives
2. cashu mints are still money transmitters

It's all-in-one viewing, editing, prompting, browsing (it has a browser), and it has a terminal (which I don't use). I also like being able to model switch when one model appears dumb or so I can pit them against each other.

Thanks for the push. I'll run some side by sides and see how I might use it.

Thinking about how I improved my writing. It's mostly about quantity and having a small impetus to write in some quantity. Writing numerous small things translates well to writing longer things and you learn to desire it.

I imagine this can translate to other subjects too. I think the most important part is the daily release hint to stop students from precrastinating and cramming.

The guys in the lab like it, but I've only used it as an implicit sub-agent.

If I were to teach a writing class today, I'd give students a writing prompt per day, releasing one prompt per day, with a minimum word limit and no maximum. The writing prompts would be voluntarily graded, strictly for feedback, if desired. Grades would be determined by written exams.

That tracks. Also, the process it speeds is the same process that assists programmers in understanding the code. So if a programmer still seeks to understand the code, it is paid in a lump sum at the end now

AI or not, the bottleneck on great code-work remains understanding the code, customer, and context. Until LLMs understand all of that as much as we do, and can prioritize and orient itself among their understanding, great work will continue to require great oversight.

After switching from Cursor to my Codex/Claude subscriptions (for cost savings), they either neuter the subscriptions a little bit, Cursor's harness is just that much better, or I'm so used to Cursor that I can't evaluate clearly.

This is a bot

Whose role is it to produce the electric Miura that I can't afford? They are fucking up.

You can view the details here: https://github.com/stackernews/stacker.news/pull/2981

There are images toward the bottom.

why?

Because wallets are the biggest onboarding hurdle. Folks come here to play with sats and it's too hard as is.

The plan is to add internal wallets in a later phase: Spark, Lexe, Ark, fedimint, Breeze, whatever else we can that doesn't require channel management in-app.

Then, if we can make it so every stacker has a great wallet, we can remove cowboy credits.