Avatar
Super Testnet
2183e94758481d0f124fbd93c56ccaa45e7e545ceeb8d52848f98253f497b975
Open source dev w/ bitcoin focus | supertestnet.org bc1qefhunyf8rsq77f38k07hn2e5njp0acxhlheksn

seems like your goal is to make op_returns cost extra for their creators, and I suspect there's a simpler way to do that: create a rule that says transactions containing op_returns must have a feerate X% larger than the average in that block

the great thing about app-layer development in bitcoin is, you don't have to reinvent the wheel

you *get* to

who writes two tests for the same function?

Does "measure twice, cut once" have an equivalent in software design?

Something like "Spec it out twice, code it once"

Why use L1 addresses as the identity? Why not use nostr keys as the identity? It seems like "identity" is what nostr keys are supposed to be, and bitcoin addresses are definitively not supposed to be that

in my experience, nostr users are almost all bitcoin maxis, and almost all bitcoin maxis have a self custodial lightning wallet. Therefore, I expect almost all nostr users to have a self custodial lightning wallet. And, as everyone in this thread is a nostr users, QED.

> Seems like the bigger nodes get the more funds are controlled to me

Not to me

I do not know which wallets have more capital in them, custodial ones or self-custodial ones

what feature of bitcoin would an ERP system use? if the idea is just "Oracle NetSuite, but denominated in bitcoin," I think Netsuite already has you covered: bitcoin is supported as a foreign currency option, and it will handle all the currency conversions for you.

hmm...so a webstore, but where you can say: 10% sats off if you (1) follow me on nostr (2) I follow you back (3) we've interacted more than 5 times there

> If you send, they need to trust you, if you receive, you need to trust their mint

I think you misidentified where the trust assumptions enter in. It's not related to whether the button you pressed is the Send button or the Receive button, what matters is whose mint holds the money. If your own mint holds all of the money, you don't need to trust anyone, regardless of whether the button pressed is "send" or "receive." I think it gets interesting when each counterparty holds their own money, and interactions happen via atomic swaps.

You might say this is "essentially no better than if you used lightning," but if you think that's a put down I disagree. Lightning is very powerful but hard to work with, partly because different backends have different APIs. Cashu unifies everything: regardless of what backend it runs on, it bolts a unified scripting interface on top, and thanks to this interface, I think it's easier for coders to programmatically authorize payments from a user's node, process refunds in the event of a protocol failure, and perform swaps.

Also, it's very useful for prototyping. Perhaps some of the things I mentioned can't be done on cashu without introducing trust assumptions. But since cashu's scripting language is similar to bitcoin script, that means if you can build a prototype on cashu, then even if it's partially or fully custodial in that environment, you can then copy the design directly into bitcoin script and make a self-custodial version. After that, it's probably wise to lift it onto layer two, but cashu gives you a taste of that in the meantime, which is very meaningful for early adopters, testers, and scientifically minded folks.