crypdough.ethᵍᵐ🦇🔊X 🏴🦀

It's an honor to launch D_D Cloud today -- a truly special occasion for myself, @futexlabs, @developer_dao, and the amazing contributors who helped me along the way. We will not just be the backbone of web3 infra, but the internet as a whole ❤️‍🔥

Resist the Digital ID. Resist mandatory Internet KYC. Resist social media app Face ID scans. Resist age verifications. Resist VPN bans. Resist the chip. Resist the biometric technological panopticon. Resist the social credit surveillance state. Resist until you can't anymore.

look I couldn't care less about agentic payments. It's just some autonomy theater so "your AI" can subscribe to a SaaS.

@hummusonrails I wrote my own small payment processor for @futexlabs This is the basic flow: - User does SIWE (only once per wallet) & fetches payment wallet address - User submits tx hash for settlement & verification -> Plan selection -> Create API keys for service

@hummusonrails I'm leaning towards these flaws being fundamental to x402's design. MPP seems more flexible i.e supports pull and push payments, but still commits the cardinal sin of moving payment submission / verification to hot paths. This causes most of the problems in both protocols imo

Why do we need a proliferation of protocols? What was wrong with x402? or... What was wrong with x402 that couldn't be adapted in the existing protocol? Protocol proliferation == more builder headaches

More on the last point: cutting the settlement and verification parts up into two phases means that correctness is thrown out the window. Tokens can be transferred or spent before settlement, leading to the service not getting paid

x402 isn't great: - Adds >2x the latency of paid API endpoints. - Handling private keys is far riskier than managing API keys. - Not viable if the per-unit costs are low enough b/c gas. - Every micropayment can be a taxable event. - Verification & settlement are non-atomic.

Silicon Valley was really so ahead of their time

GM, just released a new open source string library. Inspired by the "Put a limit on everything" rule in the Tiger Style style guide

@craigwarmke Jarvis, write me a paper on early modern philosophy and submit it to Nous. Don't stop unless the paper is accepted, make no mistakes

We're in a sweet spot, and it won't last for very long. Right now, as a non-programmer, you can build something serious. Even with claude code, it will take months. But that's where the value is. Everyone else is fiddling with openclaw setups, plucking the low-hanging fruit, or trying to build big and quitting. But what if you don't quit? You basically need to be unemployed for 2-3 months in order to build - and that's why so few people will do this. What's valuable to build now generally can't be done in mere hours or days. You have to iterate. Get feedback. Scaffold. Power through obstacles. Attack something from all angles, as a 1-man company. It's cognitively demanding - being creative and making executive decisions every few minutes like a CEO on hyperdrive. But, if you stick with it, you can collapse what would take a team over year into 2-3 months for a single person. This window for non-programmers will narrow - it just depends on how creative your idea is. The less creative, the sooner your window closes. Because lots of real programmers will be actually unemployed, and they'll be forced to innovate. That's when we get a cambrian explosion of creativity. That will be great - and I don't think we're there yet - but you want to beat the herd. So if you're a non-programmer, now's the time to build something of deep value. It's the time for idea men to speak their ideas into existence.

@glennonchain There's some identity tech out there but the reqs are pretty tight for kyc/aml. This is a big challenge for new rails because interop is a necessity. There's no guarantee that a person can be verified right away, which instantly cuts conversions for the non-stablecoin crowd

@crypdoughdoteth agree but is this more of a problem to solve with new payment rails, or with identity / cryptography?

My hot take is that payments is one of the most boring verticals to build a crypto product in right now Incumbents will simply integrate any useful parts of defi / blockchain and continue to dominate Privacy, identity, censorship-resistant tech are all much more interesting IMO

The first pure DeFi mortgage is 90% funded. No minimum amount required. Anybody can contribute and be part of onchain history. Are you in?

@resistancemoney It really did. Philosophy was my first academic interest. I wasn't really motivated as a student prior. Fortunately, reasoning well applies to many things re: hazy details -- I had trouble recalling Kant's categorical imperative in a conversation w/ @sachalansky at ETH Denver 🥲

@crypdoughdoteth I suspect it served you well, even as the details get hazy over time.

There are a ton of philosophy-trained people in tech, whose training shows, but who now poo-poo the humanities. Guy climbs ladder, pulls it up behind him. Guy writes essays about how you don't need ladders.

For “infinite software”, software the whole world needs (as opposed to bespoke software): - build cost is always ~zero - the end-to-end value derives never from the build cost in the first place - but from the safety, performance and operating experience in production.

Our friends at @developer_dao are hosting an Après Ski afternoon at our place. Come hang with us in Denver!

Sharing my @invisiblgarden session: “Multipoint KZG Proof Verification" Verify many openings for one blob commitment in a single call, slashing gas costs think (rollups, fraud proofs) Kinda surreal this is now EIP-8149: eips.ethereum.org/EIPS/eip-8149

👉 Are you heading to @EthereumDenver? Join our official IRL event, Aprés Ski, with @joinwebzero for some vibes, technical discussions and more! Spots are limited, so be sure to sign up quickly to lock one in 🔥 📆 Wed, Feb 18th 🕐 5-8pm MST luma.com/11828c74