Danilo Tuler 🐧

@GiulioRebuffo @gakonst The experience is great. But there may be decentralization compromises along the way of having that kind of experience, no? Passkeys for example are great, but attached to domains.

I just tried wallet.tempo.xyz, made a wallet and I will probably build some stuff over the weekend. However, the login experience made me reflect on some things: 1. Account abstraction is actually as easy as the Paradigm bros were telling us. 2. Ethereum needs this ASAP. we can't wait for a bunch of wallets to slowly integrate this. strongly starting to believe that this needs to happen fast and waiting is increasingly a mistake.

This once again corroborates to @cartesiproject thesis, a L2 framework with a real differentiation, a powerful alt-VM for powerful applications. Stay tuned.

In my childhood Greenland was the way to invade Europe from the US, or vice-versa. (War is the Brazilian version or Risk)

@elonmusk Not a lucid decision

Tesla will stop selling FSD after Feb 14. FSD will only be available as a monthly subscription thereafter.

GM @EFDevcon and GM to everyone at Devconnect! Our contributors are on site, so here’s who you can meet IRL to talk all things Cartesi: @felipeargento, @bmaia_expat, @GCdePaula_, @carlofragni, @guidanoli, @henrimarlon_, @miltonjonat, @PedroArgento8, @dtuler 🫡

Buenos dias 🇦🇷 Devconnect

@jessepollak @baseapp More centralizing power to chess.com 🫤

more chess coming to the @baseapp

Curious about what kind of centralized infrastructure @baseapp depends on.

@davidtsocy @baseapp Me

Who needs a @baseapp invite? Reply below

Welcome @tezos to RISC-V bandwagon. Come and take a look at @cartesiproject mature RISC-V emulator at github.com/cartesi/machin… @ArthurB @trilitech

Study @cartesiproject , IMO this is how projects should be launched. Honeypot to test their proof system, immutable contracts. Such a different way to launch a product. I hope bounty will increase and there will be enough hackers/researchers to look at their proof system👇

New project on L2BEAT: @cartesiproject PRT Honeypot. Stage 2 Appchain with a (major) twist. ⚫ Before you ape into this, reconsider: You will lose all your funds. Why? 👇

Meet the Cartesi PRT Honeypot 🍯🐝🐧 Our first rollup application equipped with the PRT fraud-proof system, putting the appchain stack to the test. It’s a key milestone toward decentralization and trustless security, in line with @l2beat’s standards. ➝ cartesi.io/blog/introduci…

@felipeargento @l2beat @cartesiproject Just one caveat. Even though the TVL is compartmentalized, potential software vulnerabilities can affect all rollups of the same codebase in a coordinated attack.

Honeypots could be a guidepost for stage-2 migration We’re ~13 days out from @l2beat 's recategorization window, and @cartesiproject 's Honeypot v2 is heading to mainnet as a fully permissionless immutable Stage 2 rollup. The contracts are in final review: once they’re live we’ll submit to L2Beat and let the public process run its course. It’s going to be tight, but the external deadline has been a great forcing function for focus. A caveat on the “stage-upgrade” debate is the asymmetry between app-specific rollups and general-purpose rollups: appchain developers can permissionlessly spin up a Stage-2 instance and choose the amount of funds that enter it. If it's fraud-proof system breaks and the application settles incorrectly, only that limited TVL is at risk and the blast radius stays confined to a single application. By contrast, a general-purpose rollup settles all apps together: turning off the security council is an existential risk since every contract and every user balance inherits the same risk profile. Stage-2 activation can be straightforward (and comparatively low-stakes) for appchains, yet a high-stakes decision for shared rollups. Still, general-purpose rollups will need a credible path away from security councils and I believe the honeypot approach can help as well. We built the Honeypot as a separate deployment that mirrors the production rollup SDK but only holds funds we’re prepared to lose ourselves. If the fraud-proof system fails, we take the hit, not end users. The pot is fully open to attack, finding bugs is encouraged, there are no multisig escape hatches, and we actively encourage white-hat attempts. I believe every rollup, including the general purpose ones, could follow the idea: - Stage 0/1 "rollup" can clone itself into a Stage 2 “honeypot.” - Projects move a piece of their own treasury or team-controlled liquidity there first. - Track how long those funds remain uncompromised and how much value one is willing to stake on their proof system. At Cartesi, we usually increase the funds gradually as we gain more and more confidence. - Publish the pot size and “days without incident” as a financial confidence signal. A visible honeypot provides a concrete benchmark: “Rollup X has had $50 million sitting guarded solely by the proof system for six months; no exploit, no counsel to veto attacks.” That data point is hard to ignore when deciding whether to remove training wheels. In sum, the honeypot: - Aligns incentives: teams signal trust in their own code before asking users to trust it. - Quantifies security: the pot size and longevity become a live, verifiable metric. - Encourages responsible progression: instead of a binary “switch,” projects can scale the pot as assurance grows. I’ll try to write an in-depth blogpost on it when things calm down a bit. But wanted to share some thoughts here first. What do you think? Should all of us be launching honeypots?

@lukaisailovic @reown_ @coinbase It's nice @MetaMask and @MetaMaskDev are working on 7715 and 7702, but it's quite odd they don't support 5792, even for the classic wallet in production (and not snap and flask)

@dtuler @reown_ @coinbase 7715 is supported by a few wallets, including MetaMask as well. As more wallets adopt 7702 and smart accounts get widely available, the adoption is bound to increase

I created @OnChessProject mostly as a demo application to experiment with EIPs that tackles UX issues. The code is based on EIP-5792 and ERC-7715. Unfortunately very few wallets support EIP-5792, @zerodev_app Kernel dropped support, @Alchemy Smart Accounts don't support.

@bmaia_expat @OnChessProject @zerodev_app @Alchemy No, EIP-7702 adds more capabilities, but all those other EIPs still need adoption.

@dtuler @OnChessProject @zerodev_app @Alchemy That's why it is so important to converge on standardization of such features. Today's hard fork addresses all those needs, Danilo?

Another annoyance is how no ERC-4337 provider support local development, and rely on testnets for development. Testnet is for testing, not for development. I hope Porto by @ithacaxyz helps to improve the dev experience for ERC-4337 (and EIP-7702).

And ERC-7715, initially proposed by @reown_ does not seem to get enough traction. @coinbase wallet gave up and went to another direction with ERC-7895.

Yes, I saw @VitalikButerin 's post on replacing the EVM with RISC-V, and I’m really glad to see this direction being explored! We built a high-performance RISC-V emulator: fully deterministic, reproducible, with complete RV64GC ISA support (both privileged and unprivileged), state inspection, and Merkle tree computation. Doing that, we’ve built deep expertise in RISC-V and EEs, and we’d be glad to contribute. If you're connected to Vitalik, @tkstanczak , or others driving this forward, please pass it along...we’d love to help. We’re Ethereum too. Everything we build is open, permissive, and public. And it’d be a pleasure to, however we can, contribute.