Once I'm happy with the idea, I go head to create an implementation spec, and chat back and forth with the implementation spec.
Here i like to talk technology tradeoffs, implementations, questions, etc..
I save both the PRD and Implementation docs.
Jun 29, 2023 • 9 tweets • 2 min read
How do we make account abstraction work? Everywhere. Cheaply.
Well, @tallyxyz and @LimeChainHQ are building a proof of concept that is going to blow your mind.
The pitch:
Universal Account Abstraction with zero setup for users and near zero additional fees for relayers.
🧵
At @tallyxyz we started building relayers for delegations and voting, and realized that building universal AA is expensive and bad UX.
But there is another impediment, it's opinionated and difficult to implement.
You need relaying infra on every network.
Oct 28, 2022 • 15 tweets • 4 min read
New theory:
As more and more L1s role out, EVM compatibility will become:
A) less important
B) a negative signal
Why such a radical change of opinion? When I previously thought EVM compatibility was critical for a blockchain success?
A mini thread-
Previously I believed you needed to have EVM compatibility to make inroads in the blockchain developer community. No one has time to learn new coding languages, especially exotic ones.
I think that *was* true, but something changed.
Jun 10, 2022 • 13 tweets • 6 min read
A quick thread on how I think you capture @gnosisSafe.
(This is a hot-take so, please, get fired up in the comments, but keep it civil)
Short answer: you silo the future.
Curious?
A 🧵...
Part 1: The future
On the blockchain, the Meme pool is public everyone can see all future possible transactions and the miners order them in the consensus process.
With Gnosis safe, before transactions go onchain, signers coordinate and send them to the @gnosis tx service...
Jun 10, 2022 • 4 tweets • 1 min read
Seems hard to imagine we don't find the bottom in that gray box...
Looks like we're already on our way....
Aug 8, 2021 • 4 tweets • 2 min read
I don't really agree with this piece and I don't think that VC heavy governances are automatically a bad thing. Not every project is a Yearn. A bunch of rich people who collaborate together in a decentralized way, are still people who want to collaborate in a decentralized way.
Them being rich and having outside influence on the thing they are collaborating on building, doesn't automatically make it bad. Some projects are bootstrapped from the ground up, some are bootstrapped from VC's. DeFi has plenty of options for everything in between.
Aug 2, 2021 • 6 tweets • 4 min read
I just got a great idea for a massive cost improvement on the L1.
Rather than passing in an JSON object on the smart contract, we pass in only IPFS hashes.
The Subgraph can load the IPFS hash which can include an arbiltrarilly large JSON object full of instructions.
It's Sunday morning. Who's wants to hear something *really cool* I hacked together yesterday?
Let me give you a hint:
Build a blockchain inside @graphprotocol subgraph that inherits the security of Ethereum but smart contracts w/ flexibility of assemblyscript? Any takers?
The idea came originally from the discussion I've been having with @auryn_macmillan on his super-tweet thread about the Poster Contract. He mentioned to me "Could we make a token?" And at first I was like, "no".
and then I was like, "oh yesssss"
Sep 16, 2019 • 7 tweets • 6 min read
@_JohnWhelan@OpenZeppelin@dappcon_berlin Absolutely! If you like going down the rabbit hole, this is a great place to start. In my demo Chat App (gsn-chat-app.openzeppelin.com) users make real Rinkeby transactions using GSN, but they are actually totally anonymous. (I put up a bad word filter just in case: 🤪)...
@_JohnWhelan@OpenZeppelin@dappcon_berlin ...but you can easily imagine with Relayers processing transactions for users, we can build some really amazing privacy protecting features. Or we can simply make our (d)app funnier, more popular, and a heck of a lot easier to use. It's a tool dev's are just starting to explore.
