Shane Mac
@shanemac.eth
Launching testnet is a huge milestone for XMTP. This is real progress toward our long-term goal of building a truly decentralized messaging protocol and network that has Signal-level security and crypto financial rails built-in.
6 replies
11 recasts
42 reactions
Shane Mac
@shanemac.eth
So grateful for @coinbasewallet, Circle, @ensdomains, @alchemyplatform, @a16zcrypto, @masknetwork & Faction for seeing the opportunity in XMTP and supporting the network as node operators. Can't wait to see who jumps in next.
2 replies
1 recast
9 reactions
Shane Mac
@shanemac.eth
Let's be real, though. Moving people to a new decentralized messaging network will be hard. We need your help to make it a reality. Crypto builders around the world have proven what owning your money means and why it matters. Messaging is next.
1 reply
0 recast
4 reactions
Shane Mac
@shanemac.eth
It's time to bring real ownership–via decentralization–to messaging. It's too important and too personal to nearly everyone in the world to entrust it to just 6 CEOs and their centralized corporations. Learn more about the XMTP testnet and where we go from here. https://paragraph.xyz/@xmtp_community/xmtp-launches-testnet-building-the-worlds-most-secure-decentralized-messaging-network
2 replies
0 recast
4 reactions
Thumbs Up
@thumbsup.eth
Interesting stuff. I have some questions though: 1. How is privacy of the USDC handled? For example can one track from whom to whom transactions are being sent or is it obscured in some way. Especially as this relates to node operators (see question 3). 2. Is there native account abstraction? If so what features are in place to maximize privacy and security (feel free to point me to a docs page that explains this). 3. Is there a mixer or other obscuring intermediary between the user and the node operator? If not, how do you prevent metadata leaking with every message? For example, at the very least you’d be able to see onchain every time a user sends a message, no? If so, and the network is not very busy, you could see back and forth txs from two addresses and infer that they are communicating, in real time.
1 reply
0 recast
1 reaction
Nick Molnar
@nickm
1. We are expecting most messages on the network to be paid for by the application, not the user. So "this message was sent by this app" is public but there is no connection to which user sent it on the network (client signatures are inside the ciphertext). 2. We do support ERC-1271 and ERC-6492 at the public identity layer. But there is no way to track which messages relate to which identities unless you can decrypt the message 3. The Payer (run by the app) sits between the user and the network and mixes together all messages from the application. Users might choose to have additional layers of disintermediation between themselves and the payer, but we treat that as optional from the protocol perspective.
1 reply
0 recast
2 reactions
Thumbs Up
@thumbsup.eth
Very nice. Thanks for the responses!
1 reply
0 recast
1 reaction
