As you may know, thanks to the support of the DHF, I've hired a team to work on making multisig actually usable on Hive. Here's the original proposal.
Backend
Now that the core of the one-stop multisig frontend is ready, most of the work has been towards getting the most important piece, the backend, ready.
The backend is operation-agnostic and only knows who needs to receive the signature request (the operation itself is encrypted for each potential signer).
So far, we've created our data models, and enabled the back-and-forth communication between the frontend and backend using socket.io. Potential signers that are already connected to the backend immediately receive the signature request, if they are not connected yet, they will receive it when they connect, if the transaction is not expired yet.
Additionally, when the number of signatures is sufficient, the backend will request a user to broadcast the transaction. This is to avoid multiple broadcast tentative for the same operation.
Although the core features of our backend are ready, we still have some work left to do on this side :
- Improving the authentication process
- Housekeeping (mostly deleting data after operations have expired)
- Notify signers after the operation has been broadcasted or has expired
We estimate that this should all be ready within a month.
Other work
Once the backend is ready, we will work on an SDK to allow easy integration by third parties, the accompanying documentation, and the integration within our own one-stop multisig dApp.