reply on: Question about multisig? \ stacker news ~bitcoin

pull down to refresh

1383 sats \ 6 replies \ @Murch 4 Oct 2023 \ on: Question about multisig? bitcoin

If you are talking about scripted multisig, i.e. one that uses either OP_CHECKMULTISIG or a leafscript with OP_CHECKSIGADD: No, because the address is a representation of the output script, and the output script includes a definition of which three public keys form the quorum. Changing the quorum will therefore change the output script and a different output script will be represented by a different address.

If you are talking about a P2TR output encoding an output that under the hood can be spent with a 2-of-3 threshold signature (e.g. using FROST or ROAST): Yes, because you can use two private keys to create three new shards of which 2-of-3 are sufficient to produce a valid signature. This allows you to rekey your wallet without moving your funds.

0 sats \ 3 replies \ @Murch 4 Oct 2023

(Reposted to fix missing negation.)

Anyway, you shouldn't reuse addresses, so you should not be too worried about being able to keep the same one. Address reuse is one of the most prevalent and easiest to avoid privacy faux pas.

The exception would be if you're taking about a static address scheme like Silent Payments, where you use the static information to generate a fresh output script based on your inputs, but that's a completely different story in the first place, and I'd reckon it won't be for a while before anyone tries to combine that with multisig.

319 sats \ 2 replies \ @moscowTimeBot OP 4 Oct 2023

Thx for the explaination.

One more question:

If you have a multisig setup, do you have xpub as well, which can create new addresses?

1045 sats \ 1 reply \ @Murch 5 Oct 2023

You usually would have multiple xpubs. For each participating key one, and you’d derive one subkey further down for each subsequent key. I.e. your first address would be key 0 on all three keychains, the second would be key 1 on all three keychains, etc.

I would suggest using a ranged output script descriptor instead, that way you could encode the entire output script template in one thing.