bips icon indicating copy to clipboard operation
bips copied to clipboard

BIP45 modification or new BIP: Multiple Multisig Accounts

Open bijeebuss opened this issue 9 years ago • 12 comments

Add an account_index to the address derivation tree, following the standard in BIP44. This way multiple multisig accounts can be created under the same master key allowing users to restore multiple multisig wallets with one mnemonic phrase.

bijeebuss avatar Aug 28 '16 09:08 bijeebuss

@maraoz @matiu

luke-jr avatar Aug 28 '16 13:08 luke-jr

@bijeebuss I suspect this is too late, but pinging the authors for the decision.

luke-jr avatar Aug 28 '16 13:08 luke-jr

@luke-jr if its too late perhaps this could be a new BIP with a different purpose index

bijeebuss avatar Aug 31 '16 03:08 bijeebuss

@bijeebuss We specifically left out the account_index from BIP45, which is based on BIP43 only (not BIP44).

@matiu is the main maintainer of copay though, which spawned the creation of this BIP, so I'd defer to his judgement.

maraoz avatar Sep 08 '16 21:09 maraoz

@matiu is not available for a few week afaik. He also was talking about documenting BIP48 on his return.

btcdrak avatar Sep 08 '16 21:09 btcdrak

Ping @matiu

jonathancross avatar Apr 25 '17 17:04 jonathancross

Ping @matiu - still formal definition of m/48’ despite broad adoption by Ledger, Trezor, ColdCard, Electrum and others for use in multisig signing. So far everyone is reverse engineering how others are doing it. This has become arcane insider knowledge that is against the principles of open development.

I’m working with others to puzzle out one spec. Let me know best way to proceed one we’ve captured it. An update to a BIP, a new BIP, or submit to be a SLIP elsewhere.

ChristopherA avatar Apr 22 '20 03:04 ChristopherA

SLIPs should only be used for altcoin stuff.

If the author of the BIP doesn't respond, a new BIP seems in order I think.

luke-jr avatar Apr 22 '20 04:04 luke-jr

As far as I'm aware, Copay has abandoned use of BIP45, and uses BIP44 (with modification of root as m/48' for multisig hardware), and without the use of multiple accounts.

braydonf avatar Apr 22 '20 16:04 braydonf

I still think this is a good idea but it sounds like there is already a workaround out in the wild that is being used. It would be better to just formalize that if it covers this use case, especially since BIP45 has already been around for a while now.

bijeebuss avatar Apr 22 '20 17:04 bijeebuss

I believe why an account index was not included or used, was because multiple wallets (and seeds) were used instead. When using hardware devices this would then require multiple devices or restoring different seeds to the device for usage. That could be potentially undesirable with hardware usage.

With the addition of an account index, there needs to be an additional backup with public keys (and account indexes) for each new set of multisig participants. I don't think that has been documented yet.

BIP45 was likely abandoned in Copay because the intention of cosigner_index could be solved by other means that would not increase the address space that needed to be scanned. The lack of a coin_type was probably also a factor.

braydonf avatar Apr 22 '20 17:04 braydonf

@ChristopherA @bijeebuss I agree it would be useful to have a specification on how BIP44 (and with root m/48') is being used for multi-signature wallets.

braydonf avatar Apr 22 '20 17:04 braydonf

@bijeebuss We specifically left out the account_index from BIP45, which is based on BIP43 only (not BIP44).

@matiu is the main maintainer of copay though, which spawned the creation of this BIP, so I'd defer to his judgement.

It appears to me that one of the BIP authors indicated here that the suggested change is counter to the intention of this BIP’s authors. I’m closing this now, as it seems unlikely that this change will be incorporated into this BIP. Perhaps it would be better to propose a new separate BIP that incorporates the change, if you still think it is worth publishing.

Please let me know with a comment if any of you disagrees with this assessment, I’d be happy to reopen it if there is still interest in pursuing a merge.

murchandamus avatar Apr 23 '24 18:04 murchandamus