Signing protocol about corefunds HOT 7 CLOSED

I'm personally okay with that. Will work on tooling immediately after the holidays.

from corefunds.

I'll update or (more likely) rewrite the tool. Nowadays we have PSBT as a standard so it's not hard to do any of this anymore - tooling can be relatively straight forward.

The tools I wrote in 2015 were mostly complex because the tip jar got spammed and we wanted to consolidate utxo. Shouldn't be a problem with a fixed fund.

from corefunds.

I think the sources of the tools and toolchains we use to build any transactions must be open source and public. I'm comfortable if the policy states that but doesn't mandate any specific tool or toolchain.

from corefunds.

The 1.21 approach is PSBT; shall we go with that? The tool to generate the transaction (which is really just collating UTXO, shuffling and estimating fees) should be open source once sign-offs are done. If 1.21's dev branch works for signing (should do), we should be able to progress with that.

from corefunds.

To clarify; the actual assembling the PSBT happens in the node, the tool just finds enough UTXO and avoids manual handling. Actually occurs there's no reason the input JSON can't be pulled out and provided so others can verify it.

from corefunds.

I don't understand the problem. If you're asking if it is ok to exchange PSBT instead of network encoding, then I think it's fine. I'll work on a small tool that converts to-and-from these formats in case someone wants to use production dogecoin-tx for example to inspect a proposal.

Enforcing tools through policy just is a liability because what if the tool isn't maintained? Every custodian just validates the transaction, how they do it is up to them. If there is a challenge I don't mind helping anyone out with the tooling they need.

from corefunds.

The core of the question is whether we're okay to use PSBT as the exchange format, yes, and then custodians can choose their tool of choice as long as it meets that.

from corefunds.