This week’s newsletter summarizes the final week of the organized taproot review, describes a discussion about coinjoin mixing without either equal value inputs or outputs, and mentions a proposal to encode output script descriptors in end-user interfaces. Also included is our regular section about notable changes to popular Bitcoin infrastructure projects.
Bitcoin Optech is planning to hold a third Schnorr and Taproot seminar workshop in London on February 5th 2020. This will cover the same material as the previous two Schnorr/Taproot workshops, which is available on this website for home study.
Member companies who would like to send engineers to the workshop should email Optech.
● Final week of organized taproot review: December 17th was the final scheduled meeting of the taproot review group. Pieter Wuille posted the slides from a presentation he gave summarizing progress, including text indicating he thought the proposal was “nearly ready”. Wuille also proposed a minor change to the tapleaf versioning. Also briefly mentioned was the discussion started by ZmnSCPxj on the Lightning-Dev mailing list about how precisely taproot could be integrated with LN to provide improved privacy and scalability.
● Coinjoins without equal value inputs or outputs: Adam Ficsor (nopara73) started a discussion on the Bitcoin-Dev mailing list about two previously-published papers (1, 2) describing coinjoins that didn’t use either equal-value inputs or outputs. Previous attempts at non-equal mixes were easy to compromise, but if an improved method was found, it could significantly improve the privacy of coinjoins by making their transactions look like payment batching. This seemed especially relevant after reports that a popular exchange was investigating users participating in the chaumian-style coinjoins created by Wasabi Wallet. Although several ideas were discussed, we think the summary by Lucas Ontivero captures the essence of the overall conclusion: “In summary, unequal inputs/outputs coinjoins knapsack is the best we have at the moment ([but] it is not as effective as equal-outputs transactions).”
● Encoded descriptors: Chris Belcher asked for feedback from the Bitcoin-Dev mailing list about base64 encoding output script descriptors so that they’re easier to copy and paste (and also so that regular users aren’t exposed to their code-like syntax). At least one reply was opposed to the idea and other replies which did support the idea each proposed a different encoding format. The discussion to date did not come to any clear conclusion.
Notable code and documentation changes
● C-Lightning #3351 extends the
invoiceRPC with a new
exposeprivatechannelsparameter that allows the user to request the addition of route hints for private channels to a generated BOLT11 invoice. The user may optionally specify which channels they want to advertise in the invoice, including both public and private channels.
● LND #3814 allows the UTXO sweeper to add wallet inputs to a sweep transaction in order to ensure its output meets the dust limit. This is designed to help support the proposed anchor outputs feature (see Newsletter #70) which will need to add inputs to a transaction in order to be able to spend low-value UTXOs.