# Descriptor serialization standard in PSBT --- # My bias - Bitcoin (only) - Compact encoding size - Simple specification --- # Problem No widespread standard for exchanging wallet setups. - BlueWallet format is textual and inflexble - [BCR-2020-010](https://github.com/BlockchainCommons/Research/blob/master/papers/bcr-2020-010-output-desc.md) is compact, but inflexible and deprecated - [Envelopes](https://github.com/BlockchainCommons/Research/blob/master/papers/bcr-2023-007-envelope-output-desc.md) is Blockchain Common's proposed standard based on CBOR - [COD](https://github.com/BlockchainCommons/Research/issues/135) is our proposal based on PSBT. --- # CBOR vs PSBT - [CBOR](https://www.rfc-editor.org/rfc/rfc8949.html) is a more complex specification than [PSBT](https://github.com/bitcoin/bips/blob/master/bip-0174.mediawiki) - PSBT is also a standard (BIP vs IETF) - PSBT is already required for every signing software - Metadata and binary xpubs for free - ~300 lines for a [from-scratch codec](https://github.com/seedhammer/bip-serialized-descriptors) for PSBT+COD --- # Problem of process - Why wasn't `crypto-output` more widespread? - many readers, but few writers (Sparrow, SeedHammer) - Why wasn't `crypto-output` BIP'ed? - What's more likely to be BIP'ed, a CBOR or PSBT standard? --- # Discussion What makes a succesful standard? - Simpler is more likely to be accepted - Learn from `crypto-output`/`BCR-2020-010` - Propose as BIP - Widespread buy-in from wallet developers - "What would make you switch?"