Censorship of OP_RETURN Transactions

[kimdhamilton]

From @ChristopherA on June 30, 2017 6:51

It has been raised by @petertodd and others that using an op_return to encode the pointer to the DDO could hurt censorship resistance, as minors could identify those transactions that are DID related and not include them.

Instead, they recommend encoding it into the script of a P2SH transaction as some form of pay-to-contract. However, I still don't have a proposal from @petertodd on the best way to do this (he wants a similar capability to prevent censorship of open timestamps transactions). Another concern about the pay-to-contract approach is that it may mean two transactions to do a DDO update. An advantage is that it may support P2SH multisig DDOs.

For the first prototypes of the DID:BTCR method, I suggest we stick to P2PKH and P2WPKH transactions and we can support more censorship resistant methods later.

Copied from original issue: WebOfTrustInfo/btcr-hackathon-2017#3

[kimdhamilton]

From @petertodd on July 12, 2017 21:33

@apoelstra Submitted a pull-req to python-opentimestamps for pay-to-contract-hash that I'll be implementing sometime in august: opentimestamps/python-opentimestamps#14

For now I'd just stick with op-return, but assume something else will be added in the future. You can easily disambiguate this stuff by pre-committing to which method will be used in the same way you commit to the txout to be spent.