Damus
nostrich profile picture
nostrich
Why I'm for loosening the OP_RETURN limits ⏎

- In practice, OP_RETURN is already unconstrained: see Carman's OP_RETURN bot or Portland's Slipstream.

- Encoding arbitrary data via OP_RETURN is far more responsible than doing so in witness data, as it's 4x more costly.

- OP_RETURNs are prunable and do not pollute the UTXO set, making them vastly preferable to faux outputs that burden every node indefinitely.

- Mining non-standard transactions that bypass mempools undermines decentralized transaction propagation, disrupts fee estimation, and increases block propagation latency (especially when compact block construction fails due to missing transactions).

- This latency disproportionately impacts smaller miners and encourages direct submission to large, well-connected players, widening the profitability gap and increasing centralization.

- Layer 2 protocols are especially sensitive to fee miscalculations, making transparent and measurable fee behavior essential to their reliability.

While I share the distaste many have expressed toward tokens, inscriptions, and BitVM (when used as a casino), I care deeply about minimizing harm to bitcoin’s base layer. Allowing larger, standardized OP_RETURNs is a pragmatic step in that direction.
Jean Do · 41w
Couldn't the same result be achieved -- over time with an option for opting-out, by doubling the -datacarriersize default with each future release? It might leave the choice (at least the perception of choice) with the individual node runner. It might offer the chance to learn over time that it is...