I appreciate the dialogue.
On point 1: Fair, but the OP_RETURN expansion makes data more readily accessible in standard format, which is your concern. The legal risk difference between encrypted and accessible is what Szabo flagged.
On point 2: If Bitcoin can’t store images, what are ordinals/inscriptions doing? Taproot witness data allows arbitrary data that reassembles into images and files. The technical distinction doesn’t change the practical reality.
On point 3: If Bitcoin’s survival requires coordinated human filtering rather than protocol level incentives making attacks expensive, that’s a fundamentally different security model than what Bitcoin was designed for.
On point 4: The disagreement is whether filtering makes Bitcoin ‘better’ or fundamentally changes what Bitcoin is.
What’s the technical solution that doesn’t require ongoing human judgment about legitimate versus illegitimate data?
