Yes, fake pubkeys increase the utxo set, which means that other than taking space in the blockchain, they also increase the space needed in the utxo DB. Not in RAM, as they will be pretty soon placed in disk cache never to be fetched again, since they cannot be spent, but still. Storing the same amount of data in fake pubkeys costs (depending on feerate) up to 30 times (if you consider 1sat/vB as feerate floor, even more for lower feerates) the cost of OP_RETURN, and way more than that compared to witness if you consider the 75% discount. So yes, fake pubkeys increase the space requirement for full nodes, but are much more expensive than that for spammers. Game theory says that if you do not give the possibility to store garbage in huge multiple OP_RETURNs, and maybe patch the possibility to do that in witness, spammers will have a much harder time spamming.

Replies (1)

Default avatar
ihsotas 3 months ago
It’s not about easier or harder. Spammers don’t care how hard it is to put stupid moneys or dogs or whatever the fuck on chain. They don’t even care how much shit costs. Higher Expense just means rarity to their twisted little brains. 1. It’s not very hard to do this to begin with. block 273,536 contains a 14.4kb image of Nelson Mandela embedded across multiple transactions using pay to fake public keys. This has no limit. 2. Limiting op_return just pushes more bloat to the utxo and is objectively worse for the chain in the long run. Bitcoin is a bad data storage system but it is designed is such a way that you can’t stop people from doing stupid shit. Making it possible to do stupid shit in a less harmful way seems better than making it only possible to do maximal harm regardless of your intentions. Wishing for people to be perfect angels is not how bitcoin operates.