I’ve never understood the bloat argument. If you have a block full of 4,000 monetary transactions, doesn’t that ‘bloat’ it just as much as 4,000 spam transactions? Is it an issue due to the amount of dust addresses spam creates? Also, uncapped OP return allows for decently big JPEGs to easily get on chain. Do other methods of embedding arbitrary data also allow for that size?

Replies (2)

JackTheMimic's avatar
JackTheMimic 3 months ago
The "set" of UTXOs would be larger if packed into pubkeys (multisigs where the pubkey outputs are the actual data not "real" keys) This puts more demand on your node from a validation and storage perspective. This means it's more expensive to run a full archival node. OP_RETUN data can be pruned as it has no effect on the chainstate.
jb55's avatar
jb55 _@jb55.com 3 months ago
no, op_return are provably unspendable, meaning you can prune them from the utxoset. If you store data in bare pubkey outputs (like the butcoin whitepaper): These are *permanent* and you have to store them forever, even on pruned nodes, since you can’t prove they are unspendable. This would be a growing fixed size burden for pruned nodes which is really bad. Thats why encouraging op_return for data is a good thing, even if it’s not the most economical compared to witness data.