the long but hopefully honest as I can explanation as I see it so far: The real debate is less technical than it is philosophical in my opinion. To us who want the purpose of Bitcoin as a singular and focused monetary system, it seems like an explicit capitulation to "just put whatever you want in the blockchain, here's an official way to do it." While it likely wont make a significant difference from a practical standpoint, it feels like crossing an ideological line. There are tons of jpegs being put into the chain, and there are a few, low cost (mostly ineffective) measures that people have been putting into place for mempool policy to attempt to mitigate on their own nodes or block templates. However, IF those mitigations were the standard are done broadly across the node network (say core implemented them the way they are trying to implement this PR to remove the limit completely), then they actually would have some effect. Yes, its largely a cat and mouse game, but nonetheless it would do something, just like OP_RETURN being set at a strict limit has ensured that 99% of transactions have adhered to that limit. In addition, node runners who want the option to set this themselves, feel that prioritizing monetary transactions over jpegs is being a good steward. This PR started a pointless battle because it not only sets the default to infinity, but it removes the configuration option completely so a node runner cant change it. Those in support of removing the limit and configuration option, are saying that because no solution would be full proof, its a waste of time. There's a worthwhile addition that those who oppose have (imo fairly) noted that some in support have potential conflicts of interest working directly with taproot wizards, invested in NFT and VM tools, or other crypto adjacent persons or projects. I agree nothing is fully effective and don't believe that "everyone is malicious" here, but standards do matter, and more importantly, how we culturally treat what is a hack and exploit to input arbitrary nonsense into the chain matters. Cat and mouse or no, all of our focus and energy should be toward protecting and engineering Bitcoin to be the best and most secure money. And if we start embracing crypto culture, we will get crypto results. In other words, "a jack of all trades is a master of none." There are plenty of arguments on both sides that are being repeated that aren't accurate. These stand out, atm: - people who want to remove the limit saying inscriptions will use this instead, when they wont because its still more expensive in fees because inscriptions get the witness discount. - people opposed claiming this will make it harder to run nodes are wrong too, having explicitly unspendable UTXOs is technically superior than big, fake signatures, and the 4mb weight limit is the same regardless of what is happening. (plus, op_return falls under original 1mb still) My analogy defending my position against: -- Just because we can make money letting people rent out the courts for parties and concerts, doesn't mean we should. If the courts are exclusively designed and used for protecting people's rights and arbitrating conflict, then we lose a core foundation that "parties and concerts" are entirely downstream of and dependent on. Maybe people still manage to get parties in the courthouse, but how do we treat them culturally regardless of effectiveness at stopping them? - Put another way, sure I think drugs should be legal, but should we be handing out needles? My simplest steelman (hopefully accurate) of the support: -- If filtering is largely futile, then if we just open up the limit to a less costly (for node UTXO set not fees) way to use bitcoin to host arbitrary data, then at least people who want to be good stewards using NFTs or the like, have the option of not creating toxic UTXOs. And if we try to stop them, they are more likely to create private mempools that don't broadcast to the network but pay directly to a miner, which causes information discrepancies with the public network and fee predictions, as well as possible mining centralization concerns if this made out of band fees become popular.