The all data is 1s and 0s argument is flawed because it implies that all data is equal. Whilst this is true from a technical standpoint it is not true from any other standpoint (moral, legal, social, etc). Just like you can see letters on a page, there is a correct and intended way to interpret those letters, and whilst you can technically interpret them differently, there is a correct meaning that everyone else will see them as.
For argument’s sake, let’s just say that you’re correct in that there is no legal problem for node runners if cp starts getting relayed around, many people are still going to have a problem contributing to the relaying and storing of this cp from a moral point of view, meaning fewer people will run nodes. There is no reality where cp has a positive impact on bitcoin, only negative. Which means any change that widens the avenue / facilitation of cp should not be undertaken.
I would argue that bitcoin could potentially be the best cp distribution protocol in the world. Anonymous upload of arbitrary data, to thousands of people all around the world in seconds without having to pay anything because this distribution can happen only at the mempool level and does not have to get mined. When 99%+ of nodes filtered out these large op_returns the p2p network was not efficient at distributing large data files because 999 in 1000 nodes were not going to relay it. As less and less nodes filter, suddenly the p2p network becomes much more effective at relaying around these large data blobs.
“The fact that you need a special software to piece together the cp vs just an image program is irrelevant.”
This is relevant because in the case of inscriptions etc, bitcoin clearly interprets them as native bitcoin data and is why there has been large increase in utxos recently. Whereas op_return’s only purpose is as a place to store arbitrary data, meaning any data put there bitcoin sees as the correct way to interpret that data.
