Oddbean

With respect, while I don’t think the 80 byte OP_RETURN thing should actually be filtered because it’s a lot point by now, there IS a critically important difference between filtering and censoring in this context: • One is a question of WHAT is allowed in the bitcoin chain, which is a universal policy. Every node does this and this is similar to arguments with RBF and the like. In a sense, this is the only thing bitcoin does, is filter with extreme prejudice, WHAT goes into the chain. — there is an argument to be had on whether the byte issue is good/bad, but it’s not censoring privacy transactions or coinjoins. It’s a filter that *happens* to catch one kind. • The other is a question of WHO gets into the blockchain. F2Pool here had the audacity to claim that a certain address, with a certain balance, is owned by some “evil” people because some govt body, without trial or due process, has declared their evil acts and demanded punishment & eviction from market activities. Regardless of whether this particular kind of filtering should be considered good, or that it potentially sets a bad precedent, they are not the same thing in terms of the danger and subjectivity of the decisions. Deciding WHAT goes into the chain is a process of defining the bitcoin system, deciding WHO can get into the chain is censorship and violating the basic tenant of neutrality. The phrase isn’t “Bitcoin is for everything,” it’s “Bitcoin is for everyone. nostr:note14hm0a4xz7v5dmsmslj6h694vmnlytc59k4mv2s254clm3zdcpzmsujjv3x

Respectfully disagree. If the end state is the same(preventing coinjoins), it feels like a distracting semantics hamster wheel to even debate filter vs censor. At best you’re totally right and it stops here. At worst this drives social consensus towards this becoming censoring by your state definition, and a startling number of people already seem on board with this. We are unfortunately in a world where waking up and Ocean, F2Pool, and Foundry all “filtering” coinjoins is a very real possibility. In this world I think an immediate and aggressive social rejection of “filtering” is a must. We are indisputably at least on the slippery slope. If we don’t charge back up, we might just keep sliding down forever.

No one has the goal of filtering coinjoins. Coinjoins are still allowed by Ocean and they aren’t taking any stance against privacy tools at all. It is only Samourai who got caught because of a particular way they use the OP_RETURN data.

All other coinjoin implementation besides whirlpool are not mathematically deterministic and are currently being demixxed by chain analysis the use of op return allows for TxO which ensures fees are paid, acts as a sybil attack defense and creates the unified values it isn’t buggy, broken or in need of a fix — people who have never coded a successful coinjoin sure seem to have a lot to say about how they should work

Where are you getting the idea that I’ve made any of these arguments. Please stop arguing with me over a conversation you had with someone else. For one, I don’t know much about the methodology and made no claim to whether it was good/bad or better/worse than anything else. I said the fact that Samourai coinjoins got filtered wasn’t because they were trying to filter coinjoins. Which based one everything we know, is perfectly true. If their method is better, then that’s cool 👍🏻 As I prefaced all of this, I think it’s a bad way to go about this filtering and the OP_RETURN issue is moot. Luke is just especially stubborn, imo. On the issue of JPEGS, I think you could safely filter OP_FALSE txs which are necessary to make all the JPEG info specifically meaningless, but still be able to include it. Which, as I understand it, means it can be filtered safely, knowing that none of the Txs with that data mean anything anyway. This filters JPEGS for clients that want to, and shouldn’t drag any privacy tools with it.

It seems as if both sides have been doing a shit job of communicating reality here. I was glad to see Luke put out a more comprehensive post yesterday - I would love to see more from him explaining his position. The original PR is actually quite interesting including Luke presenting an idea to randomize the byte limit so miners are forced to choose one. https://github.com/bitcoin/bitcoin/pull/5286#issuecomment-63180444

Seems like many of those voicing their opinions have had their judgement clouded by their contempt for Luke. He says one thing, so they defacto defend the opposite. Thanks for staying objective, @TheGuySwann, and for the rational take.

I reserve the right to be wrong but I hear the same people who decried ordinals as stupid and wrong now crying because some have opted to not include or process ordinals. We understand that this does not stop ordinal overall, but prevents these people from interacting with ordinal. Isn't this the core tenant in bitcoin and freedom thinking, that one interact with those who they see as bringing value and not with those who do not bring value? If in the end enough people see a value in ordinals they will interact and include them and if noone finds value they will be forgotten and ignored. This could extend to the future to any sort of transaction. This is an opportunity. If a group or pool decides to not include any given transaction, it falls to someone who does see a value to pick up the slack. I see no problem with this. In theory one could choose to include only those they chose to whitefish and the blockchain may splinter. Those making the decision should do so in an informed way, being aware they would lose and descendent/dependent trancactions that follow, essentially stranding those utxo's . Then as before, the chain found to be most value will be adopted and the lesser chain languishes. Bitcoin is for enemies, we don't have to agree with anyone,but by the same token, we can choose to make an informed decision to not support them.