The Bitcoin greybeards are finally entering the filter argument, a subject that is top of mind after leaked texts from Luke Dashjr suggest that one of Bitcoin’s most prominent developers wants to up his filter game.
While these developers are on the Bitcoin Mount Rushmore, they’re not really known as talking heads.
You won’t catch them making youtube videos or debating endlessly on Twitter spaces. They tend to prefer longer, forum-style communication such as the Bitcoin Developer Mailing List.
Another reminder that most of the smart people aren’t on Twitter.
I did all the trudging through the mailing list discussion to see what the smartest Bitcoin developers are saying about filters so you don’t have to. Here’s what they’re saying.
Greg Maxwell on Bitcoin filters: “The purpose of the mempool is to model what will get mined”
Gregory Maxwell (“Gmax”) is one of the most legendary Bitcoin developers of all time. The co-founder of Blockstream is responsible for coinjoin and confidential transactions as well as author of the original taproot proposal. Greg has not considered himself a Bitcoin Core developer for at least 8 years, but he still commands significant soft power over developer mindshare.
Greg argues that, when nodes’ mempools look too differently from one another (in this case, because some nodes are filtering transactions that others are not), this is harmful to the overall system.
“Significant discrepancies are harmful to the system and promote centralization and fail to achieve a useful purpose in any case. What marginal benefits might be provided do not justify building and deploying the technological infrastructure for massive censorship.”
Andrew Poelstra: Bitcoin filters beyond fees are “contrary to [the] purpose of the mempoool”
Director of Research at Blockstream and influential in the development of Taproot/Schnorr and Miniscript, Andrew Poelstra is known for rigorous formal approaches to Bitcoin and privacy primitives.
Poelstra emphasizes that there’s little functional difference between relaying a transaction and relaying a block, meaning that if you want to not relay certain transactions, you also would not want to relay blocks with those transactions in them.
By Bitcoin’s nature, you can’t really pick and choose whether you like certain transactions or not:
“Once a transaction is in a block, you need to relay the transaction if you want to relay a block. You cannot pick and choose which parts of a block you like and which parts are ‘abusive’. This is what it means for something to be a consensus system.”
Poelstra also describes the purpose of the mempool similarly to Gmax:
“The purpose of the mempool is to approximate the contents of blocks, both to help individual node operators (who would otherwise get large quantities of “surprise transactions” with every block) and to help the network (which would otherwise have poor propagation properties).
Any sort of filtering beyond that done by miners is contrary to this purpose of the mempool. This is a technical fact. It has nothing to do with “bitcoin’s ethos”, except its ethos as a consensus system, which directly contradicts your point.” (Emphasis added).
Nick Szabo: Cautious concern, but not filter happy
Designer of Bitcoin predecessor Bit Gold and the inventor of smart contracts, Nick Szbo has recently returned to Twitter after a long hiatus to weigh in on Luke Dashjr’s early stage idea to filter media out of Bitcoin transactions after-the-fact.
Szabo says he’s currently “exploring the issues,” and both camps are interpreting his exploratory statements as supporting their side.
A couple of Szabo’s tweets suggest he’s also concerned about legal issues for noderunners if obscene data goes onto Bitcoin:
But Szabo also points out that because OP_RETURN is prunable, it could actually reduce legal risks:
It seems like Szabo is just exploring this publicly for now and shouldn’t be placed in a filter/antifilter camp.
Adam Back: Anti-filter and anti-spam for Bitcoin
Inventor of Hashcash Blockstream CEO Adam Back finally stumbled into the filter debate about a month ago.
Back has recently become quite anti-filter – but also anti-spam. His argument is that Bitcoin’s fee market reduces the harm of spam.
He sums up the social and technical side of the debate in a reply to Szabo (grammatical errors kept):
“I’d argue it doesn’t matter much whether op return is 80, or 160 or 100kb, won’t affect what can be relayed and mined as the current network already will relay and mine. Setting it to 80 or 160 is clearly less socially contentious but it’s not about logic, computer science, economcs at all. People who want to filter to influence spam, know it doesn’t work technically, nor economically also, they just want to set a social message it’s not welcome.”
Most renowned Bitcoin developers think filters are worse than spam
The Bitcoin greybeards aren’t terribly concerned about spam and think filters are short sighted and antithetical to a healthy system.
They’re also less prone to ad-hominem and don’t rely on rhetoric to communicate their points.
It all hearkens back to a Bitcoin culture of old. We’ve been arguing about spam since 2010, but the discourse was much more measured and technical at the time.
Even Hal Finney chimed in pointing out Bitcoin’s permissionlessness:
“I understand that this is arguably not a good use of the Bitcoin distributed database, but nothing stops people from doing this so we should be aware that it may be done.”
