A fresh round of sparring between Bitcoin Core and Bitcoin Knots over “arbitrary data” and policy defaults is ricocheting across X, but the argument’s bones are older than many remember. As Bitcoin developer Peter Todd put it on Sunday, “Good read. tl;dr: everything that has been said about Core vs Knots has already been said almost 15 years ago.” The 2010 Fight Over Bitcoin’s Soul That Never Ended The historical through-line runs straight back to December 2010, when Satoshi Nakamoto shipped Bitcoin version 0.3.18. That release quietly introduced an “IsStandard()” relay and mining policy to “only include known transaction types,” a defensive move designed to reduce attack surface from exotic scripts. Satoshi’s own release note summarized the change tersely: “IsStandard() check to only include known transaction types in blocks.” The first debate about arbitrary data in the blockchain happened in December 2010 and Satoshi was involved On 8th December 2010, Satoshi released Bitcoin version 0.3.18, which included a standardness check, to only include known transaction types pic.twitter.com/J95ax5Cgte — BitMEX Research (@BitMEXResearch) September 29, 2025 The check ignited what many participants described as Bitcoin’s first real governance dispute. Within hours, forum users split over whether restricting non-standard transactions would neuter legitimate experiments like BitDNS or simply protect the young network. The thread, preserved by the Satoshi Nakamoto Institute, captures the core fault lines that have resurfaced in 2025. On the permissive side, user “da2ce7” argued that fees would rationalize everything: “Transaction fees will pay for the generation of the chain in the future… if [others] want to include carefully crafted transactions… they must include the appropriate compensation.” Jeff Garzik fired back that such a stance “will disadvantage people who use bitcoins… as cash as intended,” because non-currency uses would bid up fees and crowd out payments. Theymos, then pushing for minimal relay restrictions, argued miners’ incentives would bulldoze any client-level gatekeeping: “all miners have an interest in including any and all fee-carrying transactions… The restriction on relaying these transactions should be removed, at the very least.” Garzik warned that if “data spam increases TX fees to annoying levels,” currency users would decamp—and that the presence of “law-enforcement-objectionable data” would raise different, sharper risks. Crucially, Satoshi and Gavin Andresen converged on the whitelist approach as a pragmatic security default, while leaving the door ajar for purpose-built data uses. Gavin explained that whitelisting known-safe templates was “the right thing to do,” drawing an analogy to web security’s failure modes when blacklisting is relied upon.In a follow-up, Satoshi wrote: “I came to agree with Gavin about whitelisting when I realized how quickly new transaction types can be added,” and endorsed a path for small data commitments: “I also support a third transaction type for timestamp hash sized arbitrary data.” If today’s back-and-forth feels like déjà vu, BitMEX Research’s weekend recap is the missing Rosetta stone. Their thread traces the debate’s timeline—RHorning’s early pushback against 0.3.18’s new standardness rules; Theymos’s insistence that miner incentives would trump relay defaults; Garzik’s resistance to “non-currency data” pricing out money use; and community unease about what happens when immutable ledgers meet illegal content. The researchers note that Theymos even released a patch client removing restrictions at the time, underscoring how client defaults and miner policy have always been a contested, malleable layer. There are two enduring takeaways from the 2010 record. First, the “policy vs protocol” distinction—what Bitcoin can do versus what the reference implementation should relay or mine by default—has long been a pressure valve for innovation and a magnet for controversy. Satoshi’s 0.3.18 email makes plain that IsStandard() lived in this gray zone of incentives and norms, not consensus rules. Second, nearly every argument now deployed in Core-versus-Knots skirmishes had an ancestor in that first “coming-of-age” fight: fee-market neutrality versus application-layer bloat; the right to pay for block space versus the social cost of permanent data; and whether tightening defaults protects Bitcoin’s monetary function or stifles its utility for timestamping and proofs. The archive shows the spectrum clearly, from Theymos’s “remove the restrictions” stance to Garzik’s warning that generalized data “has the distinct probability of degrading service for digital cash.” At press time, BTC traded at $113,071.
