skip to content

Research • August 14, 2026

The Bitcoin Fork That Lasted Two Blocks

The soft fork that was supposed to save Bitcoin from “spam" lasted two blocks.

This article originally appeared in Galaxy Research's weekly newsletter. Subscribe to get timely insights delivered to your inbox every Friday morning.

The soft fork that was supposed to save Bitcoin from “spam" lasted two blocks.

BIP-110, formally the Reduced Data Temporary Softfork, was the brainchild of pseudonymous developer Dathon Ohm with input from Luke Dashjr, creator of the Bitcoin Knots node software, former chairman and CTO of the Ocean mining pool, and a polarizing figure in the Bitcoin community. It called for a one-year consensus-level restriction on arbitrary data in Bitcoin transactions. This includes Ordinals inscriptions (Bitcoin’s answer to NFTs), BRC-20 and Runes (both fungible token standards), and oversized OP_RETURN. The proposal contained seven rules capping most outputs at 34 bytes, OP_RETURN at 83, data pushes at 256, plus restrictions on Taproot annexes and control blocks.

The activation design targeted 55% miner signaling (1,109 of 2,016 blocks). By comparison, Bitcoin’s SegWit upgrade in 2017 wanted 95% and Taproot in 2021 wanted 90%. If 55% wasn’t reached, a mandatory signaling window opened at block 961,632, where enforcing nodes would reject any block that didn’t signal, regardless of hashrate.

Miners never came close. In the last difficulty period before the window opened, 51 of 2,016 blocks signaled. That is 2.53% against a threshold its own authors had already cut nearly in half from the standard. According to block data, every one of those 51 blocks came from Ocean.

Block 961,632 landed on Aug. 8 at 19:35:55 UTC. AntPool mined a non-signaling block that the network accepted, and BIP-110 nodes rejected it. Roughnecks, mining through Ocean’s DATUM system, produced the competing block.

The minority chain reached 961,633 and stopped. Roughnecks quit the next day and told other miners to do the same, then reversed on Aug. 10 and resumed, saying it would keep going until a “sensible POW change” arrived. The main Bitcoin chain is now several hundred blocks ahead. Strategy Executive Chairman Michael Saylor commented on the pitiful fork attempt, saying “Bitcoin worked exactly as designed” (by remaining unaffected in the face of the ill-fated fork attempt).

Ocean was the only pool signaling in favor of BIP-110. Days after the split, it disclosed that a Stratum configuration error had routed some of its miners onto the BIP-110 chain (rather than the real Bitcoin chain) for roughly 18 hours. The pool restored its default endpoint and committed to roughly 0.3 BTC in direct rebates within 72 hours. Cryptographer and Bitcoin OG Adam Back suggested the losses come out of Dashjr’s salary. Ocean’s reported hashrate has since fallen 96%, and miners have publicly called for leadership changes.

On Aug. 9, Bitcoin Improvement Proposal editor Mark “Murch” Erhardt filed a motion to the Bitcoin development mailing list recommending that fellow editor Luke Dashjr be removed as an editor. Erhardt cited Dashjr’s conflict of interest, minimal contribution, championing a contentious fork, and a coordination breakdown with the other editors.

The next day, it was done. Jon Atack merged the pull request (a one-line deletion) and longtime Bitcoin Core contributor Bryan Bishop confirmed the repository permissions were revoked. Bishop, Atack, Erhardt, Olaoluwa Osuntokun, and Ruben Somsen remain as BIP editors. Dashjr also announced a sabbatical from Ocean. He continues to refer to the majority chain as “Bpedo.” (One of his arguments for limiting arbitrary data was the alleged risk of child sexual abuse material being published onchain.)

Dathon Ohm’s response to the 2.53% tally was to accuse the large pools of collusion and announce work on a proof-of-work change to “fire the miners.” Dashjr backed it, citing ASICBoost when pressed on what was wrong with SHA-256d, the hash function at the core of Bitcoin. (ASICBoost is a technique for speeding up mining.)

OUR TAKE

Ocean’s entire pitch was miner sovereignty. DATUM exists so operators build their own block templates locally instead of trusting pool operators’ block construction. That is a useful product solving a real centralization problem. Then a config error pointed unassigned hashrate at the chairman's fork for 18 hours without operator consent, and the pool ended up writing rebate checks. A movement that spent two years arguing pools are dangerous intermediaries proved its thesis at its own expense.

The soft fork itself was never a “neutral anti-spam patch” as its proponents claimed. It was a consensus-level attempt to invalidate fee-paying transactions a minority of the Bitcoin community disliked. On top of that, the threshold for acceptance was cut from 95% to 55% and a mandatory window was designed to split the network if even that low bar wasn’t cleared. Lowering the threshold manufactured a quixotic attempt at a chain that needs (by Saylor’s math) 25 years to reach its first difficulty adjustment.

BIP 148 (whose activation in 2017 is celebrated as Bitcoin Independence Day) worked because exchanges, wallets, businesses, and holders were already behind it. Miners capitulated to an economic majority that had moved without them. A UASF (user-activated soft fork) is a mechanism for the economic majority to route around miners. It is not a mechanism for ~2.5% to route around everyone.

The PoW pivot makes this all worse for Ocean. Miners rejected BIP-110, and the response was to try to change the electorate. And ASIC resistance may not even be the point. Per one of the Knots strategy channels, the stated logic was to adopt an altcoin’s algorithm so ASICs would exist for the new chain on day one. Basically, cannibalize the hardware of an asset Bitcoiners would call valueless (they use an indelicate term for these; it rhymes with “bitcoin”). Bitcoin Gold ran this experiment in 2017 and nobody bothers citing it as a cautionary tale anymore.

LukeJr Discord screenshot

Luke Dashjr laying out proof-of-work design criteria in the Bitcoin Knots Discord channel. Source: Bitcoin.com News

What to watch is whether Ocean’s 96% drawdown proves permanent. The pool building the most credible answer to Bitcoin mining centralization bet itself on a fork its chairman championed, and lost.

You are leaving Galaxy.com

You are leaving the Galaxy website and being directed to an external third-party website that we think might be of interest to you. Third-party websites are not under the control of Galaxy, and Galaxy is not responsible for the accuracy or completeness of the contents or the proper operation of any linked site. Please note the security and privacy policies on third-party websites differ from Galaxy policies, please read third-party privacy and security policies closely. If you do not wish to continue to the third-party site, click “Cancel”. The inclusion of any linked website does not imply Galaxy’s endorsement or adoption of the statements therein and is only provided for your convenience.