Twelve treasury requests paid and seven refused, in the voters' own words

Twelve paid, seven refused

The treasury paid twelve withdrawals across these five epochs, ₳44.2M together, and seven others reached the end of their voting periods with nothing. A twentieth was ratified inside the window for payment after it. A withdrawal needs 67% of DRep power, counted with abstaining power left out and power that cast no ballot as no. The lowest share behind a payment was 68.4%, the highest behind a refusal 63.0%.

More than half the money went into a single withdrawal, ₳25.4M to Intersect at 75.7%, for a year of its operations from June 2026 to June 2027: the member organisation, technical stewardship of the core repositories, incident response and release coordination, at a dollar figure its text puts below the previous year’s ask. Most of the rest is upkeep. The highest share of the twelve, 79.3%, went to the Mithril certificate scheme that lets a node bootstrap without replaying the whole chain, the lowest, 68.4%, to a lightweight data node.

What the voters wrote about the seven

The largest refusal was Cardano Builder DAO, ₳20M at 11.0% for a fund that would allocate money to builders against usage targets. SASA_nagamaru_ながまる voted no because, it wrote, the treasury should stay directly accountable to the DReps rather than hand funding decisions to another layer. The committee members who voted no wrote that no definitive administrator and custodian was named. Alchemy asked ₳10M and closed on 5.3%. The Cardano Foundation wrote that its no did not evaluate the proposal: the proposers had announced they no longer wished to proceed.

Strike Finance Liquidity Deployment asked ₳9M as treasury capital rather than a grant and had the tightest head count of the window, 65 yes against 64 no, at 20.5% of power. The Eternl DRep Committee abstained, writing that the room left under the spending limit “needs to be spend on voted Intersect budget items”. The Cardano Foundation voted no because liquidity of that scale should go through an ecosystem wide framework rather than one protocol. Blockfrost’s transformation to not-for-profit asked ₳9.8M to turn the service into a community owned not-for-profit over eighteen months and reached 43.3%. The Eternl DRep Committee voted no with a question: how a monthly cost of 13,600 dollars per employee is justified, and what the service earns today.

The smallest of the seven was Reimburse Ikigai Info Governance Action Deposit., ₳103,000 to reimburse a ₳100,000 deposit lost in 2024 to a node bug on one of the first info actions, plus ₳3,000 for the staking rewards forgone. At 63.0% it came nearest to passing. YUTA voted yes, writing that the loss “may not be a protocol bug, but it was at least due to poor design”. The Eternl DRep Committee voted yes because many DReps had already voted yes, “so we assume it’s legit”. A request for ₳4.0M for a payment card drawing on balances held on chain reached 54.4%.

How Yoroi and EMURGO voted

Yoroi W₳llet at ₳575.3M and EMURGO at ₳299.1M, the largest and the fourth largest DReps at the end of the window, cast identical ballots on all nineteen: yes on eleven, among them ten of the twelve that were paid, and abstain on the other eight, among them six of the seven that ran out. Each abstention came with the same short rationale. Yoroi wrote that it was withholding its vote “in light of the current situation” and would reassess “once conditions allow for a considered decision”. Blockdaemon at ₳334.4M, the third largest, abstained on every one of the nineteen with no rationale in the record.

Neither rationale says what the situation is. In the epochs that followed, and recorded here after the fact, both companies named it in their own words: a security incident at SecondFi. On 3 August 2026 EMURGO announced on X, for itself and the Yoroi DRep, its resignation from the Intersect board and its intention to deregister as a DRep in due course, adding that users migrating their wallet would be undelegated from their DRep.

Abstaining power leaves the count, while delegated power that cast no ballot stays in it as no. On Strike Finance, ₳2.29B of DRep power abstained, ₳0.61B voted yes and ₳1.37B no. On Blockfrost, ₳1.78B abstained, ₳1.51B voted yes and ₳1.16B no. Those shares were set by the DReps who voted yes or no, together with the silent power counted against, and without Yoroi, Blockdaemon and EMURGO.

A lower minimum for the committee, and why

The minimum size of the Constitutional Committee fell from seven members to five at the start of epoch 643. Reduce the committeeMinSize parameter from 7 to 5 had been ratified one epoch earlier with 80.1% of DRep power behind it, on 137 yes votes, 22 no and 17 abstain, against the 75% it needed, and six committee members voted for it with none against. A single resignation earlier in the year had left six members where the protocol required seven, and everything needing a committee ruling stopped until the seat was filled.

YUTA wrote that only four candidates had run for the four seats up for election, “despite a candidacy period of more than one month and repeated appeals”, and that the committee sat at exactly its minimum. The Eternl DRep Committee wrote that it did not want single members to carry more weight at five, “on the other hand there are currently not many volunteers for this currently unpaid job”, and voted yes “to keep governance going”.

The hard fork passes the pool vote by 1.4 points

Hard Fork to Protocol Version 11 (‘van Rossem’ Hard Fork) was ratified at the boundary to epoch 643 and took effect one epoch later. It delivers the Plutus operations the cost model change of the previous window had already priced. DReps put 77.6% behind it against the 60% it needed. The pool side was the tight one: 383 pools voted yes, none voted no, 48 abstained, and the share was still 52.4% of stake against the 51% needed, because on a hard fork every pool that does not vote counts as no. The Eternl DRep Committee’s whole rationale was one line: the wallet is hard fork ready. Across the window the count of DReps that had voted in the last 12 epochs fell from 341 to 318, and the power delegated to DReps by ₳115.7M.

The request that did not fit under the limit

Cardano PRIME arrived in epoch 642 asking the community for ₳120M for a twelve month programme for decentralised finance on Cardano. The spending limit for epochs 613 to 713 stood at ₳350M, and by the end of epoch 646 withdrawals since epoch 613 had used ₳335.6M of it, leaving ₳14.4M. Its text makes the withdrawal conditional on a limit with enough room to cover it in full.

That limit was on the ballot at the same time. Net Change Limit, submitted in epoch 640 to replace the ₳350M cap with one of ₳500M, stood at 62.4% as of the close of epoch 646. Three withdrawals ran to the same boundary and stood, as of that close, at 51.8% for Scalus, 33.8% for Bifrost and 26.7% for Global Order Book. The review of epochs 647 to 649 takes all of them from there.