Our last issue (#16) closed under the heading “The floor rose, now comes allocation.” Van Rossem had lifted the protocol floor to 11, and on top of that floor a set of votes on who receives limited funds was heading toward a July 28 deadline. We wrote that far and left the outcome to the following week. That week is this one. The short version: one of the five cleared the deadline. The remaining four expired without reaching the threshold. In the same week, more than 32 million ada actually left the treasury. And immediately after the allocation question settled, a new one was submitted — this time not about who receives, but about what the people running the network keep: a proposal to lower the minimum fixed fee a stake pool earns from 170 ada to 75 ada.
Executive Summary
- Allocation settled: Of the five proposals we tracked last issue, only Se7en Labs’ Daedalus maintenance proposal passed, with 77.06% DRep approval against a 67% threshold. It stood at 54.72% when we published #16. The other four expired in epoch 646.
- Funds actually moved: In that same epoch 646, five withdrawals were enacted, totalling 32,877,129 ada. This was not merely a week of decisions — it was a week of disbursement.
- Next: the operator’s share. Intersect’s Parameter Committee submitted a proposal to reduce
minPoolCostfrom 170 ada to 75 ada (−55.9%) and to complete a two-step increase to Plutus memory limits. Ratification requires 67% of DRep stake plus 51% of SPO stake. - The bridge cleanup got a date: Wanchain notified the attacker that returning 90% of the stolen NIGHT by 12:00 UTC on August 6 would let them keep 10% as a white-hat bounty, with no civil claims pursued. NIGHT fell 10.4% on the week.
1. Market Pulse
| Asset | Jul 25 | Aug 1 | WoW |
|---|---|---|---|
| BTC | $64,099.00 | $62,820.09 | −2.0% |
| ETH | $1,861.26 | $1,860.59 | −0.0% |
| ADA | $0.16386 | $0.168048 | +2.6% |
| NIGHT | $0.020245 | $0.018148 | −10.4% |
BTC drifted lower, ETH barely moved, and ADA alone rose 2.6%. It has been a while since ADA gained in a week when the majors did not. That said, this was a narrow round trip inside the $0.16 range, not a change of direction. The week’s real movement was again in NIGHT, down 10.4% — following last week’s −27%, meaning the post-incident rebound did not hold. The gap between those two numbers mirrors the week’s content: on one side, procedure moved forward; on the other, the cleanup continued.
2. Allocation Settled — Four Expired, One Passed
The outcome
Epoch 646 began at 21:44:51 UTC on July 28. The five proposals we flagged last issue as “expiring July 28” were resolved there.
| Proposal | Outcome |
|---|---|
| Se7en Labs: Daedalus Wallet Maintenance and Improvements 2026-2027 | Ratified |
| Blockfrost’s transformation to not-for-profit | Expired |
| Cardano Enterprise Adoption: Ticketing Platform | Expired |
| Cardano Builder DAO | Expired |
| Alchemy by Sundial × Charms: Cardano-Native Bitcoin Treasury | Expired |
The Daedalus maintenance proposal was the one that passed, requesting 1,785,333 ada. DRep approval reached 77.06% (148 votes) against the 67% needed. Votes cast against stood at 3.41% (19 votes), with 14.41% not voting. The Constitutional Committee recorded 6 in favour, 0 against, 1 not voted.
What deserves attention is how that number moved. When we wrote #16 on July 25, we reported this proposal at 54.72% — more than twelve points short — and noted that “unless votes move substantially before the deadline, most will expire rather than pass.” In the event, this one proposal crossed 67% in the final days. The other four did not.
Not just decided — disbursed
In the same epoch 646, a separate set of five withdrawals was enacted.
| Proposal | ada |
|---|---|
| Intersect: Governance coordination and technical work | 25,400,000 |
| Mithril Protocol | 3,810,423 |
| Hardware Wallet Maintenance 2026 | 1,310,960 |
| Intersect Technical Steering Committee Support | 1,193,000 |
| MLabs Core Tool Maintenance & Enhancement | 1,162,746 |
| Total | 32,877,129 |
These went through the 2026 budget process, were ratified in epoch 645, and were enacted in 646. Ratification and enactment are distinct stages: a ratified proposal disburses at the following epoch boundary. This week is when that disbursement happened.
Placed side by side, the picture is clear. Items that came through the budget process are being disbursed on schedule, while items proposed individually outside it largely failed to reach the threshold. Proposals that secured a framework agreement first passed; proposals relying on individually gathered support did not reach 67%. That contrast is the honest reading of this week’s results.
Last issue we wrote that this “is not dysfunction — it is the threshold working as intended.” Four expirations support that view. But it is worth adding: a threshold working as intended does not mean the four rejected proposals were poor. Sixty-seven percent is a high bar, and good proposals fail it when support does not consolidate. Failing to be selected is not the same as being rejected.
The Daedalus withdrawal has not been disbursed yet. Ratification occurred in epoch 646; the funds are expected to move in epoch 647, which begins at 21:44:51 UTC on August 2.
3. The Next Question Is What the Operators Keep
From 170 ada to 75 ada
As the allocation question closed, a new parameter change was submitted in epoch 646. Intersect’s Parameter Committee bundled two independent changes into a single action.
minPoolCost: from 170,000,000 Lovelace (170 ada) to 75,000,000 Lovelace (75 ada), a reduction of roughly 55.9%- Plutus memory limits (part 2 of 2): per transaction from 16,500,000 to 17,500,000 units (+6.1%); per block from 72,000,000 to 77,500,000 units (+7.6%)
The second item completes a cumulative 25% increase begun earlier. The first is the one that will generate discussion.
minPoolCost is the floor on the fixed fee a stake pool receives each epoch. Every pool declares its own fixed fee, but cannot set it below this value. From a pool’s epoch rewards, the fixed fee is deducted first, then the operator’s margin, and the remainder is distributed to delegators. So the number is a cost from the delegator’s side and a size-independent guaranteed income from the operator’s side.
What changes at 75? Put simply, delegators to small pools keep more, and operators of small pools are guaranteed less. The fixed fee represents a larger share of total rewards the smaller the pool, so that is where the effect concentrates. At the same time, the fixed cost of running an additional pool falls, which changes the arithmetic for operators running many pools. How to weigh those two effects is what the coming weeks will argue about.
Our own position, stated plainly: this is not a proposal that resolves into “good” or “bad.” A high floor protects the viability of small operators while depressing delegator yields. A low floor does the reverse. Either choice imposes a cost on someone, which is why the question has been debated for so long. What is really being asked is not a number but a judgement about whose sustainability takes priority.
Why SPOs get a vote this time
This is the part that matters most from an SPO seat.
Intersect states the ratification requirement as 67% of active DRep voting stake plus 51% of active SPO voting stake. Parameter changes require SPO votes only in specific circumstances.
CIP-1694 defines a group of “security-relevant protocol parameters.” Any proposal touching that group requires an additional SPO vote, even for parameter groups SPOs do not normally vote on. That group is: maxBlockBodySize, maxTxSize, maxBlockHeaderSize, maxValueSize, maxBlockExecutionUnits, txFeePerByte, txFeeFixed, utxoCostPerByte, govActionDeposit, and minFeeRefScriptCostPerByte.
Read that list again: minPoolCost is not on it. What is on it is maxBlockExecutionUnits — the Plutus memory limit bundled into this same action.
So the structure is this. On its own, the parameter setting the floor on operator income would have been a proposal operators had no vote on. Because it was bundled with the block execution unit limit, the security-relevant rule engaged and 51% SPO consent entered the ratification requirement. Operators ended up holding a vote on their own compensation — as a consequence of what it was bundled with.
Whether that is deliberate design or coincidence is open to argument. The practical conclusion is not. On this proposal, SPOs do not merely have opinions; if they do not vote, they are not counted.
Voting has only just opened. As of 10:14 JST on August 1, DRep approval stood at 0.9% (9 votes), votes cast against at 0% (0 votes), and 95.8% had not voted. The action expires in epoch 653, so there is ample time for the debate.
4. The Oversight Seats Fill, and a Window Opens on the Constitution
The 2026 Constitutional Committee election, which we noted last issue was due to complete its audit by July 27 with on-chain reflection expected in epoch 646, proceeded as scheduled. An independent audit completed, and the four elected members are Philip DiSarro, Leandros BSP, Marek Mahut, and Cardano Curia. A governance action appointing them was submitted on-chain in epoch 646.
That action updates the four committee seats whose terms expire at the end of epoch 653, with new terms running to epoch 799 under the committeeMaxTermLength parameter. The action’s own expiry is also epoch 653 — the seats are designed to be filled before they fall vacant.
Alongside this, Intersect published a Constitutional Amendment Portal (cap.intersectmbo.org) in alpha. Using wallet authentication, community members can propose amendments, raise issues, and discuss constitutional changes. The Constitutional Committee reviews whether proposals conform to the constitution; the route to changing the constitution itself has until now been in a place few people could see. The process for choosing who reviews, and the process for revising the standard they review against, both moved forward in the same week.
5. Midnight Watch — The Bridge Cleanup Gets a Date
12:00 UTC, August 6
The Wanchain bridge outflow we covered last issue has a sequel. Wanchain notified the attacker that returning 90% of the stolen NIGHT by 12:00 UTC on August 6 would allow them to retain the remaining 10% as a white-hat bounty, with no civil claims pursued. An on-chain message from the exploit wallet, or a dedicated email address, was given as the channel.
The incident itself is as we described. On July 20, roughly 515.2 million NIGHT were drained. The cause lay in the bridge’s TreasuryCheck validator, which built the signed message by concatenating 14 variable-length fields without separators or length markers, leaving field boundaries ambiguous. A legitimate signature authorising roughly 3,110 NIGHT therefore also passed for a withdrawal of 203,001,692 NIGHT — a factor of about 65,000.
What changed this week is that the story acquired a date. For two weeks the position had been investigation and suspension; with a deadline attached, it is now a matter of waiting to see what happens. If August 6 passes without a response, Wanchain loses its reason to withhold civil claims.
Whether such offers get accepted is genuinely hard to predict. But putting the terms and the deadline in public has value on its own. When negotiations proceed privately, users learn nothing until a result arrives. With the conditions visible, at least it is clear what is being waited on.
The price did not recover
NIGHT fell from $0.020245 on July 25 to $0.018148 on August 1, down 10.4% on the week, following the prior week’s −27%. Last issue described a partial recovery from the post-incident low; that recovery did not hold.
Much of the drained supply has already been sold into the market, and this coincides with distributed tokens entering circulation in sequence. The opening of recovery negotiations could support the price, but it did not this week. Until the deadline passes, that factor stays suspended.
We wrote last issue that the security of a chain and the security of the routes in and out of it are separate problems. One thing is worth adding this week. Cleaning up after a broken bridge cannot be processed by the chain’s own procedures. Neither Cardano nor Midnight has a mechanism for recovering funds from an external bridge. So what remains is a deadline, an email address, and a negotiation. What you place outside the world that settles on-chain is something to decide before you cross.
6. The Name of the Next Floor, and the Design Beyond It
The info action naming the Protocol Version 12 hard fork “von Bergen” is still in voting. As of 10:15 JST on August 1, DRep approval stood at 23.88% (50 votes), votes cast against at 0% (0 votes), and 72.46% had not voted. It expires in epoch 651.
The name honours Fabian von Bergen (Zyroxa) who, per the action’s own text, has been a community member since the ITN, is the SPO of the Tempus Stake Pool, and has been a Cardano Ambassador since the earliest days of the programme. The name of the protocol floor becoming the name of someone who has run a pool is the kind of action worth recording from an operator’s seat.
The design beyond it is also coming into view. According to Intersect, the initial Dijkstra rollout will take place across several phases, introducing Nested Transactions, Linear Leios, and Peras, with the first two phases targeted to mainnet by the end of 2026.
There was also movement on how fees themselves are set. The review period for Input Output Research’s 2026 mid-year report draft, published July 16, closed on July 30. Per that report, the Fee Market Design specification has progressed to a draft CIP (CIP-1194), with simulations showing urgent transaction latency reduced by roughly 18%.
Fees are a cost to users, income to pool operators, and a congestion control mechanism for the network — three faces of the same number, where optimising for one alone always starves another. This week the fixed component, minPoolCost, went to a vote while the method for setting the variable component advanced as a design specification. The question of how a pool’s income is determined is being touched from both ends at once.
7. Risk Dimensions
| Area | This week | What to watch next |
|---|---|---|
| Treasury | Four expired, one ratified; 32.88M ada separately enacted | Whether the four return as revised proposals, and what changes |
| Pool economics | minPoolCost 170 → 75 ada in voting; 51% SPO requirement |
SPO turnout — heavy non-voting leaves the requirement unmet |
| Execution environment | Plutus memory limits entering step two | Measured effect of higher limits on block propagation |
| Oversight | CC appointment action submitted (expires epoch 653) | Whether reflection completes before terms lapse at 653 |
| Constitution | Amendment portal live in alpha | What amendments actually get proposed |
| Boundary (bridges) | Wanchain set a 12:00 UTC August 6 deadline | Whether there is a response, and the next step if not |
| Distribution (Midnight) | NIGHT −10.4% on the week; rebound did not hold | Pace of circulating supply growth and the recovery outcome |
| Protocol in practice | No confirmed implementations using PV11’s new builtins | Carried over from last issue |
8. What to Watch Next Week
- Vote movement on
minPoolCost— both DRep and SPO, and particularly how far SPOs move against a 51% requirement. - Disbursement of the Daedalus withdrawal — whether 1,785,333 ada actually moves in epoch 647 (from 21:44:51 UTC August 2).
- 12:00 UTC, August 6 — Wanchain’s deadline. A response, or a lapse.
- Resubmission of the four expired proposals — whether they return with revised terms.
- Constitutional Committee appointment — progress against the epoch 653 term expiry.
Closing
This week Cardano answered the question our last issue left open. Allocation settled, and one of five passed. In the same week 32.88 million ada left the treasury. The threshold worked as intended.
And as soon as the answer arrived, the next question was submitted. This time it is not about who receives, but about what the people running the network are paid. Thanks to what it was bundled with, the proposal to lower minPoolCost from 170 to 75 is one on which SPOs hold a vote.
Last issue we wrote that “what gets built is decided not by the floor, but by the votes.” This week those votes turned toward the feet of the people who build the floor. The next ones to be counted are our own.
Sources
- Intersect MBO — Weekly Update #122 (2026-07-31; ratification requirements, CC audit completion, amendment portal, Dijkstra phases): https://intersectmbo.org/news/intersect-weekly-update-122-july-31-2026
- Intersect MBO — Weekly Update #121 (2026-07-24; baseline data for last issue): https://www.intersectmbo.org/news/intersect-weekly-update-121-july-24-2026
- CIP-1694 (security-relevant protocol parameters and the additional SPO vote): https://github.com/cardano-foundation/CIPs/blob/master/CIP-1694/README.md
- Intersect Constitutional Amendment Portal (alpha): https://cap.intersectmbo.org
- Input Output Research — Cardano Vision 2026 mid-year report draft (published 2026-07-16, review closed 7/30; CIP-1194 fee market design): https://forum.cardano.org/t/ior-cardano-vision-work-program-2026-mid-year-report-draft/155740
- Wanchain sets an August 6 deadline (Crypto Times, 2026-07-31): https://www.cryptotimes.io/2026/07/31/wanchain-sets-august-6-deadline-for-cardano-bridge-hacker/
- Technical analysis of the Wanchain bridge exploit (signature encoding flaw): https://www.cryptotimes.io/insights/wanchain-night-bridge-exploit-signature-flaw/
- Governance action listing (status and deadlines): https://gov.tools/governance_actions
- On-chain vote tallies: AdaStat (https://adastat.net) as primary, cross-checked against Koios
- Prices: CoinGecko (https://www.coingecko.com/)
Transparency Note
Governance approval percentages are taken from AdaStat as the primary source and cross-checked against Koios, with only matching values reported. Capture times are given in the text. Voting continues until expiry, so these are not final figures.
Approval percentages are expressed against the ratification denominator, which excludes abstentions and includes non-voters. Where we write “against,” we mean votes actually cast against; non-voting is not counted as opposition. Non-votes do work against ratification in the arithmetic, but that is distinct from an expression of opposition. For the Constitutional Committee appointment action we report no approval percentage, because the denominator differs between sources.
We report no figures for SPO voting on the minPoolCost proposal, as we could not confirm a second, cross-checkable tally. That SPOs are a voting body on this action, and that the requirement is 51%, come from Intersect’s published statement and CIP-1694.
Prices are CoinGecko daily snapshots on a UTC basis, comparing July 25 and August 1. Last issue used a snapshot from a different moment, so its July 25 figures differ slightly.
Readers may have seen references to a Midnight submission to United States regulators. The company’s article reporting that filing is dated July 1 and is not a development from this week, so it is not covered here. Similarly, a developer headcount figure for Midnight and a change to NIGHT collateral terms by Indigo were both set aside, as we could not confirm primary sources.
This article is for informational purposes and is not a recommendation to buy, sell, or hold any crypto asset.