Hello @middleway.eth !
I’m currently developing Milestone 2.
I’m facing issues with integrate again currently all optimization techniques I’ve used before (simulation introduced some new problems), but I’m sure that I will be able to make last commits to make it reviewable within today or tomorrow morning.
Currently we have a Milestone 3, which can be reviewed already.
About Milestone 3, I was thinking to make docker fully automatic (it’ll always install latest stable version from github of the project through docker compose), what do you think about this idea?
Project status Update:
Milestone 2 on-hold.
I’ve encountered a loop bug related to trace_callMany using CoW RPC mainnet, which I cannot resolve myself.
Tried to troubleshoot it in this two days, but at this moment I am not getting any more errors, logs, or tracebacks which could be helpful to resolve the issue.
Already tried to use Infura, which seemed to work okay.
Will mention the bug itself on discord.
Dear @april, we appreciate your motivation and wanted to thank you for the efforts you’ve made to this grant.
Unfortunately we found it to not be a good fit this time, and therefore we’ll be terminating the reminder of the grant.
This has been well communicated and discussed in advance, so I hope the reasoning is understood.
Wanted to clarify that any progress updates regarding a grant should be publicly posted on the grant thread. The lack of communication was the main driver for the termination, together with initial assessments that were not so positive.
We’ll be reviewing any progress that has been made so far if publicly posted here in the coming week, and will be open to changing our view.
Given the review outcome we’ll be open to consider appropriate compensation and potentially reconsidering the termination of the grant.
I appreciate the response and the opportunity to clarify a few things before proceeding further.
1. Milestone 3 Was Publicly Communicated and Ready for Review in October 2024
On October 1, 2024, I explicitly posted on the forum stating that Milestone 3 was ready for review.
The core alerting system was completed on September 17, 2024, along with a working Docker version.
There was no response rejecting Milestone 3 at the time—instead, it was left unreviewed.
2. Milestone 3 Was Deprioritized Per Felix’s Request, Not Due to Inactivity
Felix directly instructed me to focus on Milestone 2 instead of finalizing automation for Milestone 3.
The final Docker/Devcontainer improvements were made on December 4, 2024, based on his explicit request.
If prioritization changes occurred internally, I was following the DAO’s instructions, not neglecting progress.
3. Milestone 2’s Final Step Was Uncommitted Due to Internal Direction
The last step of Milestone 2 (fully changed bundle_simulation.py) remains uncommitted because I received the termination notice same morning as I were commiting it (I previously adviced Felix and Chim about the commit process and when it will be commited, no one of them objected, but accepted).
If I had committed it earlier, it would have been available for review…
4. The Termination Process Violated Governance Rules
Grant decisions require a governance proposal (CIP).
This termination was decided unilaterally, without a Snapshot vote or governance discussion.
The agreements specify that the DAO forum is just one possible interface, not the only required one—official communication happened on Discord & GitHub as well.
Where is the governance record proving that this termination was properly approved?
5. Request for Clear Confirmation on Payment & Governance Compliance
Given that Milestone 3 was completed and available for review months ago, will it now be evaluated for payment?
If not, please clarify why the work was not rejected at the time and why I was instructed to prioritize Milestone 2 instead.
If the termination is final, will this decision go through the proper governance process as per CoW DAO’s policies?
I would like a clear response on these points before deciding my next steps.
There’s no reason to review milestone three before completion of milestone two. Milestone two represent the core of the project, so any further work on alerting and productionazation is meaningless without completion of prior milestones.
The grant was approved in Jul 2024. The total grant duration estimated by you was 11 working day. We’re having this conversation 6 months (!) after your estimated completion date.
Technically the grant should have been canceled automatically long ago according to the Timeline and Completion Extension Clause.
I acknowledge that further communication with the you after the completion deadline might have been interpreted as extension even though an extension have not been formally communicated. To be fair, we’ll not consider the grant automatically canceled.
Internal grants processes are NOT required to comply with CoW DAO’s CIP process, so this point is clearly false.
You actually bring up a valid point, at the time that this grant was approved, there was no definition for a termination process. This point has been addressed in the new Grants T&C which define a committee snapshot vote for terminating grants.
In order to try and offer a path to closing this grant on good terms, I’d suggest you take until the end of the week to complete any work you’d like on the github repo, finalize PRs for review and make an official submission here in the thread.
We’ll queue your submission for review and come back with feedback.
As a clarification, milestone three is not required to be considered for payment if milestone two is not considered complete.
We’ll reserve the right to offer partial compensation as a show of good will even in case we’ll consider the milestones incomplete - but would not commit to such compensation.
In case that not all milestones can be considered complete, we’ll proceed to to a snapshot vote to ratify the the official termination of this grant.
I apologize for announcing the termination of this grant without prior communication (even though it didn’t represent violation of any grants conditions that applied to this grant).
I hope you’d find the proposed course of action satisfactory, and hope that it’ll lead to a closure of this episode on good terms.
Thanks for your response and for acknowledging the communication issues. I appreciate the opportunity to clarify the next steps.
1. Finalizing My Submission
I will finalize the GitHub PRs and make an official submission in this thread by the end of the week, as suggested.
Since Milestone 3 was largely completed in September 2024, and Milestone 2 was in its final stage, I will ensure that everything is properly documented for review.
2. Clarification on Payment Conditions
You mentioned that Milestone 3 is not required to be considered for payment if Milestone 2 is not complete. However, Milestone 3 was fully completed as an independent work item and was publicly communicated for review in October 2024.
While I understand that Milestone 2 is important to the overall system, the work completed for Milestone 3 remains valid and separate—just like in any modular development process where multiple components depend on each other.
If milestones are not treated as standalone deliverables, then either:.
.The term “milestone” must be explicitly defined in the grant agreement, OR
.The entire project should be published as a single deliverable and paid for in full upon final completion—not split into abstract “milestones”.
Internal priorities shifted when Felix instructed me to focus on M2 instead of finalizing M3. This means any delay in M3 was a management decision, not a failure on my part.
Since you mentioned the possibility of partial compensation for incomplete work, it’s important to clarify:
.Why would fully completed work (M3) not qualify for full payment?
.What is the actual basis for considering “partial compensation”?
.Who determines that, and under what criteria?
3. Addressing the Project Timeline & Inconsistencies in Deadline Enforcement
You mentioned that I originally estimated the work would take 11 working days, but that we are now 6 months past that deadline.
However, the formal snapshot vote for this grant only closed on August 2, 2024, which means that the project could not have officially started before then. (Snapshot Vote Link)
Furthermore, Milestone 1 was fully completed and published on September 5, 2024, which was already past the 11-day estimate.
Despite being beyond the estimated timeline, this milestone was reviewed, approved, and fully paid without issue.
This proves that the deadline was not being rigidly enforced at the time and was already acknowledged as flexible.
Given the project’s complexity, it was clear early on that a longer timeline was necessary. That’s why Felix and Chim continued discussions with me rather than enforcing an automatic cancellation.
Since you acknowledge that continued communication could be interpreted as an extension, it’s important to fairly assess the actual work done rather than focusing on an outdated estimate.
4. Clarification on the Upcoming Snapshot Vote
I understand that if all milestones are not considered fully complete, a snapshot vote will be held to ratify the official termination.
Can you confirm that this vote is strictly about the grant termination and will not impact payment decisions for completed work?
Since there were no termination rules at the time this grant was approved, this vote will essentially serve as a retroactive approval—please clarify if it will affect any financial considerations.
I appreciate your time in addressing these points, and I look forward to submitting my final work by the deadline.
The fact that some work is possible to be completed independently doesn’t mean that it is required to be considered for payment irrespective to the completion of prior milestones.
I think it is completely logical to consider a grant proposal as a single body of work with logical order to the sequence of milestones.
I find it hard to contest the assumption that a milestone that deals with alerting has any meaning without the completion of a prior milestone dealing with the actual core of the grants: data retrieval, processing and handling. There’s no way to “productionise” the system without its core component.
Payment according to milestones comes mainly in favor of the grantee, avoiding long delay between work and payment. This doesn’t in any way mean that a grantee is free to pick and choose which milestone to execute first and expect payment in accordance that order.
Again the whole idea of the proposed course of action to to offer a path for concluding this grant on good terms, so I find it unnecessary to discuss the fact that we didn’t stick to the official Timeline and Completion Extension Clause. Obviously this doesn’t form the basis to assume there was an official extension of the grant timeline.
I can confirm that the review will be made on a pure technical basis without considering milestone submission delays or the potential eventual termination of the grant.
I’ll be finalizing my submission as requested and will post it by the deadline.
To avoid further misunderstandings, I’d like a clear confirmation before proceeding:
.If M2 is reviewed and accepted, will M3 be paid in full?
.If any part of M2/M3 is considered complete, will it be paid accordingly?
Also, regarding milestones:
M3, as an alerting system, is built modularly and could fit into any other system, just like any properly structured software component.
The technical independence of M3 (and any other part of the project) was never disputed previously—only its payment.
I want to ensure that the review process is structured and leads to a fair resolution. Since I followed DAO instructions, including shifting focus to M2 as directed, I need to understand what to expect regarding payments.
I’m not interested in debating theoretical milestone structures—only in ensuring that completed work is fairly reviewed and paid accordingly.
I appreciate a clear response on these points so we can move forward efficiently.
If M2 is considered complete it will be paid.
If M2 AND M3 are both considered complete they will both be paid.
M3 cannot be considered complete independently of M2 because productionzing a system without its core component is impossible.
The review will be conducted fairly to the extent you feel Felix treated you fairly so far, as Felix will be the reviewer. (which I’d argue he did way beyond any expectation).
Hello @middleway.eth ,
Got it. I’ll finalize and submit everything as planned.
Just to clarify:
From the start, the project was defined as an SDK with a product-ready tool in mind. The evaluation process should take into account the original application description, which was previously discussed and adjusted accordingly. M3 is not just a productionizing step—it’s a fully functional alerting system, built as a modular component, just like the rest of the project. This was never disputed until now.
Since Felix will be conducting the review, I assume that means there will be clear criteria for determining completeness, ensuring the evaluation aligns with the project’s application intended structure.
I’ll wrap things up and submit as expected until the expected deadline.
I have now finalized and submitted both Milestone 2 (Bundle Simulation) and Milestone 3 (Alerting & Productionization) as per the original grant agreement. The implementation has been committed to a dedicated branches in the main repository and is available for review.
What This Submission Includes
Milestone 2: Bundle Simulation
Greedy algorithm for bundle merging
Simulation using trace_callMany with stateDiff
Comparison of actual vs. optimal user refunds
(For exact changes regarding last commit, see it’s description )
Milestone 3: Alerting & Productionization
Fully implemented alerting system (Slack, Telegram, logs, completed September 17, 2024)
Initial working Docker setup (completed September 17, 2024)
Finalized Docker/Devcontainer adjustments (completed December 4, 2024)
Deployment instructions
Both milestones are modular, functional, and independently reviewable.
Next Steps
Please confirm receipt of this submission and provide an expected timeline for review.
Once reviewed, please confirm approval or any required modifications.
Following approval, I expect payment per the grant agreement terms.
Appreciate the confirmation. Just to stay aligned, I’d like to get some clarity on the expected review timeline. Since “early March” is a bit open-ended, can I assume that means a final review no later than March 7th?
Additionally, if the milestone is approved, will payment be processed in the next payout cycle, as per the standard grant process? I just want to make sure everything is structured properly so that we can wrap this up efficiently.
Let me know if that timeline aligns. Looking forward to the review process.
There’s no reason to commit to a timeline without being sure we can meet it.
Felix is the technical lead of CoW Swap and frankly has some higher priority items on his agenda.
Especially since this grant is being delivered months after schedule, where it is almost irrelevant to the original need which led to posting the RFP.
I’ll update once a final timeline can be confirmed or when the review is completed.
I see what’s happening here, and I want to be very clear.
The grant was approved, and the work is fully delivered according to the original agreement.
The excuse that Felix has “higher priorities” is not a valid reason to delay payment indefinitely.
If this project was so “irrelevant,” why wasn’t it canceled months ago?
Instead, CoW DAO had me continue working, and now that it’s done, you’re stalling the review process.
At this point, it’s clear that this is no longer about a technical review but an intentional stalling tactic to avoid payment.
I expect transparency. If CoW DAO claims to be a serious entity in the Web3 space, then act like one.
Give me a real review deadline. Not “when we can.” A concrete date.
If no review date is given, I will escalate this issue publicly across broader DAO communities.
I have no issue pushing this into Optimism, Gitcoin, Messari, and other governance spaces where DAO funding fairness is taken seriously. This isn’t just about one grant—it’s about whether DAOs can change priorities, delay payments, and get away with it.
I respect structured processes, but this is not a structured process—it’s deliberate obstruction.
By the way, I appreciate the attempt to downplay the situation. I understand that from a governance perspective, it’s easier to act as if this is just “one minor case.” Unfortunately, that narrative doesn’t align with reality, nor with the fact that CoW DAO had to rewrite its grant rules entirely after this situation arose.
Your case is actively discussed every single fortnight on grants committee meetings.
We have had issues with not just your grant achieving satisfactory outcomes. We have a fiduciary duty to CoW DAO to ensure that Grant funds are spent wisely and achieve their intended outcome. Rewriting of the terms and conditions began in November in preparation for the new grants cycle - not as an output of your specific case - though it certainly reinforces some of the decisions that were made on process.
The “Felix has other priorities” - is quite true. @middleway.eth I’d strongly suggest that another reviewer be allocated for this or at least a plan to derisk this so that it can be seen to promptly.
Grants Committee members raised with @middleway.eth internally that the method by which the grant was terminated initially was sudden, and that this could have been done better. This has been acknowledged @middleway.eth within this thread.
Grants Payments happen by routine at the end of the month. As you can most certainly appreciate, co-ordinating payments and the synchronous manner of operating a multi-sig necessitates overhead so it has been a long standing practice to make payments once a month, although exceptions have been made for exceptional circumstances (like making payments to auditors to get contracts audited that were holding up product launches).
You are seeing deliberate obstruction where there frankly is none. I agree that there needs to be a concrete date by which there is assessment made, stick to the facts and articulate solely this point, as the rest is simply emotion and conjecture - belligerence will get you nowhere.
I appreciate the insight, but let’s stick to the facts.
The grant was approved and fully delivered as per the original agreement.
The “outcome” metric was never mentioned before completion, and shifting the goalposts now isn’t fair.
You’ve confirmed that this case was actively discussed in committee meetings (I assume, for the last several months), yet no concrete resolution has been reached.
The idea that Felix has other priorities is understandable, but a new reviewer should have been assigned months ago if that was the case.
I’m fully aware that grant payments happen at the end of the month, but I’d like a concrete review date well before then.
What is the exact deadline for the review? If this is actively discussed every fortnight, there should be a clear timeframe.
If a payment is expected in the March cycle, I expect confirmation that this will happen if the milestone is approved.
I’m not interested in debating whether this was obstruction or not—the only thing that matters is getting a clear answer and finalizing the review.
Let’s get this resolved properly. Looking forward to the exact review date.
We looked into finding an alternative reviewer but couldn’t find a good fit.
Given the circumstances, we can commit to posting the review feedback by April 15th which will allow you to respond or fix any points that come up during the review, up to the end of April.
We’ll be queuing payment of all eligible milestones for in April’s payment cycle.
Understood. I’ll be expecting the review feedback by April 15th and the corresponding payment cycle in April. I trust that this will be handled transparently and without further unexpected delays.
Looking forward to the next steps, thanks for the answer and your time.
What is the exact deadline for the review? If this is actively discussed every fortnight, there should be a clear timeframe.