Funding Request: Aragon Zero Knowledge Research guild 2023

[p4u]

Dear Juliette,

I wanted to share some insights on your query, from my perspective as someone who is actively immersed in developing offchain voting solutions for the Ethereum ecosystem.

To start with, zk-POPVOTE, as I see it, is a ground-breaking voting solution with the potential to introduce onchain, token-based anonymous voting to DAOs. The emphasis here is on “token-based”, which is a capability that neither OVOTE nor any other current solutions can provide.

In scenarios where DAOs prefer to structure their governance based on ERC20 or NFTs (as opposed to a list of signers like a Multisig or OVOTE), while also having gasless and/or privacy capacities, a system such as zk-POPVOTE becomes indispensable. While minor tweaks could bring up potential alternatives, the core concept is the same. For more details, you might find my recent article useful: Road to Anonymous Binding Voting for Ethereum

The PoC research being conducted by AZKR and Aztec could potentially revolutionize DAO governance. However, I’d be remiss not to mention that this is not a short-term achievement due to the heavy computation required on the browser scope (proving several Ethereum storage-proofs within a zkSNARK is quite resource-intensive). But the encouraging news is that what seemed impossible a few months back is slowly becoming a reality.

Regarding Tally CLI, it indeed doesn’t directly benefit Aragon OsX. But, it serves as a crucial tool for testing and evaluating the proof-of-concept, an essential phase in the development of such systems.

In my view, AZKR should continue to push the envelope in this direction, without the pressure of delivering a finished product in the near to medium term. The complex nature of this endeavor calls for patience, research, and time.

Lastly, I’d like to comment on TLCS, which is crucial for enabling “shielded” or “secret until the end” voting, a very important feature for any kind of democratic governance. At Vocdoni, our goal is to integrate this service in the future, as it enhances our current solution.

Do bear in mind that this response is strictly my personal viewpoint, not endorsed or reviewed by any member of AZKR. I genuinely hope this provides some clarity and a better understanding of the objectives at hand.

[juliettech]

Super helpful @p4u , thank you! I was definitely missing context on many of what these solutions were, so your explanations are great.

Great to hear Vocdoni is also benefitting from the findings of AZKR. The work being done is certainly innovative and worth pursuing for the greater industry. I’d still push for a clearer strategic plan for pursuing one solution over another internally and ideally a stronger collaboration between AZKR and Product to ensure these solutions come to production as soon as possible. Hopefully we can see more of this for the next quarter!

[juareth]

Hey Alex, thank you for your transparency around the OKRs. I would like to also provide publicly me individual feedback.

O1. Generalization of Nouns voting solution to zk-POPVOTE:

Your ongoing efforts in both browser and server approaches certainly make sense, given the constraints of Product Development capacity. Personally I think under ideal circumstances, focusing on one solution at a time before moving on to the next might be more efficient. However, pushing forward with both approaches this quarter seems like a reasonable strategy. As you’ve suggested, fostering closer relationships between the Guilds could indeed enhance our coordination and efficiency when that time comes.

O2 Implementation of Time Lock Cryptographic Service (TLCS) from Alpha to Beta

You know already my concerns with this service, but let me put them in written form just for visibility’s sake.

Whereas I understand the importance of this service for “Shielded Voting” and that requirement for the Nouns project, the chosen solution is not, in my opinion, the right approach. It will take the Guild into long and heavy conversations and support for something that was never the main goal, and that concerns me. As I mentioned, I believe architecting this as a rollup focusing on one scheme that can be proven onchain afterwards makes more sense rather than going in the League of Entropy approach.

Of course, I might be wrong, and would like to trust you on this one this quarter and explore the public service path you’ve started preparing. Just be mindful, this was never the main goal, and if you see it starts taking too much time, it will go in detriment of the main objectives for the Guild.

O3 Implementation of ZkRegistry from Alpha to Beta

I really like this one. I can’t see other way of solving this problem and solves something lot’s of players in the ecosystem are having. Also, converting this into an EIP seems like the correct approach.

Very happy with this approach, please contact with the Growth team and ping me if you need help explaining it.

O4 Maintain a reasonable level of dissemination activities

As every quarter, your Guild always exceeds at increasing awareness of the zk technology not only through talks at conferences, but also internally. I really like hearing your team members explain this technology that enables so much. Thank you for this contributions, really appreciate it.

---

In other order, I think it might be interesting touch other points along the quarter that could make the team more aligned. I’m sure you already plan on doing that as “business as usual”, I guess I just like them seeing written down for everyone acknowledgement of work being done and that those more “peopleware” problems are also being addressed. I’m talking about better defining the role between research and implementations/deliveries, or roadmaps doubts the team might have regarding the role within Aragon. Either way, wether it’s part of a OKR or not is another question, but would love seeing you putting some love to those concerns in one way or another.

Again, thank you for your hard work,

Carlos

[Orion]

I think it would be good to showcase any proof of concepts, products don’t have to be finished to see progress. A community screencapture could show proof of concept and talk about the pros and current hurdles still yet to be developed. I think it is important to communicate to the less technical majority of progress being made.

[alex-kampa]

Hey Carlos, thanks for your comments. I have now added the following paragraph at the end of the OKR post:

“In addition to the explicit OKRs stated above, the guild will keep working to ensure a smooth inter-guild coordination and effective contribution to the Aragon project. This includes, but is not restricted to: keep the regular meetings with Growth, reinstate the regular meetings with Product, contribute to the Product roadmap definition and execution, participate in Aragon’s discussions, etc.”

[alex-kampa]

@Orion

I think it would be good to showcase any proof of concepts, products don’t have to be finished to see progress. A community screencapture could show proof of concept and talk about the pros and current hurdles still yet to be developed. I think it is important to communicate to the less technical majority of progress being made.

Thank you for your suggestions, completely agree with you that we also need to communicate to a less technical audience. In September, we plan to have a public session to present the results in a way that is accessible to a general audience.

Please note that during the execution of the project we have participated in 2 Twitter spaces, have issued 4 progress reports (cf Aragon-Aztec Private Voting - Documentation - Main page - HackMD) and are preparing a final report that will also be made publicly available.

[Orion]

Hi Alex,

Good to hear about the spaces and the reports - were the twitter spaces recorded?

[Yakitori]

Hello, have you thought about marketing those research topics? I’d be interested to know which ones could be sellable and accrue (monetary) value to the Aragon network.

[Orion]

i think anonymous voting is something one could charge a fee for. People would pay maybe a small fee to be anonymous vs no fee for publicly voting?

[Yakitori]

Regarding private voting, I think it could be very, very easy to monetize it, given the latest news regarding SEC and CFTC investigations of DAOs and their members. A great killer app for Aragon!

[p4u]

Regarding Anonymous Voting, the Vocdoni Guild is currently developing an Aragon app plugin that will facilitate offchain, gasless anonymous voting. We aim to deliver a fully functional version of this plugin, inclusive of a user interface, within the forthcoming months. You can find the technical proposal here: Anonymous Vote for DAOs with Vocdoni - HackMD

As for monetization, it remains a viable option. To create anonymous votes on the Vocdoni blockchain via the Aragon app plugin, users will need to acquire certain tokens. The cost of these tokens will be dependent on factors such as the size of the census, the duration of the election, and more. While we have not yet determined the specific tokenomics of this token, one possibility is to maintain a peg or bridge of some sort to ANT. This way, each time a vote is cast (whether in the WEB2 or WEB3 space), ANT holders would receive a portion of the revenue generated.

Of course, our intention is not to introduce obstacles to DAO governance. Therefore, our proposal includes allowing a certain number of votes to be cast free of charge. This would essentially result in the “big players” subsidizing the governance costs for the “small players”, thus enabling the implementation of a freemium model.

EDIT: Sorry if I was a bit intrusive here. I just wanted to share with you that we are exactly working on this direction. In order to achieve the full potential of Anonymous voting, we would need to use the existing AZKR prototypes and research. So, this needs to be actually a join effort

[p4u]

@juareth

O3 Implementation of ZkRegistry from Alpha to Beta

I really like this one. I can’t see other way of solving this problem and solves something lot’s of players in the ecosystem are having. Also, converting this into an EIP seems like the correct approach.

Very happy with this approach, please contact with the Growth team and ping me if you need help explaining it.

I would like to comment on this point, since this is something I have been evaluating and talking about since long ago. The fact that ECDSA is non-deterministic is a big barrier on the zkSnarks voting solutions. So such a registry is a need, not only for Aragon but for the whole ecosystem.

To this end, I would propose to actually implement a useful solution, not only for the AZKR purpose, but for the whole crypto space. It might be a win-win, since the more Ethereum users registered on this smart contract, the less friction for using it. Be aware that entering on this registry will cost Gas, so the user should see the utility of it, also outside the Aragon context.

As far as I know, the current Registry proposal is made out of keccak hashes, a non-friendly zkSnark hashing function. I understand Noir+UltraPlonk can handle this hashing, but it is very expensive for most of the current zkSnark libraries and frameworks due to the enormous number of constraints required (compared to Poseidon or Mimc).

I would like to ask the AZKR team to consider other needs (including Vocdoni’s) and try to approach a solution that can be useful for everyone. Maybe it would require some more time, but it will be a work that we can reuse in the future.

[alex-kampa]

@p4u

I would like to ask the AZKR team to consider other needs (including Vocdoni’s) and try to approach a solution that can be useful for everyone. Maybe it would require some more time, but it will be a work that we can reuse in the future.

We are completely aligned with this. The current zkRegistry is just a first iteration, done with the specific needs of the Nouns DAO project in mind. The next step will be to find out what additional features will be needed to make it as useful as possible for the entire ecosystem, including of course Vocdoni.

[alex-kampa]

@Yakitori

Hello, have you thought about marketing those research topics? I’d be interested to know which ones could be sellable and accrue (monetary) value to the Aragon network.

As a research group, our focus is on creating prototypes and proof-of concept implementations. Topics such as marketing, product development and monetization are not among our core competencies. However, we are ready to provide any support necessary to other groups within Aragon to develop products and services based on the prototypes we have worked on. I also think that Vocdoni is better suited at the moment to monetize their services, cf p4u’s comment above.

[roger]

Q2 activity report

Summary

Despite the OKRs achievement of Q2 is under Q1’s (72% vs 90%), the key accomplishments made during Q2 make us face Q3 very optimistically. We believe that the deviation is mostly due to the uncertainty when setting the objectives.

Work done

As the number of KRs per objective shows (6 out of 13) the efforts have been focused on the design and implementation of the Nouns proposal Proof of Concept (PoC). The public periodic reports issued every ~3 weeks can be found here.

To the best of our knowledge, this PoC, which we plan to deliver in the coming two weeks, will be the first working prototype enabling on-chain, (highly) trustless, anonymous voting with fairness (i.e. no partial tally possible).

Main deviations

AD-KR-415 Delay-relayer service PoC with vote batching and gas costs paid by third parties

Due to time constrains and its heavy dependencies on preceding tasks, we decided to postpone this KR to Q3. The 10% is because we already have a preliminary design.

AD-KR-417 Successful integration of additional services (delay-relayer and TLCS) into MVP

This is the task we are working on now and the only one remaining to be able to deliver the PoC. Once done, we will set it to 45%. The fact of having postponed AD-KR-415 to Q3 already limited the maximum score to 50% (one out of two services to be integrated). The remaining 5% is due to some small steps that we wanted to automate even further.

AD-KR-420 All Guild Leads and Product Leads are satisfied with the quality of collaboration with AZKR

The results of the survey we run at the end of the quarter shows a wide range of opinion in the key question “During Q2, were you satisfied with the quality of collaboration with AZKR?” (average: 50% -the score of the KR). The answer to the questions regarding strengthening collaboration in Q3 show good will, but concerns about the time available to do so.

AD-KR-423 Survey to assess our knowledge transfer tasks conducted (pending from Q1)

Despite the number of dissemination activities made by AZKR, we decided to skip this KR because the lack of a proper training program (which is out of scope) precludes from being able to make a meaningful evaluation in this regard.

Next steps

As already explained here, the main objective for Q3 is to keep refining the PoC and prepare it to be integrated in Aragon OSX in Q4. We will also address AD-KR-415 (the Delay-relayer service).

[luis]

Anonymous voting (and even anonymous token holding) seems like a key enabler for DAOs to keep working, even under regulatory uncertainty. I also do think DAOs would pay for it, and in this case a protocol can be built that accrues fees from that usage. Makes tons of sense.

[Dcasares]

Curious how this would work, would it be a subscription model or a per vote system? Would be interesting to see how that product evolves

[alex-kampa]

Curious how this would work, would it be a subscription model or a per vote system? Would be interesting to see how that product evolves

This is obviously a good question in the context of product design. However as we are a research team this is out of scope for us, as we are very much focused on technical issues for the time being.