Compound Grants Program

I am glad we could work together on this!

1 Like

We’re excited about running this experiment too!

The questions you pose are all very good and the community should get an answer to them. Let’s see if I can do a good job of answering them.

Why should the community, and COMP token-holders trust you? What are your qualifications, and background? What is your experience in the Compound community to date?

I spent the last three years investing at Digital Currency Group (DCG). During that time, I invested in 100+ companies and projects (some went on to be very successful, and some, umm…did not go on to be very successful). Early this year, I left DCG to work on something that I think is critically important and all at once underinvested in: network governance. That’s how I find myself here today.

Unfortunately, I didn’t have the time to actively participate in Compound’s community while I was at DCG. Since early this year, however, I have been lurking on the forum and Discord on a daily basis. Going forward, I plan to be a far more active member of the community. (Less lurking and more posting).

What about the other members, that disburse funds?

We tried very hard to make sure members of the committee represent the interests of various stakeholders in the community and/or have experience in providing funding to projects in the space. Jesse is an investor at Variant (and a16z before that). Getty, Monetsupply, and Sam are independent members of the Compound community. Ken has deep experience with the grants process thanks to his work for the Ethereum Foundation. Finally, Ryan works at Coinbase, and before that, was an investor at CoinFund. I’m biased, of course, but I think we have some very high quality committee members!

Should the members be active Compound community participants? Some names I don’t recognize from Discord or these forums.

Active community members are absolutely critical to the success of the program. Few know what areas/people Compound should provide grants to as well as they do. We tried our best to form a committee that’s well-rounded: it has both active community participants as well as folks who know how to provide funding to projects and ideas. Having said that, we’re always open to hear from other community members who’d like to get involved.

Nobody on the list (to my knowledge) has a deep familiarity with the protocol codebase, which might be useful in quantifying grants for technical development.

That’s certainly a weakness of the committee today. We’d absolutely love to hear from people who are familiar with the codebase.

In the case of a protocol development, do you envision a grant before work is completed, after it is completed, after it is merged into the protocol? At what stage of development, would somebody apply for a grant? In the workflow, this isn’t made clear.

This is a question we hope to have an informed answer on after we get data points from running the grants program for several months. Certain applications, for example, would benefit from receiving a grant before the work is completed. Other applications may benefit from receiving a grant after the work is finished. Applications that have several intermediate stages may benefit from a hybrid approach: part of the grant is given upfront, part is given once intermediate stages are completed, and part is given once the work is merged into the protocol. We hope to have a better answer to the question after running the pilot and seeing what works and what doesn’t work. After several months of experimentation, best practices should start to emerge.

Since we don’t have the best practices yet, we’ll want to start with the largest funnel we can to get as many data points as possible. To that end, when we start, we would accept applicants who are at any stage of the development process: those who haven’t started the work yet, those who have completed some of the work but have several more steps to go, and finally, those who have completed the work and merged it into the protocol but haven’t yet received payment for a job well done.

It seems like somebody would (or should) apply before completing development work (or another type of contribution). Would you disburse funds, partially disburse funds (with the remainder at a later stage), or communicate a “provisional” grant/acceptance?

I think the above answer covers this question too.

Any steps in mind to eliminate potential fraud?

Being a victim to fraud is the cost of doing business, particularly in this space. While we will take utmost care in screening out bad actors, mistakes may occur and losses may be incurred. That said, a small potential loss shouldn’t prevent us from doing something that has a chance of delivering a large potential gain to the community.

To keep the probability of loss as low as possible, we will likely need to verify the identity of each applicant prior to disbursing a grant. That is how we can ensure no baddies make it through the grants process. If this is indeed the final approach we take, we will be extremely careful with protecting the identity of applicants (the only people who would see their identity are members of the grants committee). But we’re still thinking this part through, so input from the community would be very welcome!

5 Likes

Thanks for the comments @dabar90!

…but I did not see repairing oracle solutions and compensating damaged users (or at least a statement) among the listed priorities.

Repairing oracle solutions would certainly fall into the scope of the program (my guess is it would fall into the protocol and parameter development category). As far as compensating damaged users goes — do you think a grants committee should be handling that, or would a separate governance proposal be the better option?

I would prefer that the interests of your team be aligned with the interests of the users and that these be priorities. That will be win-win situation.

We’re in agreement! But at the same time, we want to keep the scope of the program manageable so as not to boil the ocean. We will be listening to user requests closely: if there is something they want to improve on Compound but lack the technical knowledge to do so, we will encourage them to find someone with the technical knowledge, apply for a grant, and make the improvement.

And i dont like expressions like this because it sounds to me like you’re addressing a board of directors of a joint stock company. The very purpose of the DeFi and DAO concept is for users to manage the protocol, so you can start with that.

I think that’s a completely fair comment. At the same time, I do think the term stakeholder is apt because it captures users who use Compound but aren’t actively managing it. How else do we call these people? Do their interests matter?

5 Likes

This an extremely important point. We are well aware that most grants programs are annoying to apply to, require some degree of lobbying, and in general, are a dreadful process for the applicant. We want to make the process as quick and easy for applicants as possible.

We’re still finalizing the process, but perhaps I can share some of the details that will hopefully encourage you to apply if the proposal passes.

Applying will be fast. We will ask applicants for only the most important information. The application won’t take more than 30 minutes.

Approvals will be fast. We won’t require applicants to jump through hoops to receive a grant. They won’t need to lobby all of the reviewers. That would stink, and it would prevent applicants from applying ever again. Our default position is applicants are here because they care about improving Compound. With that in mind, we will be optimistic with approvals. (We’re not gatekeepers; rather, our goal is to open as many doors as possible).

You don’t need to be well-connected. But you do need to care about Compound and have the ability to make it better.

Hope that dispels some of your concerns!

5 Likes

Thanks @sukernik and everyone else involved in bringing this forward. A well-structured COMP grants program should help encourage a broader set of developers and community members to come forward and contribute to the ecosystem. As you note, the need to front audit and other development costs is often a prohibitive barrier to all but the most well-funded development teams. A grants program could help address this issue and bring more diversity and breadth to the ecosystem. It could also fund smaller, one-off “growth” projects that are too small or otherwise not practical to bring through the full governance process.

While delegating authority to a committee does have certain downsides (as @jared notes), I think in this case it’s worth the tradeoff, for a few reasons. First, any updates to the core protocol would still need to be passed through the formal governance process before being merged. The committee itself would only be empowered to disburse funds to help defray some of the upfront development and audit costs. Second, for grants not involving core protocol development, the relatively small size of the budget - together with the 4-of-7 multisig design - further limits the downside risk of bad actors trying to abuse the system. And finally, as discussed above, the program could meaningfully broaden the set of community members who are able to contribute to the protocol, which should in turn enhance overall protocol health. Collectively, these factors weigh in favor of the committee structure for this specific use case, at least on an experimental basis.

Operationally, the program should aim for efficiency while also not becoming overly bureaucratic (per @kermankohli’s comment above). One way to help strike this balance is to scale the amount of reporting, disclosures, etc. that a grantee has to provide alongside the amount of funding requested. For example, it seems reasonable that a project requesting $250,000 should have to provide more robust disclosures and reporting than a project requesting $5,000. A basic tiering system along these lines could help to reduce the risk of fraud while also not over-burdening smaller projects and discouraging them from participating.

Finally, the proposal could ideally aim to quantify the success metrics a bit further. This exercise could be informed by lessons from the Uniswap grants program or from other more long-standing grant programs in the industry (e.g., Ethereum Foundation). Perhaps the program could even include incentives for the committee members that are tied to achieving certain metrics (though perhaps that’s better suited for a v2 of the program).

Overall, this looks like a great first step towards creating a COMP grants program. We look forward to hearing other feedback from the community.

Jeff at a16z.


For disclosures, please see Disclosures - Andreessen Horowitz

6 Likes

further limits the downside risk of bad actors trying to abuse the system

FWIW I would be more concerned about the filtering bias than malicious behavior, but I’ve found a lot of the points on this thread pretty compelling. And actually, if a record of all the applications for grants are publicly visible, as well as all the decisions (i.e. if its totally transparent), I think it might work pretty well.

10 Likes

While I believe that a Grants committee would be a new positive to the Compound ecosystem, I think there are a lot of issue which should be addressed with the current proposed program prior to initiating it.

Membership

Firstly, to echo Robert, I find the committee membership questionable. Some of members have never posted on these forums or in the Discord as far as I can tell, and I don’t believe that any member is familiar with the Compound protocol codebase.

While in a more broad grants committee this may not be problematic, the scope of the committee is defined as the Compound ecosystem. For such a scope, an intimate involvement within the community is a must for each member in my eyes. Also, having at least one person with a strong understanding of the Compound codebase is a must (I would recommend @adambavosa for this role).

Grant Process

There should be a more concrete process defined for the lead and reviewers to commit to follow. Under the current post, there doesn’t seem to be any actual rules by which the grants committee must follow—I don’t see how Governance can entrust $2M to a committee without any predefined process and rules to follow.

Building on the idea of processes, there should be an off-chain method for the community to vote on CGP related issues such as amendment to defined processes and changes to the committee.

Practicality

Something which should be considered is the practicality of distributing $1M per quarter on the Compound Ecosystem. Compared to the broad $750k of the Uniswap grants program, this is a very large influx of spending, which I’m not sure could be done effectively as of now. As one of the main active developers within the community, I think that $1M of spending per quarter within the ecosystem could cause chaos. I would rather see grants spending scale up to a sustainable amount.

11 Likes

These are really good comments. Thanks @arr00! Responses below.

Firstly, to echo Robert, I find the committee membership questionable. Some of members have never posted on these forums or in the Discord as far as I can tell, and I don’t believe that any member is familiar with the Compound protocol codebase.

Since we posted the proposal, we’ve had a community member reach out and volunteer to join the committee. The company they work for does understand the Compound codebase, which should improve the committee’s grasp of the technical side of things. If you’re reading this, know Compound’s codebase, and want to join the committee too, please don’t hesitate to reach out! (I’m larry#7198 on Discord).

@adambavosa: if you’re interested in joining, let me know!

As far as some of the members not posting on forums/Discord goes — this is something we really went back and forth on. One question I have is should being an active community member be a prerequisite for being on this committee and potentially future grants committees? You can sort of argue both sides. On the one hand, active community members know what the treasury needs to provide funding for better than anyone. On the other hand, someone who is an active community member may not actually be good at running a grants program!

Then there’s this question: what does it mean to be an active community member? Is an investor who owns and thinks deeply about COMP but doesn’t have time to be part of the Discord / forums not an active community member? We certainly don’t have all the answers to these questions, but I think as a community, it’s worth thinking through them to set norms that define how future committees of this nature form.

There should be a more concrete process defined for the lead and reviewers to commit to follow

Absolutely. We didn’t post the process as part of the original proposal to keep it digestible (my sense is the longer it is, the fewer people read it in full). The original plan was to post the proposal, get community feedback, post it for an on-chain vote, and — if the proposal passes — to share a process grants applicants should expect to go through. But since you asked, here is the tentative process! It’s tentative because it hasn’t been tested in practice yet. I’m 100% sure parts of the process will change as we go about administering the program.

Grants Process

  • We want to make this a speedy process for applicants. Apart from the initial ramp-up period, we want to get the time from application submission to grant disbursement to 10 days or less (8 days for lead to review and 2 days for reviewers to vote and disburse the funds)
  • The lead will review every application and send an “approve” or “deny” decision along with a brief explanation to the reviewers. Reviewers are expected to review the lead’s email and reply with an “approve” or “deny” vote as per their independent judgement
  • Depending on the outcome of the reviewers’ vote, the grant will either be denied (no disbursement) or approved (grant is disbursed). If the grant is approved, reviewers should send it to the recipient no longer than 2 days after they first received the lead’s email
  • Multi-sig will require majority vote to pass (4/7 or more votes)

Building on the idea of processes, there should be an off-chain method for the community to vote on CGP related issues such as amendment to defined processes and changes to the committee.

That’s a great idea (I took note of it when you first mentioned it on the Clubhouse community call). We’ll work on setting up an off-chain vote for making small changes to the committee.

Something which should be considered is the practicality of distributing $1M per quarter on the Compound Ecosystem.

To your point, it’s hard to know how many applications we will receive before the fact. We may get $250k or $5mm worth of grant applications — the only way to find out is to launch the program and gather data! Our thinking was it’s better to play it safe than sorry: if $2mm is too much, we will simply return the funds to the treasury at the conclusion of the program. The outcome we wanted to avoid is we get something like $1.5mm in fantastic grant applications per quarter, which would require us to go through the governance process once more to ask the community for more funding. That would seriously slow the process down and make it less appealing to grants applicants.

2 Likes

Fantastic update @sukernik, excited to see this move to a proposal :clap:

3 Likes

First of all, I would like to express my respect to you for making this proposal.

I also have a lot of respect for the members involved in the proposal.

I would like to share my thoughts on this proposal.

I believe that Compound will collaborate with existing lenders in the future, but only if the governance of Compound remains well-diversified.

A report on Defi from the Bank of St. Louis

https://research.stlouisfed.org/publications/review/2021/02/05/decentralized-finance-on-blockchain-and-smart-contract-based-financial-markets

There is a concern that this proposal may run counter to this.

Even if you get the right people to make the right decisions, it is still “centralized”.

What I think is the best thing about Defi is that it is a system where no one can intentionally manipulate the Compound.

The following is an alternative idea to the current proposal.

・A mechanism to conduct small-scale proposals through an off-chain.

・Include in the proposal the cost of the audit and an estimate of the cost in advance.

If the community agrees to the “Grants Program”, I will follow it.

4 Likes

Update - 3/8/11
Thank you to everyone who weighed-in on the proposal! The community gave a lot of really good suggestions, a good portion of which we will incorporate into the final proposal. We’ve made the following updates after listening to the community:

Proposal updates

  • Committee membership: to improve the technical sophistication of the committee, we’ve replaced Ken with Nick Cannon from Gauntlet. Gauntlet has successfully passed several technical updates to the protocol. The committee will benefit from having them onboard. We’ve also swapped in Min Teo from ConsenSys and Leighton Cusack from PoolTogether, who will be taking the place of Ryan and Sam (shout-out to @massnomis who posted the original idea for a grants committee). Playing committee Tetris wasn’t fun, but we think it was well-worth the effort — thanks to feedback from the community, the committee is way more well-balanced than before. The final committee will include:

    • Larry Sukernik (Sheepshead Bay, LLC)
    • Jesse Walden (Variant)
    • Getty Hill (Grapefruit Trading)
    • Monetsupply (Independent)
    • Aparna Krishnan (Opyn)
    • Leighton Cusack (PoolTogether)
    • Min Teo (ConsenSys)
    • Nick Cannon (Gauntlet)
  • Grants program amendments: during the community’s first Clubhouse call, @arr00 made a wonderful suggestion. If we need to make changes to the grants program after the proposal passes, it wouldn’t be practical to set up a formal governance proposal just for making small changes (e.g., swapping committee members around). Instead, @arr00 proposed the grants program set up a Snapshot page for the community to vote on amendments to the program. Setting up the Snapshot will require writing a bit of code (querying the amount of COMP delegated to addresses that vote). If the proposal passes, we will provide a grant to any community member who can help write this code and set up the Snapshot! While we’re on the topic of the grants program structure, we’d like to remind everyone that none of the grants program committee members are being elected into permanent roles. Once the pilot program wraps up in six months, the committee will be disbanded, allowing any community member(s) to set up a new committee with a fresh slate of members.

We will be posting the proposal on-chain tomorrow from Gauntlet’s address, 0x6626593c237f530d15ae9980a95ef938ac15c35c. Why are we using Gauntlet’s address? Gauntlet has ~126k COMP delegated to them, allowing us to more quickly post the proposal (as a reminder, Compound governance requires addresses to have 100k COMP to make a proposal). We’re excited about providing grants to the Compound community, so the quicker we can start, the better!

If you are in favor of the proposal, please vote once it’s posted on-chain! (You will see the proposal tomorrow at compound.finance/governance/proposals).

9 Likes

@sukernik this is great; can you post the address of the multi-sig, and have the members announce their addresses that they will be using (using some social channel, even Discord)? It is a good diligence step to ensure that COMP gets sent to the right place, and that it’s usable / without issues.

4 Likes
3 Likes

@sukernik Thanks for the detailed responses and addressing the concerns of the community. I’m excited to see this as a proposal.

3 Likes

Speaking from experience of having been an undergrad who applied to several EF + Maker grants any money makes a difference. What makes a bigger difference is the community and support. In fact, a big reason of why I got into DeFi is the Maker grant program and the support from the other people who were building things through the grant program at the same time! If we can create a community for the people who are building things in the grant program, I think that can go a long way! :smiley:

3 Likes

The goal should be to make it as accessible as possible to people who are in need of it!

That said, I think this is significantly quicker than other grants or incubators that I’ve applied to in the past and if we can actually have decisions within 10 days or less that’s brilliant!

For more context, as a scrappy first time founder I applied to several grants and incubators. Any money when you are small makes a difference and goes a long way. 2 * $10k = $20k enough to cover cost of a simple audit.

Grants:
EF grant application took months. The Maker grant process took 3-4 weeks. Non crypto grant programs required connections. Some university based grant programs took months. The Dfinity grant program took months. I still applied to every single one of those. $10k across several grant programs starts to add up!

Incubators:
YC takes a couple months of work for $125k and 7%? (not sure if this has changed since).

I still applied to all those grants and incubators because the application process (especially for YC) very clearly helped me think through parts of the business that I hadn’t previously. They forced me to crystallize my ideas.

The most important thing in the process is to make sure the grant program is tailored towards helping builders succeed. I think people will invest time into the process if they are able to see how the application itself helps them succeed.

5 Likes

Great suggestion.

Multisig address: 0xF1D8c2eED95D5fC2EaDe4E6Bb15a5969453E89a9

Multisig members announcing their address

4 Likes

As well as thinking this is a good idea, I’ll just throw in the newbie question here:
What is the quorum threshold, and/or other criteria for success currently? (Mildly surprised I don’t see them on the vote page. I see the green bar, but I’m not sure that counts for anything)

1 Like

Tnx for answer, with your statements and little research I am totally support selected team and proposal.I hope this will accelerate the development of the Compound protocol and improve communication and align the interests of long-term users and early investors.

1 Like

Thanks for being in favor of the program!

The quorum threshold is 400k COMP (you can find more in-depth documentation here).

Just to walk you through the process in a bit more detail, the proposal currently has ~1.37mm votes. If that doesn’t change, the proposal should pass and be queued for a two-day timelock. After the timelock passes, the proposal can be implemented.

1 Like