AIP-1: Arbitrum Improvement Proposal Framework

(Posting on behalf of TreasureDAO as a member of its DAO council - Karel Vuong, Co-Founder.)

Summary

TreasureDAO, on behalf of its delegates, is voting AGAINST AIP-1 in its current form. Treasure asks that Arbitrum resubmit this proposal with (i) greater clarity over the purview of the special grants program, (ii) breaking AIP-1 into several proposals rather than one.


Rationale

As a DAO, we have learned firsthand how difficult it is to maintain a fast-paced organization while upholding decentralized governance. We have empathy for other builders in this position, who must juggle their operational plans and community input at the same time. It is an arduous, unenviable model to which builders in web3 remain committed to in order to facilitate the decentralization of technology.

We do not find AIP-1 to be inherently problematic in terms of the token allocations. The ArbitrumDAO should ensure ample funds are allocated to mission critical programs, and the Arbitrum Foundation team supported by Arbitrum’s original creators, Offchain Labs, is in the best position to evaluate what the ecosystem needs at a core protocol level to continue growing. However, we agree with the voices in our community requesting that more detail be provided regarding the special grants program. Early proposals for ArbitrumDAO will set best practices for future governance. Providing transparency into how funds will be allocated, particularly for a budget of this size, will set a positive norm for the DAO’s own grant administration through the treasury. The Foundation should set an example for the community on what appropriate grant administration looks like so that the treasury funds are put to best use.

Moreover, AIP-1 is multiple proposals packaged into one and is currently being decided by a razor-thin margin. A proposal as multifaceted as this one should only be enacted with overwhelming support from the community. It sets a bad precedent for the DAO to normalize proposals containing rider conditions. Governance should be nuanced and granular.

We discuss both arguments in detail below and conclude with quotes from our community members that helped inform our decision-making.

(i) Special Grants

Details into this grants program are vaguer than we would have hoped. Treasure evaluated this proposal by the standards with which we try to conduct our own internal governance and found the explanation to be lacking. It is unclear if our community would have approved this proposal if submitted in the context of TreasureDAO. As much as we support Arbitrum and appreciate the ongoing help they provide our team, we have a duty as a delegate to evaluate this proposal by the standards that earned us trust among the ARB token holders.

The proposer references “voter fatigue” as the rationale. We agree with this argument and would add that it would be unwise for ARB stakeholders to exercise too much influence over core protocol decisions. The overall community does not have the same technical expertise as the Arbitrum team itself, and subjecting blockchain design to popular consensus would likely have bad results. Ethereum has eschewed design by committee for a reason.

Nevertheless, ARB is a governance token, and the credibility of ArbitrumDAO as a decentralized organization depends on governance proposals such as AIP-1 being fully transparent. We identified several problems with the presentation of the grants program in AIP-1:

  1. insufficient discussion of the kinds of grants that would be approved;
  2. missing explanation of vesting or other timelines, if any;
  3. insufficient explanation for how grants would be reported to the community; and
  4. no proposed accountability framework by which the community could evaluate the efficacy of the program over time.

We ask that the proposer resubmit the proposal with a detailed response to these questions. In particular, we would like to see more discussion about the purview of the grants program, which includes examples of the types of grants that would be approved or not approved.

(ii) Splitting into Separate Proposals

AIP-1 contains a lot of separate action items for the DAO to consider. It should be split into granular proposals with their own rationales and plans. The proposals should be broken into:

  • Security Council
  • Directors
  • Special Grants
  • Reimbursement for Foundation Setup Costs
  • Data Availability Committee

Allowing the DAO to vote on these issues separately is important to nuanced governance and also ensuring that a proposal is able to pass. Bunching too many items into a single proposal causes parties to vote “no” simply because one part may be untenable.

Arbitrum’s Response
We appreciate Arbitrum providing a detailed response to this issue. We agree with the points being made: setting up a DAO requires preliminary legwork that has to be done, and the initial setup cannot be logistically completed by a DAO.

We also agree that Arbitrum should have a large ecosystem fund available for completing special grants. The issue is not the size of this fund but rather transparency of the program. The response from the Arbitrum team is a justification of the size but does little to address the primary controversy. It also does not explain why formation of the DAO requires $3.5 million in reimbursements. Moreover, AIP-1 should be broken into several proposals. We do not think it sets a good model for future governance to have the first proposal enacted by the DAO include several rider provisions not germane to the central controversy here.


Community Voices

The Treasure community has been engaged in a passionate debate surrounding AIP-1 on Twitter and in our Discord channels. We wanted to acknowledge some key community input that informed our decision-making. We also would like to thank all of the community members that weighed in on this discussion and helped us reach a decision that we feel best represents the interests of the entire delegation.

Support for Grants

The Treasure community is in support of a grants program run by the team. Rahulio Jones commented,

“if the funds are used well, it will help Arbitrum accelerate and become dominant. Extend the lead over competitors. Use funds to attract and keep highly desired talent.”

Uncle Pennybags echoed this sentiment and compared the importance of ARB grants to Optimism’s program:

“Arbitrum grants will also be a pivotal moment for [the ecosystem] in general. Just like how OP eco was propelled by the generous OP emissions handed out to every defi protocol within the eco.”

DaddyDuckNugget agreed with Arbitrum’s reticence at turning mission critical questions fully over to the DAO:

“[It is] impossible to develop a project at the scale of ARB without a governing centralized body. [Full] DAOs are [incredibly] slow and at the whims of biased token holders who might not have the best interest at heart when it comes to governance. Decision-making requires leadership and ARB won’t be fully decentralized until the product is 100% complete.”

Transparency

Some critics like Chris Blec have accused the Arbitrum team of “stealing” funds from the DAO. The Treasure community strongly rejects this polemical framing. Concerns were more about the manner in which the proposal was presented.

CJ explained it well by saying:

“I don’t think any serious ecosystem participant is buying that [nefarious intent] narrative (though I’m sure it gets Twitter engagement - I think the focus here is on the additional disclosure and projections required that would accompany the funding (e.g., what are the criteria for the special grants programs, how large is the program as a proportion of the overall 750mm tokens, a breakdown of startup costs already paid vs. what is still outstanding, projections for admin and operational costs). Remember that a “no” vote can also mean “revise and table the proposal again.” Similar to how Treasure has done in the past when the community wants amendments and additional information.

He went on to suggest the following guidelines for a future proposal:

“[Arbitrum] should draft a full business plan and exhibit deck and let the community digest that to make an educated decision rather than asking for almost $1bn in funding over a weekend.”

NFDoggo agreed and said,

“best to do a mea culpa and reset.”

Some in the community added that a revision of AIP-1 should also have some explanation of why the ArbitrumDAO setup costs totaled $3.5mm. It is unclear why the fees were this high. Omitting a detailed breakdown of these costs would set a bad precedent for future governance.

We would like to note here that the TreasureDAO council does not fully agree with this community perspective. Expecting a DAO to provide a business plan that accounts for every line item of a program is not advisable for a few reasons. Holding a token does not automatically qualify an individual to weigh in on how much things should cost. We have noticed from our own experience that non-experts tend to grossly underestimate the cost of expenses like game development. Moreover, allowing the community to have input into the minutiae of program funding could have disastrous consequences if the DAO decides not to spend adequate funds on legal fees or administrative necessities. It is more appropriate to grant a full amount towards a particular program and outline categories of expenditures rather than each line item.

This approach is distinct from that established by AIP-1. The proposal provides the ecosystem fund with a blank check without a discussion of the kinds of programs that will be funded. A better approach is to request allocations from the ecosystem fund on a case by case basis or for categories of expenditure with a clear transparency framework. We also notice that AIP-1 omits any discussion of the timeframe over which the grants will be dispersed and adds that “further funding of the administrative budget wallet shall require approval of an AIP by the ArbitrumDAO pursuant to the AIP process.” Leaving the door open for future funding warrants an explanation for the length of time that Arbitrum expects the initial allocation of funds to cover.

TreasureDAO’s own model for its ecosystem fund is to submit specific proposals on new programs that should be funded using MAGIC. We understand that programs like “core protocol development” will require a sizable amount of funding. Too much input from the community would be inappropriate here. For example, the token holders are not in the best position to judge how much it costs to build a competitive L2 or what resources are required to do so. However, the need for technical expertise in certain areas of decision-making does not absolve Arbitrum from making an effort to explain the purview of the grants program in more detail than provided in AIP-1.

Moving Tokens

A debate took place on the proposal forum in which ARB token holders were asking why the Arbitrum team preemptively moved the 750mm tokens into a new wallet and may have been moving these tokens onto CEXs before the DAO vote went through.

NFDoggo made a compelling argument in the Treasure Discord that it is unreasonable to assume that ArbitrumDAO could have gotten off the ground without preliminary work. He said:

“Everyone seems to be jumping on the fact that certain decisions have already been made - which the Arb team directly addresses in the feedback, pointing to the ratification language. The team took certain actions in preparation of launching the foundation and a first step for the DAO is to simply have the members ratify those decisions. [It’s] just legal housekeeping.”

NFDoggo ends by saying that the correct perspective that the ArbitrumDAO should take now is to focus on the desired amendments to the proposal rather than pointing to necessary preliminary actions that were already taken. He also raises a good point that AIP-1 effectively decentralizes the Arbitrum ecosystem more than the status quo, which is inarguable.

Like NFDoggo, we do not think that movement of the tokens was done in bad faith. Establishing market makers before the token went live on CEXs was in the best interests of every ARB holder to ensure deep liquidity across exchanges.

16 Likes