Research Summary: A Decision Model for Decentralized Autonomous Organization Platform Selection: Three Industry Case Studies

Governance and Coordination
1

TLDR

  • A growing number of Decentralized Autonomous Organization (DAO) platforms, such as Colony and Aragon, have emerged and drastically simplified the process of bootstrapping a DAO.
  • The choice between them is a selection problem which requires a balance of a number of criteria such as governance, developer support, and the tools offered.
  • This study builds a decision model for the DAO platform selection problem by considering 82 criteria and 28 alternatives.
  • The study examined three cases. Each considered the following features a priority: infrastructure decentralization, being on-chain, upgradeable contracts, token-based voting, a transparency portal, fund allocation, scalability, upgradability, reputation-based voting, the ability to upgrade governance, extensibility, permissionlessness, shared resources, and proposals.
  • The first case study focused on dOrg. They are a decentralized autonomous collective that provides DAOs with product development and operational services. They build DAOs without a centralized middleman mediating between freelancers and employers.
  • dOrg states they use the DAOStack platform to develop DAOs for their clients. In this case, the Decision Support System (DSS) suggested Colony was the most feasible platform, followed by Aragon and DAOStack.
  • SecureSECO project was the focus of the second case study. They are a collaboration between five companies and five universities that aims to make the worldwide software ecosystem safer. Its community is managed by a DAO, which decides a trust calculation mechanism.
  • SecureSECO states that Aragon and Colony were their top potential DAO platforms. The DSS suggested that Colony was the most feasible platform. Aragon, DAOStack, MakerDAO, MolochDAO, and Kleros were scored as the second to sixth potential solutions respectively.
  • Aratoo was the subject of the third case study. They are developing a noncustodial DeFi wallet for managing cryptocurrencies. Their organization needs a DAO for protocol governance, value accrual, and voting implementation.
  • At first, the DSS suggested infeasible solutions, so the authors adjusted some hard constraints into soft constraints so that they could receive a solution. The DSS suggested that Colony was the most feasible platform. Aragon, and DAOStack were scored as the second and third potential solutions respectively.

Core Research Question

  • Can the proposed decision-making model solve the selection problem when choosing between DAO platforms?

Citation

  • E. Baninemeh, S. Farshidi, & S. Jansen. A Decision Model for Decentralized Autonomous Organization Platform Selection: Three Industry Case Studies. 2021. Retrieved from https://arxiv.org/pdf/2107.14093.pdf.

Background

  • Decentralized Autonomous Organization (DAO): An organization form that takes advantage of blockchain as its based technology, partially or thoroughly. It is usually characterized by the spirit of democracy and decentralized governance.
  • DAO platforms: In order to reduce the heavy and complicated process of constructing blockchains or peripheral programs, as a BaaS service, DAO platforms provide DAOs with various functions modulus that enables launching DAOs to become easier and cheaper considerably.
  • Multi-Criteria Decision Making (MCDM) problem: A problem solving-method that considers a set of decision criteria and evaluates a set of alternatives to make a decision. (Farshidi, 2020).
  • ISO/IEC 25010 standard: A product quality model to evaluate software products. The product quality model defined in ISO/IEC 25010 comprises the eight quality characteristics: functional suitability, performance efficiency, compatibility, usability, reliability, security, maintainability, and portability.
  • The Blockchain technology stack: The Internet Layer; The Blockchain Layer; The Application Layer, including DApps, DAO Layer 1, and DAO Layer 2. (Machart, 2020) DAO Layer 1 refers to “independent DAO Platforms that are no-code and provide community coordination tools” (such as Aragon, DAOstack, DAOhaus, and Colony). They lessen the technical ability requirements for DAO users. DAO Layer 2 refers to “Services offered by DAO platforms”.

364×629 50.3 KB

(Baninemeh et al., 2021)

Summary

  • Main Research Question (MRQ): “How can knowledge regarding DAO platforms be captured and organized systematically to support decentralized autonomous organizations with the decision-making process?” Regarding MRQ, the authors collected five categories of knowledge that corresponded to their research questions. The categories of knowledge include DAO platforms, DAO features, mapping among the DAO features and quality attributes, Quality Attributes, and mapping among the DAO features and the DAO platforms.
  • RQ1: “Which DAO concepts should be considered as DAO features in the decision model?” Regarding RQ1, the authors raised 77 Boolean and 5 non-Boolean DAO features (available and accessible on https://dss-mcdm.com/).
  • RQ2: “Which DAO platforms should be considered in the decision model?” Regarding RQ2, based on literature and expert interviews, the authors ended up with 28 DAO platforms.
  • RQ3: “Which software quality attributes can be used to evaluate DAO platforms?” Regarding RQ3, the authors investigated DAO features with the ISO/IEC 25010 standard and extended ISO/IEC 9126 standard as two domain-independent quality models.
  • RQ4: “What are the impacts of DAO features on the quality attributes of DAO platforms?” (Baninemeh et al., 2021) Regarding RQ4, a Boolean adjacency matrix (Qualities × Features → Boolean) is applied, where DAO features can be part of many quality attributes, including Functional appropriateness, Operability, Interoperability, Functional correctness, Ownership attributes, and Functional completeness.
  • RQ5: “Which DAO platforms currently support the DAO features?” (Baninemeh et al., 2021) . Regarding RQ5, the first table below shows the Boolean Features (FeatureB), DAO Platforms (Platforms), and the “BFP” mapping. The second table below shows the NFP mapping between the Non-Boolean DAO Features and Platforms.

840×1128 240 KB

(Baninemeh et al., 2021)

838×1148 235 KB

(Baninemeh et al., 2021)

806×990 226 KB

(Baninemeh et al., 2021)

  • DAO Feature Requirements: A set of weights (WMoSCoW = {wMust, wShould, wCould, wWon’t}) according to the definition of the MoSCoW prioritization technique (Consortium et al., 2014) is applied for decision-makers to prioritize their DAO feature requirements.
  • The authors evaluated the decision model’s efficiency and effectiveness by case studies from three industries: 1) Web3 development - dOrg; 2) Open-source software security - SecureSECO; 3) Decentralized finance (DeFi) - Aratoo.

Method

  • The authors used a mixed-methods approach to understand and inform the development of their decision model.
  • The Multi-Criteria Decision Making (MCDM) analysis approach for the DAO platform selection problem: It receives platforms and their features as input, then applies a weighting method to prioritize features by the decision-makers’ definition, and finally gives a rank and solutions as output.

988×501 99.6 KB

(Baninemeh et al., 2021)

  • The MCDM approach above is based on a framework (Farshidi et al., 2018) that implements a Decision Support System (DSS) (Farshidi et al., 2018) for MCDM problems. It follows the six-step decision-making process (Majumder, 2015): 1) Identifying the objective; 2) Selection of features; 3) Selection of alternatives; 4) Selection of weighing method; 5) Applying the method of aggregation; 6) Decision making.
  • Expert Interviews: The primary source used to build the decision model. The authors contacted ten domain experts ranging from blockchain and DAO experts to software architects.
  • Document Analysis: The authors’ source for identifying features from each of the platforms. Composed of around 59% web pages, 11% peer-reviewed articles, 16% documentation of the platforms. The remaining 14% are a collection of videos, white papers, forum discussions, and books. These sources span the early years of DAO concept (2014) to the time the research was made (2021).
  • The steps for conducting case studies: 1) Requirements elicitation; 2) Results and recommendations; 3) Analysis.

Results

  • The table below shows features requirements of the three case studies:

884×917 291 KB

(Baninemeh et al., 2021)

  • The table below shows DSS solutions of the three case studies.

697×521 70.9 KB

(Baninemeh et al., 2021)

  • The decision support system (DSS) suggests Colony, Aragon, and DAOStack as feasible solutions for the three case studies examined by the authors because they each support all “must-have” priority features.
  • The results seem reasonable as they all are in the top-5 list of popular solutions in the market (According to Table 7 provided in the paper, the top-5 list of popular solutions are Aragon, Colony, MakerDAO, DAOStack, and Tezos) and have relatively high maturity levels.
  • All three of the case studies assigned priorities to infrastructure decentralization, on-chain, upgradeable contract, token-based voting, transparency portal, funds allocation, scalability, upgradability, reputation-based voting, governance upgrade, extensibility, permissionless, shared resource, and proposals.
  • Flexibility on feature requirements generated more alternative solutions.
  • The case study participants confirmed the effectiveness of the DSS. They stated that it was useful for generating a shortlist of feasible DAO platforms. It saved time and costs, letting more detailed feature requirements be met. It was also helpful for finding their primary concerns.

Discussion and Key Takeaways

  • Subjectivity needs to be considered. However, a DSS promotes objectivity and dismisses subjectivity.
  • Security issues on a DAO are crucial, including cybersecurity, voting procedure, and voter manipulation. The immutable nature of blockchain also leads to important DAO security issues, particularly, it would lead to a supreme difficulty to modify a bug in code, which is fatal when a DAO is under hack attack.
  • MCDM approaches can address decision-makers’ concerns and priorities. Benchmarking and statistical analysis give generic results and comparisons without considering decision-makers’ preferences.
  • Within a range of MCDM approaches, the proposed decision model has a lower cost than alternatives given the number of criteria and alternatives.

Implications and Follow-ups

  • The authors proposed a website (https://dss-mcdm.com/) in the study, which is running and keeping its knowledge base up-to-date and validated. They are considering building a community that will maintain and curate the system’s knowledge base.
  • The authors built six decision models based on the framework they created to model the selection, including database management systems, cloud service providers, blockchain platforms, software architecture patterns, model-driven platforms, and programming languages. They confirmed the effectiveness and usefulness of the DSS in addressing MCDM problems.
  • The authors plan to work on building trustworthy decision models to address the consensus algorithm selection problem and the self-sovereign identity framework selection problem in the future.

Applicability

  • The research provides a decision model to evaluate DAO platforms in the market, rapidly giving systematic directions for DAO platform selection.
  • More platforms and features can be added to the decision model.
9 Likes
2

@Astrid_CH thanks so much for this incredible comparison – before we launch into the rest of our questions, I’d love to know what you got out of doing this series of DAO-related articles, what was the most useful information you gleaned from doing this? I’m also curious about some of the more informal DAO platforms, how does something like Aragon compare to Mirror.xyz for example?

3 Likes
3

@jmcgirk Thanks for your questions and mentioning informal DAO platforms like Mirror.xyz, discussing various platforms and tools always brings us closer to the panorama. Comparing these platforms allows me to understand the inclination of the DAO market more, the most useful information I got from these papers is that I found the most popular and some useful features of DAOs, as they are partially manifested by platforms’ popularity. The tables and case studies listed some features and their corresponding concerns, which is crucial for DAO’s participants. Furthermore, I can use these features to design my new DAO project, the process of researching these features pushed me to consider more possibilities and stimulated my creativity. I think Mirror.xyz is a writer/creator-friendly platform, besides its functions for writing, it provides DAOs (that may be built on DAO platforms, such as Aragon) with a well stretch of fundraising. While DAO platforms provide more complicated organizational structures build blocks. I think they could be complementary to each other. However, the amazing gas fee could be a barrier for some of them to thrive.

4 Likes
4

Hi Astrid, I know this is a while ago, but the work is highly related to my current work. Do you know where I can find the specific definitions of all the features of DAO they mentioned in the paper? I was able to track to this webpage (Decision Model Studio), but all these features (like political decentralization; infrastructure decentralization) need clarification and definition on my perspective, especially how they collect the data to decide whether it has this feature or not. Please let me know if you have access to this resources, or have any connection with the original authors, thank you! @Astrid_CH

4 Likes
5

In ‘laywoman’ terms, this paper centres around a methodical way to select a DAO platform based on specific criteria and demonstrate application using three case studies. Am I correct? @Astrid_CH?

While there were several features/criteria discussed, governance stood out for me. If funding is the lifeblood of a DAO, it is probably safe to say governance is the wheel on which DAOs run. This makes me wonder, among the different criteria that the authors discuss in their paper, do they mention participatory governance? Considering how important this criterion is, do all the top 5 DAOs listed have features that enable participatory governance for DAOs? No doubt, how active (well-oiled) a DAO is will determine its progress, considering the countless inactive (rusty) DAOs.

In this comment, I attempt to expand on features to be considered when choosing a DAO platform by zeroing in on participatory governance in DAOs. I start off with my definition of a DAO and participatory governance. I briefly talk about different governance models and the type adopted by the top 5 highlighted by the authors. Seeding further conversation, I conclude by identifying more areas to explore relating to DAO platform features and participatory governance.

What is a DAO?

Adding to that already said by the Authors, @Astrid_CH and many others, I’ll still offer my definition of a DAO. Decentralized Autonomous Organizations are digital-native organizations formed by entities (natural and artificial) having a common goal, with funding from a common purse reaching decisions through participatory governance. More on participatory governance soon.

DAO governance models

To capture how vital governance is to DAOs, one could say DAOs exist primarily for their governance. This is backed by the fact that early DAOs were actually formed to govern how to spend funds on investments. DAOs often adopt a multi-layered governance model that involves:

1. Discourse/proposals: DAO members post their proposals on places like forum discourse, discord, telegram, Reddit, etc. Community members often begin governance at this stage by debating or discussing the topics raised.
2. Off-chain Voting: Leveraging off-chain trackable metrics like likes, replies, reactions, etc, members show their position on the matter through off-chain voting.
3. On-chain voting: This is the stage that permanently records the final decision, often also the execution stage.

What is Participatory governance?

At the risk of sounding like Abraham Lincoln, former President of the United States, participatory governance to me is governance of the governed by the governed. This governance is governance done from an informed state.

What’s the level of participatory governance in the top 5 DAOs?

The findings from the authors’ research show us that platforms have 100% coverage for the feature of an active community, but have 46.43% coverage for the feature of Direct DAO governance. This can only mean the level of participatory governance in DAOs is low for the top 5 DAOs. Yet, society is quick to campaign for democracy, conveniently ignoring the risk of the dictatorship of the majority.

Some available alternatives include quadratic voting. They have already successfully trialled this in Colorado in an experiment led by Glen Weyl. And currently successfully utilized as quadratic funding in Gitcoin’s grant rounds.

There exist some challenges with adopting these governance models. One future research point I wish to highlight is the incorporation of a decentralized dispute resolution system in existing and future DAO governance models. Traditional dispute resolution systems are obviously not uniquely suited to efficiently handling DAO-native disputes.

6 Likes
6

A Decision Model for Decentralized Autonomous Organization Platform Selection

What the available platforms offer is a a new way to create and manage companies, organizations, and projects without a centralized authority. As a result, DAO platforms are the newest vehicle for a variety of operations, including economic and social experimentation, the pursuit of new organizational structures, and the implementation of governance solutions. It is important that the right DAO platform is selected, because it will be the backbone of the company, organization, or project.

Creating and managing a company, organization, or project without a centralized authority is possible The Blockchain industry is booming and seems to be on the cusp of a technological revolution. With so many different decentralized autonomous organization platforms, it can be difficult to decide where to invest your time and resources. There are some important considerations when making this decision including the cost of development and the scalability of the platform.

Blockchain development has become cheaper and more efficient in the past few years, but are DAO able to release funds, invest capital, and distribute profits to shareholders automatically. The DAO is an interesting new way for organizations to be managed by a smart contract.

Blockchain is disrupting nearly every industry, changing the way people and businesses approach data, information, and transactions. This is especially the case for centralized organizations that have to deal with a governing body or a top-down hierarchy. The most obvious example of this is the banking industry, which is moving to a decentralized system with blockchain.

The new way of doing business, open-source, and also autonomous why does anyone need to make a decision about which D.A.O. platform to use? The answer is that if you are going to be voting on anything in the life of your organization, you will need the platform which makes the voting process most straightforward. You have to have a place to vote on it, and then you have to have a way for the vote to be counted.

One of the most popular D.A.O. platforms is Aragon. This D.A.O. platform is built for companies, and it includes features for governance, fundraising, and accounting
There are many different decentralized autonomous organization platforms, here are some of the factors to consider when choosing:

-Are there any developers in your team?
-Do you have a preference for a type of platform?
-Are you looking for a platform with a robust framework?
-Are you looking for a platform with a strong developer community?

The best decentralized autonomous organization platform can be difficult to select because it is difficult to compare all the platforms, I believe that the best way to select the best platform is to start with a decision model and briefly discuss the decisions for decentralized autonomous organization platform selection.

-Determine your motivations and goals

-Select the business needs of the organization

-Identify the organizational goals

-Identify the technology requirements

-Identify the levels of decentralization desired

-Identify the desired features

4 Likes