Research Summary: What is a DAO? Conceptual Foundations

[danielo]

TLDR:

• DAOs have become a popular concept, albeit with divergent and often conflicting definitions.
• To address this concern, the authors reviewed the pre-history and early history of DAOs to frame central themes in understanding DAOs. We then identified five conceptual lenses that allow us to conceptualise the telos of DAOs;
• conducted a study with 155 participants to identify the key ‘qualities’ associated with DAOs;
• and arrived at a definition of DAOs that satisfies our research purposes.

Core Research Question

How can we conceptualise DAOs to map the common paths towards becoming a DAO and the problems that block that progress?

Citation

Ospina, D. and Bohle Carbonell, K. (2022). What is a DAO? Conceptual Foundations. Mirror. What is a DAO? Conceptual Foundations — Danielo

Summary

• DAOs have become a meme, one used to describe everything and nothing. We focus our research on elucidating the telos of DAOs
• Multiple ideas associated with DAOs have been present in narratives that preceded them.
• DAOs were originally conceived as non-humane organisations, but influential definitions of the early 2010s included a human element and emphasised censorship resistance. The definition has continued to evolve and diverge since.
• Across different definitions, DAOs have come to be associated with a broader exploration of coordination between (autonomous) agents who seek to satisfy certain needs and aspirations.
• Fluid boundaries mean DAOs fit poorly within the definition of Organisations, but a newer concept, Organisationality, proves more appropriate. Organisationality frames three key criteria: characterised by interconnected instances of decision-making, these instances of decision-making are attributed to a collective entity or actor, and collective identity is accomplished through speech acts that aim to delineate what the entity or actor is or does.
• There’s a set of ideas and values that further qualify DAOs; we’ll refer to these as the Ethos of DAOs.
• DAOs exist both as entities at a specific point in time and as a constant process of change. DAO are both entity and process.
• This dual nature (entity and process) poses a complication to conceptualising DAOs.
• By framing DAOs as Communication, we can study the patterns of communication in a collective and elucidate both the entity and process dimensions.
• DAOs, as communication networks, are subject to the principles of complex systems. Ethos directly shapes the emergent patterns and structures that we see in DAOs.
• A Pol.is survey resulted in 4 statements about what characterises DAOs (Ethos of DAOs) with 64% or more participants agreeing:

1. Decentralised power: no single source of authority.
2. Autonomous: self-sovereign, not bound to an external coercive force.
3. A common goal, vision or set of values that are (being) worked towards.
4. A shared treasury controlled by a decentralised voting mechanism.

• We conclude that DAOs are collectives that exhibit organisationality, expressed and evolved through communication events and processes, and shaped by an Ethos that highlights the 4 qualities mentioned above.

Method

We conducted a literature review, facilitated workshops with 15 participants (including inviting 4 professors to input into the conceptual development), and conducted a Pol.is Survey.

The survey was distributed through our networks on Twitter and LinkedIn as well as 20 other DAO-related communities. 155 participants provided 31 statements and cast 1,829 votes to the question "What makes a DAO a DAO?”.

Results

A majority grouping (97 participants) converged on 5 key statements with 64% or more agreement (2 of said statements we identified as permutations of the same concept and condensed into 1).
See Pol.is report here

And full paper here

Discussion and Key Takeaways

We conclude that while DAOs are not organisations (in a traditional sense), they are collectives exhibiting some degree of organisationality, enacted through communication events and processes.

In addition to organisationality, we frame DAOs a both static entities while simultaneously evolving and striving to uphold certain values (an Ethos). Conceptualising DAOs as communication networks enables us to resolve the tension between the current entity and its evolution (between what is and what is yet to be).

Finally, we identified a generally shared Ethos of DAOs that highlights decentralisation of power, autonomy, a common goals, vision or set of values that are (being) worked towards, and a shared treasury controlled by a decentralised mechanism.

Implications and Follow-ups

This study provides a conceptual foundation to discuss DAOs and bridge previous research on organisational studies and related fields.

Further work is needed to delineate the limitations of the proposed conceptualisation of DAOs when it comes to applying insights from previous research that didn’t take into account the conceptual foundations (and distinctions) hereby identified.

Applicability

This work is currently being applied to create a tool to map, assess, and plan DAOs. Additionally, this research is being referenced to conceptualise and develop assessments of DAO Community Health.

Other applications are yet to be explored, and we’re happy to engage and collaborate on their development (contact us through @rndao__)

[kingdamieth]

SourceCred is a DAO & an open-source community tool to measure & reward value creation. Can a DAO die? if yes what can result to the death of a DAO? if no then what happened to sourcecred?

[danielo]

Interesting question. From the definition we have, I guess that as long as there’s communication happening, decisions are being made and said decisions/acts are attributed to SourceCred, we still have some degree of organisationality, so technically SourceCred is still ‘alive’ as an organisational collective. Then, whether the above aligns with the Ethos of DAOs (or how much) will determine its degree of DAOness

[Mr.Nobody]

@kingdamieth

Can a DAO die? if yes what can result to the death of a DAO? if no then what happened to sourcecred?

Wait, what happened to SourceCred? It doesn’t seem dead.

[Tolulope]

Thank you @danielo for this timely summary.

We recently discussed DAOs and legal wrapping on the chat so your research summary is just in time. Following your takeaway that DAOs are not organisations in the traditional sense, do you have any thoughts on DAOs attempting to gain legal personhood?

Do you think the attempt to give legal personality to a DAO still allows it to remain a DAO, especially in line with the principle of decentralisation in the DAO ethos your survey posits?

Also, for DAOs that are incorporated, sometimes, the incorporation tends to disenfranchise some of the DAO members who are outside the jurisdiction of the regulatory authority. Would you agree that this is an infringement of DAO ethos?

[kingdamieth]

I clearly understand you and thanks for your conclusion with this

@danielo

Then, whether the above aligns with the Ethos of DAOs (or how much) will determine its degree of DAOness

I think their level of organisationality is Low compared to the past and looking at this from my own perspective, it might affect the level of technicality of their product.

[danielo]

It’s not dead but the ‘organisation’ stopped operating in the way it was (no further attempts to pay contributors, anyone who contributes does so as part of an open-source project) and they’re maintaining Twitter and doing some minimal discord moderation.
You could argue that it’s still a DAO except for the ‘shared treasury controlled by a decentralised voting mechanism’ part of the Ethos

[danielo]

IMO, ideally the whole legal wrappers can be avoided/reinvented, as whether DAO-Ethos-aligned or not, nation-state law is a problematic legacy system. In practice, each individual and collective will have to negotiate their way through it.

I see legal wrappers as a necessary evil for limited liability (given the legal uncertainties of the industry) but in an ideal world we can stop giving personhood to legal entities (it is suicidal that fictional entities can have rights like humans and more rights than say animals or the biosphere at large).

Also, since DAOs are born global, it’s weird to attach them to the local (a nation state). But again, we haven’t got the right structures in place here.

A legal wrapper does put the DAO under the authority of a Nation State (it’s debatable whether without one the DAO is still controlled by a nation state or several as the individuals still are exposed to it/them.) So the Autonomous part is put into question here.

Then, if the legal wrapper allows for the decentralisation of power (eg permissionless joining, decentralised governance, etc.) it might not be fully anti-DAO. So we can still try to apply (some) DAO ideals to it.

But then, for me, the goal is not to be more DAO, the goal is better human coordination. As long as we don’t call everything that’s not a DAO a DAO, then we can still have a shared language to advance the discussion. And that’s the contribution this piece aims to make

[danielo]

I would caution against defining that too black and white. Indeed most DAOs are quite chaotic but they’re also very early stage experiments on a new form of coordination. And six months or 2 years from now we could be looking at a very different picture.
Already there are some more ‘organised’ outliers

[s_ben]

@Mr.Nobody

@kingdamieth

Can a DAO die? if yes what can result to the death of a DAO? if no then what happened to sourcecred?

Wait, what happened to SourceCred? It doesn’t seem dead.

As a long-time SourceCred (SC) contributor dedicated to keeping the dream alive, my response here is biased, but hopefully useful in testing some of the conceptualizations presented.

@danielo

Interesting question. From the definition we have, I guess that as long as there’s communication happening, decisions are being made and said decisions/acts are attributed to SourceCred, we still have some degree of organisationality, so technically SourceCred is still ‘alive’ as an organisational collective.

This definition feels validating, as it articulates a vague sense I’ve had since SourceCred (the organization) began winding down a couple months ago. While the number of contributors has declined, there still exists a core group that have continued to meet and make decisions (e.g. the community call is still happening every Wed), engaging on Twitter (including Twitter spaces), participating in convos on discords, forum threads like this one, etc.

There are also presumably lots of conversations about SourceCred I’m not aware of. SC has ‘broken through’ as a ‘brand’, reaching relatively broad awareness. I’ve heard stories of people knowing about SC in non-crypto/web3 OSS projects, non-tech people, etc. It’s also being actively used by a decent number of projects. Some of which are deeply invested at this point. For instance, MetaGame, the first community to integrate SourceCred, has expressed interest in being ‘one of the SourceCred stewards’.

I should also add here that the SourceCred community decided to release the SourceCred trademark to the public nearly a year ago, in the aftermath of an unsuccessful hard fork. And that the code is MIT licenced. So there doesn’t appear to be a legal barrier to ‘being’ SourceCred (I am not a lawyer this is not legal advice).

@danielo

It’s not dead but the ‘organisation’ stopped operating in the way it was (no further attempts to pay contributors, anyone who contributes does so as part of an open-source project) and they’re maintaining Twitter and doing some minimal discord moderation.
You could argue that it’s still a DAO except for the ‘shared treasury controlled by a decentralised voting mechanism’ part of the Ethos

Basically agree, but if we zoom out, things get more complicated? For one, a number of projects are still distributing meaningful amounts of money using SourceCred. And SourceCred contributors can earn rewards in those instances, often by supporting said SC instances. For instance, I’m earning from at least five instances: Maker, MetaGame, Giveth, Token Engineering Commons (TEC), and (if eligible) SCRF.

SCRF is thinking about potentially increasing payouts. Would a core SC contributor (me) writing a long post (this one) increase SCRF’s confidence in their ability to maintain their SC instance, leading to a greater probability of increased payouts? I’m not writing this post for the money (at least consciously). But feasibly I am coordinating with organizationality with other SC contributors around SCRF (talked about SCRF at last SC community call, have RT’d SCRF Tweets about SC, etc.). May RT this post from the SC Twitter if people are cool with it.

Additionally, on a purely technical level, likes on this post will increase my Cred score, giving me more DAI and giving my likes greater influence in the Cred graph.

So technically, I am directing some non-zero amount of SCRF’s budget with my words It’s a stretch, but if we expand our definition of ‘shared treasury’ to all SC budgets in the ecosystem, and ‘decentralized voting mechanism’ to likes (each post is a mini election), SC meets this requirement in a diffuse, fuzzy way?

@kingdamieth

I think their level of organisationality is Low compared to the past and looking at this from my own perspective, it might affect the level of technicality of their product.

Agreed. We’ve got a pool of devs that have contributed in the past, some knowledgable devs supporting instances, and our recently departed lead dev is around to review the occasional PR. But active development is halted for now. I think this is OK for the near to medium term, as the core components (CredRank + main plugins) are relatively mature and hardened. Which I go into in my recent talk at MetaFest (which for disclosure I earned Cred/XP → SEED). The basic functionality of the product hasn’t changed really in ~2 yrs. In that time numerous projects have battle tested a good chunk of the solution space, found bugs, built features, etc. I think it’s relatively low risk to explore if using existing plugins and not building novel functionality or radically new use cases. However all codebases do require at bare minimum basic maintenance over time, in particular the data fetchers calling external APIs. We shall see

I will throw out an interesting possibility: that SourceCred is already a DAO, in the Bitcoin sense. A decentralized protocol. While Bitcoin gave censorship-resistant value transfer, SC gives (if governed correctly) censorship-resistant value reward. Monetary sovereignty → Valuation sovereignty. This is actually how I’ve viewed SC since the beginning, and perhaps why my view has often diverged. Where others see a failing SaaS startup and leave, I see a hardening protocol giving me increasing SC income, expanding possibilities for projects needing stability. If curious, below is an early more creative expression of that vision, where I explore SC as a 'Community-Store-of-Value (CSoV).

SourceCred as Store-of-Community Value (SoCV)

I realize this may be a stretch conceptually, but AI (and PageRank is considered pseudo-AI by some), has a way of reifying concepts And the possibility of an anti-fragile income via SC is very motivating for me.

DALLE-2 ‘group of researchers sitting in a circle watching a field of possibilities erupting from black hole’ + SC logo

Phew! Ok sorry for long thread, lot to say on this apparently

Really liking the proposed DAO conceptualization as well. Resonates with my experience in DAOs and feels like the most useful definition I’ve seen to date.

[danielo]

I’d propose a distinction between the output of an organisation/ organisational collective (e.g. the SourceCred protocol/product) and the collective itself. e.g. a hammer can keep existing even if the group that manufactured hammers dissipates. And hammer owners can keep talking about hammers and using hammers but that doesn’t imply a Hammers DAO exists.

So while SCRF, MetaGame, etc. can continue using the SourceCred protocol, the SourceCred org could be alive or dead.

In this case, to determine whether SOurceCRed is still a DAO, one key criterion of the organisational collective definition is the shared instances of decision making (and the attribution of said instances to the collective entity).
Applying this, SCRF using SourceCred does not entail any collective decision of SourceCred users at large, nor is the way SCRF is using the protocol attributed as a decision taken by SourceCred itself.

I understand there is still a Github repo with centralised control i.e. one or a few people decide whether to merge open source contributions (not certain this is the case but for the sake of argument I’m running with it).

If decisions are made that affect the collective (i.e. the code used by the different SourceCred users), then there are collective instances of decision making, which would validate SourceCred as an Organisational Collective.

But is it a DAO?
If there’s an aspiration to decentralise power in managing the github repo, then we could probably say that SourceCred has some degree of DAOness (decentralised contributions and an aspiration to decentralise approval of which ones get made “official”). But if the ability to make contributions to the centralised repo ceased and there was no intention to decentralise power, then it wouldn’t be a DAO.

[s_ben]

@danielo

I understand there is still a Github repo with centralised control i.e. one or a few people decide whether to merge open source contributions (not certain this is the case but for the sake of argument I’m running with it).

There are I believe like a half dozen owners of the repo. Can’t verify, as GitHub doesn’t make that info available to regular contributors like myself. But iirc it’s basically the main devs over the course of the project.

@danielo

If there’s an aspiration to decentralise power in managing the github repo, then we could probably say that SourceCred has some degree of DAOness (decentralised contributions and an aspiration to decentralise approval of which ones get made “official”).

A couple months(ish) back, when Thena was making their exit, I made it a point to raise this issue in a governance call (who controls GH access), and pushed for rough consensus. Note: there were more formal governance mechanisms in place, but their legitimacy was arguably questionable at that point, and in any case we lacked the bandwidth for formal governance. Rough consensus was the most legitimacy we were going to get. Thena agreed to give ownership to whoever the community decided on. I raised hammad (METADREAMER) as the only candidate I could think of with the ability and potential interest. Those on the call agreed he would be acceptable. Thena made METADREAMER an owner on the call. We discussed how this could enable a ‘coup’ scenario, where any one of the owners could revoke permissions of the other owners. We agreed that wasn’t ideal, but that it was better to decentralize as much as possible. So arguably SC has aspirations to decentralize power. If not the means. Indeed, all formal governance has ceased for now.

I should note that last I heard METADREAMER is working on MetaCred, which broke off from SourceCred nearly a year ago to pursue its own, more crypto-native vision of the product. But these days appears to just be integrating SC output with Coordinape and other reputation systems to create a “pattern language”. Note: I’m not a part of this project (which is not public) and this may be out of date .

There are also 122 forks of /sourcecred. While there may or may not be any forks being actively developed, the ability to fork definitely shapes the decision landscape.

Could someone fork SC and hire devs in Bangladesh on Upwork to update the data scrapers and patch the occasional security fix for a couple grand a year?

For what it’s worth, I also think non-code contributions are important here. The ability to change the code is the clearest ‘line in the sand’. Without that it’s hard to argue SC can “manufacture the hammers”. However, decisions around comms are impactful as well. For instance, what @sourcecred tweets to its 3,621 followers. Or the decision I’m making now to share a bunch of knowledge and information in this post. To what extent are comms deciding how people spend their valuable attention? Perhaps I’m delusional here (entirely possible), but on a personal note, my decisions and those of others in the community feel impactful.

@danielo

Applying this, SCRF using SourceCred does not entail any collective decision of SourceCred users at large, nor is the way SCRF is using the protocol attributed as a decision taken by SourceCred itself.

Agreed. What SCRF decides is its own decision. I’m writing this post in part to give it as much information as possible, so that SCRF and others can make informed decisions around SC.

I will add that ‘attribution’ is a very blurry concept for me. Presumably scientists are better at this, but observing narratives emerge around SC over the last several months, many seem more about desired (or feared) futures than an objective account of causality. E.g. seeing some leadership attribution bias probably.

Right now, I would say SC organizationality is likely low enough to not meet this definition of a DAO. Though we could see something with more organizationality emerge

[kingdamieth]

DAOs were executed to be trustless and for a period being, They worked effectively without need of trust, hence bringing up real issues about the worthiness of current administration hypotheses. The limitations of DAO’s doesn’t discredit trustless associations/organizations. What are the difficulties that should be tackled assuming trustless associations are to succeed? While The DAO’s administration might have fizzled, Do we predict the chance of “CAOs” that are represented by centralized, hierarchically-organized smart contracts?

[kelsienabben]

I have a piece with a colleague on “lifecycles” of DAOs here (including resurrection): The Lifecycle of a DAO: Inside a Cultural Phenomenon

[kelsienabben]

And DAOs :) feedback welcome: DAO Design Patterns:. Components that constitute… | by Kelsie Nabben | BlockScience | May, 2022 | Medium

[Tolulope]

Thank you for sharing this interesting article.
I love the explanation about the cultural aspect of “gm” and what it represents.
It also provides a great answer to @kingdamieth’s question on whether DAOs can die. DAOs die when their mission comes to an end or they have fulfilled their mission.

[Elbeth]

It’s very interesting to read different perspective of participants objective views on what DAOs are. Invariably, all these still corroborate some key characteristics of a DAO.

Would you like to elucidate how the “qualifies” were generated? Was is from body of knowledge or from individuals?

[Elbeth]

@danielo

But then, for me, the goal is not to be more DAO, the goal is better human coordination. As long as we don’t call everything that’s not a DAO a DAO, then we can still have a shared language to advance the discussion. And that’s the contribution this piece aims to make

Absolutely, this speaks to Owocki posture on “it’s all about coordination challenge”

The “livingness” is fuelled by emmergent factors of a complx system which is a characteristics of a DAO.