Governance

Governance is the process by which a DAO maintains its coherence in the present, guides the group into the future, and reviews its past actions. These domains are covered by executive, legislative, and judicial governance. By definition, all types of governance fall into one or more of these branches. Each branch of DAO governance can be further broken down into hard and soft protocols. Hard protocols are automated by the backend distributed computing software everyone shares. Soft protocols include automated protocols in frontend software that users elect to use, as well as cultural norms that guide members' behavior.

In DGF all governance is determined through REP-weighted democracy. REP denotes reputation tokens, which are gained by peer validation and staked upon assertions in the Validation Pool.

Following the requirement that REP be domain specific in order to give the tokens the authentic meaning of reputation, governance through REP tokens will usually require a distinct type of REP token, gREP, instead of the main type of REP token the DAO uses to measure the reputation of expert workers, wREP.

DAO governance can be analyzed for its effectiveness by appealing to principles explained in REP tokenomics.

Overview

Governance is the explicit process for how an organization is organized. Governance is the structure, superior to the individual, which makes an organization become and remain coherent. Governance is the set of rules or protocols for behavior that dictates the limits of behavior, after which an individual is expelled from the group. Freedom is the apophatic inverse of governance. Governance is group power; freedom is individual power.

The idea of governance is best understood through its etymological roots in the Greek word, kubernetes, which means guidance, steering, control. Governance is necessary for any organization to keep coherent, but it is always an overhead cost. As such, ideal governance is minimized under the constraint that it remains effective. Analogously, piloting a ship is most efficient using the smallest possible adjustments that can effectively steer the ship to its goal.

Governance as control can be separated into the static concept of ownership and the dynamic concept of power.

Ownership (i.e., property or financial equity) in a contemporary DAO is determined by tokens. Tokens are digital objects that are algorithmically minted and transferred. Pseudonymous control of tokens in a DAO is accomplished using public key cryptography. Digital signatures allow someone with the secret, private key to prove ownership of a digital token. This asymmetric encryption tool allows individual privacy while maintaining maximal transaction transparency in a decentralized network with no leader.

Power in a contemporary DAO is achieved with smart contracts. A smart contract is a software-program that encodes business logic that can be far more complicated than any traditional legal business contract. Such automated contracts self-execute at the speed of electricity. Therefore, much more sophisticated business arrangements are now possible on scales previously unimagined both large and small. For example, every device in the Internet of Things can dynamically negotiate with every other micro-component on the globe, in unlimited multiparty arrangements.

Governance, from a higher perspective, means guidance of the group, not just individual tokens. Raising our attention from the business details of token ownership and smart contract operations to the more abstract level of analyzing the justification of those business operations, analyzing the foundations of group consensus, that is the objective of legislative governance. As both ownership and power dynamics will necessarily evolve in time, we also need to govern how the rules change, and moreover, how to change the rules for changing the rules. Token minting and ownership is 0^th-order governance. The business achieved by smart-contract-enabled token transference is 1^st-order governance. How we change smart contracts and backend logic for minting tokens is 2^nd-order governance. How we change the way we change contracts is 3^rd-order governance. 1^st and 2^nd-order governance is executive governance, while 2^nd and 3^rd-order governance is legislative governance, which are discussed in separate sub-sections below.

There is one higher level order of governance that is necessary in any practical application of power. Every system is flawed. Flawed means that the intent of the system is not captured by the formal protocols that are specified. In fact, it is an incontrovertible fact that every human attempt to fully understand anything has failed to some degree. Every actual instantiation of any system is therefore even more flawed. Mistakes will be made. Every system of governance in history has therefore found it necessary to build an institution for dealing with this reality. The process of stepping outside of the system must be built into the very system. In sophisticated social systems, this is called the judicial branch of government.

Governance is broken down into executive, legislative, and judicial governance. This triad follows the information theory breakdown of information transmission, processing, and storage, respectively. Our goal is to specify mechanisms for achieving all these types of governance in a decentralized context.

Executive governance

Main page: Executive governance

Executive governance consists of automated policing of peer work. The mechanism for executive governance is the Validation Pool using binding (tightly-coupled) votes on the acceptability of actions that affect the DAO.

Executive governance in a DAO is the active, direct control of ownership in the organization. Executive governance consists of policing who is inside or outside the group and how much power insiders have. Executive governance is therefore the accounting of the lists of owners of the various types of tokens that have power through the DAO’s smart contracts. Inasmuch as a DAO is truly decentralized, this control must be ultimately democratic. Therefore, some sort of token-weighted voting is necessary. Inasmuch as executive governance is the process of executing the regulations the organization has already agreed to follow, executive governance should be automated as much as possible. In a DAO this means algorithmic enforcement using smart contracts run on a decentralized computing platform.

Legislative governance

Main page: Legislative governance

Debating and voting on updates to DAO operating parameters and smart contracts (hard protocols), and cultural norms (soft protocols). The mechanism for legislative governance is the Validation Pool using a series of votes on proposals to adjust existing protocols. The proposals are recorded as posts in the Forum. The series of votes on a single proposal gradually change from optional, non-binding (loosely-coupled) polls of the REP holders to binding (tightly-coupled) votes that determine new law. A tightly-coupled vote means when you vote against the majority with your REP tokens, you lose them and they are redistributed to the winners through the Validation Pool mechanism. Loosely-coupled votes merely register voters' opinions without redistribution of REP.

Legislative governance in a DAO is primarily concerned with updating the P2P software that automates executive governance. The primary legislative governance activity is adjusting the parameters of the algorithms in the smart contracts. Such governance will always be needed, to optimize any DAO’s interaction with the dynamic market in which it functions, because we are not capable of creating a perfect system that can anticipate the future. More rarely, but still inevitably, governance will take the form of more dramatic updates to the software, such as updating UIs, updating smart contracts, and most rarely, updating back-end logic that profoundly reorganizes the functioning of the DAO.

Secondarily, legislative governance sets the culture for how members can thrive in the group. Legislation determines the rules as well as the rules for changing the rules. This secondary function of legislation is meant to correct the failures of the DAO, which are inevitable.

Judicial governance

Main page: Judicial governance

Judicial governance of a DAO consists of reviewing past actions and decisions by revaluing the Forum. Since the Forum is technically a WDAG (weighted directed acyclic graph), judicial governance is technically the process by which the WDAG is reweighted. This judicial governance mechanism is necessary for instituting an evolutionary structure which converges on ever greater security.

Judicial governance in a DAO is fundamentally a matter of approving token ownership in the DAO. The more visible, secondary function of judicial governance is the review of REP token accounts. Judicial governance allows a DAO to re-evaluate REP holdings by slashing accounts that were later determined to have harmed the DAO, or augmenting accounts that had a later positive affect.

Inasmuch as business and social decisions are inevitably flawed, this function is necessary for the long-term stability of a DAO. It is necessary for a mechanism of review to exist, but in a healthy DAO it will rarely be used, compared with the other branches of government.

Reputation tokens should be slashed for two basic reasons: first, when members violate explicit protocols, even though the violation was not detected by automated executive policing; second, when members betray the more abstract values the DAO shares which have not been explicitly encoded in automated protocols. Judicial governance also allows more accurate accounting of power after a DAO updates its values or protocols. Judicial governance is therefore fundamental in incentivizing members to behave well, because if it is functioning, judicial governance imbues REP tokens with some of the crucial qualities that characterize authentic reputation: future-orientation and non-fungibility.

Forum references allow review

The Forum is a collection of posts (comments, proposals, work evidence, etc.), each one potentially having a set of references to older posts. As such the Forum is a DAG (directed acyclic graph), since the posts are nodes in the graph, and references are directed edges between posts. The Forum graph is acyclic, since older posts cannot reference newer posts. The Forum is weighted by the REP that each post earns during the REP Token Minting Mechanism stage and the strength of each edge reference. The Forum is reweighted every time a new post is added, allowing a decentralized means of review.

Forum Reference Mechanism

Main page: Forum Reference Mechanism

The Forum Reference Mechanism allows DAO members to re-weight the Forum. This means new posts can give or take REP from older posts through a weighted reference. This allows DAO members to review previous actions that had previously been given a particular quantity of REP tokens, and either slash or augment the posts' holdings depending on whether they are later deemed less or more beneficial to the DAO than had originally been decided.

With two basic reference functions, DAOs have complete control over the process of reweighting the Forum WDAG.

Reweighting primitives

The Forum can be completely reweighted through a process of adding references. Reweighting the WDAG means that any given WDAG can be changed to any other choice of weights by leaching REP from some posts and donating REP to others. This is the REP review process.

Donation primitive

?? Add explanation, figures & code link

Leaching primitive

?? Add explanation, figures & code link

Incinerator primitive

?? Add explanation, figures & code link

??

Motivation for the existence of the incinerator:

If every WSC post is required to donate through references at least ${\textstyle p=10\%}$ of its REP to governance, then that REP is always available for revaluing the WDAG in response to governance proposals that the DAO agrees on.

Then WSCs can be updated in response to such approved proposals.

The workers automatically use those updated WSCs (unconsciously) because their UIs automate that choice, If no governance needs to happen at the time of that particular WSC (governance should be rare compared with work) then their 10% is simply burned directly in the incinerator.

If every WSC is referencing the same 10% in governance, then no individual loses relative to the whole. Every worker's total REP is affected the same, so their REP salary is not affected by the 10% loss (by the Quantity Theory of Money in economics). They do not notice the 10% lost whenever it is incinerated because there is no actual loss in REP salary if everyone participates in the same governance "tax".

However, when REP is donated to someone, instead of incinerated, then this is an actual cost to the group, since the REP salary is then shared with any new REP that is not burned. But since everyone loses 10% of their REP automatically, since the tax affects everyone equally, not just the particular worker whose WSC was used to donate REP, then.

The ${\textstyle p=10\%}$ parameter is arbitrary. Governance can decide to make ${\textstyle p=1.0\%}$ if the DAO wishes governance to be slower, or ${\textstyle p=90\%}$ if they desire it be faster. Any fixed percentage will not affect the WSCs' remuneration in REP salary fees.

However, if a DAO does change this parameter, then there will be an effect on how workers are remunerated until the lifetime of the REP token clears all the old tokens from the system.

Finally, the incinerator makes the reweighting primitives complete. E.g., suppose the DAO wishes to move 10 REP from post1 to post2. Then post3 uses 10 value & leaches from post1 with strength 1.0 and donates to post2 with strength 0.5 and donates to incinerator with strength 0.5.

Then in the end, post1 has value 0, post2 has value 10 and post3 has value 0, as desired.

It costs 10 REP to re-distribute 10 REP.

Given the constant 10% burn argument discussed above, this doesn't actually cost the poster of post3 any meaningful REP, since everyone loses 10%.

If everyone burns p=10% of their tokens, then nobody loses anything. If only one single person burns 10% of their tokens, then that person does suffer: they lose 10%. But if everyone burns 10%, every time, then they're all in the same situation and nobody loses, relative to each other--the REP salary is the only thing that matters, not the absolute number of REP tokens. (To put a fancy name on it, REP is a type of "gauge". While money is closer to absolute.)So if the WSC demands that everyone always cedes 10% then nobody loses. That's how governance can work without punishing anyone. The system can incorporate governance if it constantly/automatically includes p% references in everybody's WSC

Completeness

??Add explanation & proof

The primitives are 1. donating, 2. leaching, and 3. burning. Together these three functions are complete in the following sense. To say a set of reweighting functions are reweighting complete means those are the only functions you need to be able to revaluate/reweight the WDAG in any possible manner. The example discussed shows that new posts are necessary, but those new posts end with 0 weight, so they don't count in the WDAG. (Because they don't count in the REP salary.)

Governance process

??If every WSC post is required to donate at least 10% of its REP to governance, then that REP is always available for revaluing the WDAG in response to governance proposals that the DAO has ratified.

Then WSC can be updated to execute those proposals over the course of many work events, which gradually reweights the Forum.

The workers automatically use those updated WSCs because their UIs automate that choice,  

If no governance needs to happen at the time of that particular WSC (governance should be rare compared with work) then their 10% is just burned directly.

If every WSC is referencing the same 10% in governance, then every worker's REP is the same value (by the Quantity Theory of Money in economics). They do not notice the 10% lost whenever it is incinerated because there is no actual loss in REP salary if everyone gets the same governance "tax".

However, REP that is donated to someone is an actual tax, since the REP salary is then shared with any new REP that is not burned. But due to this system of everyone losing 10% of their REP at initiation, the tax affects everyone equally, not just the particular worker whose WSC was used to donate REP.

Finally, the 10% parameter is arbitrary. Governance can decide to make that 1% if the DAO chooses a slower governance process which demands greater deliberation. Or it may choose 90% if you they value faster governmental response. Any fixed percentage won't affect the workers' remuneration in REP salary fees, since every worker is under the same REP reward regime.

However, if a DAO does change this parameter, then there will be a change in how workers are remunerated, until the lifetime of the REP token clears all the old tokens from the system.

Hard protocols

Main page: Hard protocols

Hard protocols??

Soft protocols

Main page: Soft protocols

REP tokenomics

Main page: Reputation tokenomics

REP tokenomics is the study of the economic consequences of DAO governance mechanisms of the REP minting mechanism and the Forum review mechanism. These mechanisms are the major factors in the executive and judicial branches of DAO governance, respectively. By mathematically determining the rate of REP minting, distribution and redistribution, REP tokenomic analysis helps guide governance decisions to choose operating parameters which optimize incentives for behaviors that serve a DAO's goals.

Transcendental values

Main page: Transcendental values

It is important to recognize that even though governance is the obvious force needed to keep an organization coherent and stable, it is not the most important force in this regard. The values and goals of an organization are ultimately more important than the explicit rules for keeping an organization solvent. This effect is particularly pronounced in decentralized organizations: when roles are less specialized and authoritative leadership is less important, members give greater importance to abstract values in maintaining harmony.