New Category: Governance

[Crypto2099]

Currently all CIPs and CPSs must be assigned by their author to one of the following Categories

• Meta | For meta-CIPs which typically serves another category or group of categories.
• Wallets | For standardisation across wallets (hardware, full-node or light).
• Tokens | About tokens (fungible or non-fungible) and minting policies in general.
• Metadata | For proposals around metadata (on-chain or off-chain).
• Tools | A broad category for ecosystem tools not falling into any other category.
• Plutus | Changes or additions to Plutus
• Ledger | For proposals regarding the Cardano ledger (including Reward Sharing Schemes)
• Catalyst | For proposals affecting Project Catalyst / the Jörmungandr project

Now that Cardano has officially entered the "Voltaire Era" with the Conway ledger and Chang Hard Fork introducing on-chain governance, there is a surge in interest and attention on the CIP/CPS process to facilitate governance discussions for both dry (technical, on-chain) governance as well as wet (social, off-chain) governance.

We already have 4 merged CIPs (CIP-100, 108, 119, 1694), 1 merged CPS (0007) in addition to 4 proposed CIPs (CIP-120, 129, 136, #936) and 1 proposed CPS (#935) that all specifically address facets of both on- and off-chain governance. We have arrived at this point approximately 2 months after on-chain governance going into effect.

I would argue that this raises enough attention to warrant the inclusion of a new category for CIPs and CPS specifically for those related to Governance.

[adamrusch]

Yes, please! We have had discussions in the Cardano Civics Committee about how we can use CIPs to specify governance rules and procedures that cannot or should not be part of the Constitution document. Having a "Governance" category will provide a way to address these issues.

[intertreeJK]

Seems reasonable to me.

[Hornan7]

I fully support the introduction of a dedicated "Governance" category for CIPs and CPSs. As Cardano's governance continues to evolve in the Voltaire Era, it is important to establish a clear framework for proposals that address governance-related topics, without necessarily enforcing them through a constitution. This new category will be particularly valuable for organizing CIPs related to governance standards, such as codes of conduct for DReps or the Constitutional Committee, as well as other practices or metadata standards.

[gitmachtl]

Seems reasonable to me.

johnny, how far are you with your ideas to restruct the CIP tree in general? as we talked about it in the CLI/API wg?

[rphair]

An important question that I think must be very clearly answered before proceeding with the use of the CIP process to document governance standards:

• To what extent would Governance CIPs allow proposal authors to bypass the mandated governance process requiring voting consensus to create & modify its own rules... given that CIPs, though they can be challenged by the community, are ultimately reviewed & approved by a much smaller number of agents?

How can it be proven that this is not a risk? If this is a risk, what would mitigate it? ... strict "Path to Active" criteria requiring that such CIPs (and maybe some CPSs) follow the generally recognised governance vote to become Active?

Since there already seems to have been some agreement about this proposition in advance (@Ryun1 @perturbing at this time we still haven't heard from you on this issue), as in #935 (comment) I need to present as the "Devil's Advocate" to try to pre-empt criticism from the community... however uninformed or unjustified it might be... that Governance CIPs will be a way for influential proposers to bypass the usual Governance process in passing their preferred propositions.

Unless someone can disqualify this risk conclusively, I think we need to place strict definitions on:

• how Governance CIPs and CPSs stand in relation to the Governance process (p.s. I think @Hornan7's definition as a "framework" ... "without necessarily enforcing them" ... could serve well here);
• how these CIPs are declared Active;
• how these CPSs should be interpreted (since e.g. CPS-???? | Social Governance #935 looks to me like a Governance rule-book);
• who is expected to be involved in Governance CIP/CPS PR review and acceptance.

[rphair]

restruct the CIP tree in general

@gitmachtl please will you document exactly what's being referred to in #937 (comment)?

[Crypto2099]

@rphair I think you bring up a ton of very good and valid points of consideration in #937 (comment).

Regardless of the rule setting that we determine here, I do still firmly believe that a governance category for CIPs/CPS in general makes sense given the number of proposals at this point that specifically address it as a topic.

However, particularly w.r.t. #935 where we very much are addressing a hybrid between on-chain and off-chain expectations, perhaps it would be best that the CPS formally define that every proposal aiming to address Social Governance must include criteria for an on-chain Info voting action in order to garner enough on-chain approval signal to justify it being merged or at least toggling it into Active status. It would probably make sense that the CIP would need to be merged as Proposed prior to being submitted for on-chain voting as this would give us a canonical version of the text that could be referenced in the vote.

Potential/Suggested Language for #935:

Path to Active for Social Governance CIPs

Any proposed CIP that addresses social governance and does not have a clear and direct
on-chain implementation MUST include, as part of its Path to Active section, a clearly
detailed procedure for submitting the CIP for on-chain approval via an Info action.

The criteria for ratification of any social governance CIP SHOULD meet or exceed the
thresholds currently defined in the Constitution and on-chain Protocol Parameters for
ratification of similar protocol parameters or Constitutional modifications.

Proposals SHOULD only be submitted for on-chain voting after they have been formally
merged under Proposed status so that a canonical reference to the language of the CIP as
proposed may be referenced in the Info voting action.

Social governance CIPs MUST NOT be marked as Active by the CIP Editor body until
evidence of an on-chain vote meeting the above criteria has been submitted as a final and
sole modification to the CIP document content.

[adamrusch]

@rphair @Crypto2099 You are talking about a component that I believe is vital for "Social Governance" CIPs - and which I think would be wise to include in the Governance Category definition. Expanding CIPs out in this way is a logical step as every CIP defines behavior that must be taken off-chain to impact on-chain behavior in some way, shape, or form.

Every legal system relies upon multiple sets of laws and the on-chain governance of Cardano is no different. Constitutional Laws are supreme and trump all other consideration. Common Law is important for helping judges render consistent rulings but because we do not (yet) have an arbitration system in Cardano it will have to be a future workstream. CIPs are basically a form of Administrative Law where the community has entrusted our CIP editors to help define how various procedures will work.

What we have now with on-chain governance is the opportunity to also use CIPs to record Statutory Law, or procedures that are ratified through the legislative process of Cardano. If we have a specific procedure put into place of how the DReps and/or other governance actors of Cardano can vote to ratify and activate Social Governance CIPs we make them legitimate rules of governance. This is why you will see that the Social Governance proposals that came from Mr. Dean (Social Governance - Budget Guardrails) and myself (Social Governance - Voluntary DRep Code of Conduct) include explicit vote mechanisms.

We've discussed extensively in the Cardano Civics Committee the fact that while we want the Constitution to be a living document that responds to the needs of the community, we also want to make sure it doesn't become so large and so complicated that is inaccessible. If we didn't already have the CIP process we would need something very similar to it for "Statutory Law" that the governance actors of Cardano can use for setting and updating rules or best practices as necessary.

The final point that I will mention is the nature of voluntary engagement that are central to blockchain ecosystems such as Cardano. Most CIPs are written from a Best Practices standpoint to encourage participants to bring their activities in line with community standards so that they will be legible on the network. I expect that we will find the same to be true for many of our "Statutes" in Cardano, such as the Voluntary DRep Code of Conduct, which we cannot strictly enforce on a permissionless system but which we hope the community will uphold by rewarding those who exemplify the standard.

[rphair]

This hour we have yet another submission... and they have been coming in daily, even over the weekend... of a CIP categorised as Governance without any agreement yet that proposals like this are even within the scope of the CIP process.

Somehow the authors of these proposals have been encouraged to charge ahead with the CIPs that are suddenly being submitted. These proposals — which, as they are currently written, appear not to require the due (voting) governance process to merge them — would have huge impact on human affairs in the Cardano space, with organisational specifications that are completely non-technical and therefore patently incompatible with CIP-0001 which has long required that CIPs are technical documents.

Maybe the 2 remaining active editors @Ryun1 and @perturbing have not contributed to any of these discussions yet because (2 separate possibilities): somehow they have been separately & privately briefed on this behind-the-scenes "change" or they're as surprised as I am about this being sprung on the editors with no explanation, no notice, and no precedent.

Until someone can at least point what conclave decision made it seem inevitable that CIPs should bear this bureaucratic burden — over points of politics and business, i.e. issues that the CIP process for 4 solid years running has never gone near — I am removing the Triage tag for these submitted "Governance" proposals.

This doesn't disqualify them but effectively removes them from tomorrow's & future meeting agendas until this GitHub issue has been addressed, with some editor quorum & community consensus: both in tomorrow's CIP meeting (https://hackmd.io/@cip-editors/100) and in the comments here.

The time freed up at tomorrow's CIP meeting will help us focus on this issue specifically. @Crypto2099 I will leave it to you to invite the authors of these several new CIPs to the meeting. Please plan first to brief all the other editors on the privately arranged decision to use the CIP process — long agreed to maintain technical standards — to maintain bureaucratic and financial structures such as budget policies, human committees, and all the other features in this onslaught of new CIPs:

• CPS-???? | Social Governance #935
• CIP-???? | Social Governance - Budget Guardrails #936
• CIP-???? | Social Governance - Voluntary DRep Code of Conduct #939
• CIP-???? | Trust-Minimized Cardano Budget Process #940

[rphair]

Some more points to consider at the meeting tomorrow & any time in the discussion here:

1 - The 4 potential Governance proposals submitted so far have been written & submitted as if the CIP process and repository were already cleared for them to be reviewed and posted here. Again, I think any such decision would have been reached in conclave — with most CIP editors not being included in the discussion — because otherwise the first question coming up would have been, "Why the CIP repository?"

Before proceeding, editors & advocates of the CIP process therefore also need to consider the logical alternative of a completely separate repository to debate & document these governance propositions. Effectively we would have that anyway, since here's how the CIP archive would look if the suggested Governance category were included on top of the CIP process that exists today:

• the Governance category = addressing human factors like bureaucratic composition, budgetary constraints, etc., rather than technical propositions.
• every single other category = addressing exclusively technical propositions & generally avoiding human factors.

2 - It was straightforward to call all prior categories "Cardano Improvement Proposals" because it was obvious that Cardano (the blockchain) was the thing being improved. However, we would have to make an exception here as well for Governance proposals: since the proposed improvements are all around bureaucratic and economic factors: certainly important, but not directly to the blockchain, network, or tooling.

3 - The only agreed-upon use case of the CIP to document the basis for governance propositions, in my recollection, has been as the basis for hard forks which must in turn be confirmed by a governance vote. This has the advantage of supporting this use case without requiring a small army of CIP editors: since there is a natural limit to the rate at which such technical propositions can be developed.

However, we would need that small army soon enough once Governance CIPs became a requirement for budget & bureaucracy standards... since the ways people would want to influence others about assignments of power & money would be practically infinite.

[Crypto2099]

Source: https://roadmap.cardano.org/en/voltaire/

It has been the intent of both the CIP and Governance processes to work together both in the roadmap and public forums for years. Arguing that we never addressed social governance in the time and era before we had on-chain governance to lend it credibility is not a valid justification to continue with "business as usual".

I would also argue that many of our other existing CIPs deal with human factors like bureaucratic composition. CIP-25 and 68 just describe steps that you must take in order to make a "standards compliant" NFT token that can be recognized by various explorers and marketplaces as valid. In the same way, #936 describes technical requirements that must be included in a proposed Budget Info Action in order for it to be considered as valid.

CIP-0052 literally describes purely bureaucratic composition of how dApps must submit information to auditors to be audited. A 100% off-chain process.

[adamrusch]

@rphair Apologies if this is coming off as sudden. I've been talking with others about the idea of the DRep Code of Conduct as a CIP for a while now but had not taken the steps to formalize it until I approached Mr. Dean @Crypto2099 and asked him for help with ensuring that I met the right standards. He explained that we would need to make sure to write a CPS for this to match and it that it would fit well with the Budget Process Guardrails CIP that he had been discussing with Mr. Solomon, so we worked together to bring these to the table for discussion.

I'm happy to join any CIP meeting to explain what I'm trying to do for my part, if it would be helpful and it fits within my schedule. I just want to assure the CIP editors that this was never meant to be a disruption to the system but rather a way to best facilitate the vision of Voltaire governance as I thought we were on the path to formalize it.

[rphair]

@Crypto2099 #937 (comment): It has been the intent of both the CIP and Governance processes to work together both in the roadmap and public forums for years. Arguing that we never addressed social governance in the time and era before we had on-chain governance to lend it credibility is not a valid justification to continue with "business as usual".

We have recently attempted to validate the premise that CIPs would be used for Social ("human factors") Governance before, with no evidence of any such requirement in the community. Here is the last time I publicly asked (2 months ago) for an inventory of how "CIPs [would be] used as official references for governance proposals":

• Participate in Governance: Add CIP Wiki as reference developer-portal#1317 (comment)

and here is a complete list what came up (as answered by @Ryun1 & @Hornan7, and reviewed by a few people who work with/for the CF):

• governance metadata CIPs (i.e. Governance Metadata Standards)
• governance actions for Hard Fork initiation
• protocol parameter changes

This is consistent with the highlighted phrase in the Voltaire promotional page above (written in the year 2020):

To that end, the Voltaire era will add the ability for network participants to present Cardano improvement proposals that can be voted on by stakeholders, leveraging the already existing staking and delegation process.

... since this refers to network, rather than human, behaviour.