CIP-0084? | Ledger evolution (meta)

[JaredCorduan]

This CIP aims to make the Cardano ledger a registered category of the CIP process.

---

(rendered proposal in branch)

[michaelpj]

I still think ledger events should pretty clearly be "out". They don't affect the operation of the Cardano network. A competing implementation could omit them and be perfectly correct.

[rphair]

This looks like a definitive statement from the Ledger team so I'd be ready to approve if these issues are addressed, also pending any further input from @michaelpj & others 😎

[rphair]

All the details that I marked have been filled in & elaborated, so I'm happy to approve this while keeping an eye on further review.

[KtorZ]

I think the "Specification" would benefit from a short section about what it takes for a ledger CIP to become active. Are you expecting people to submit PRs to the ledger, or would you rather avoid people doing so (and have only the core maintainers do the work that's agreed on CIPs)?

[JaredCorduan]

Are you expecting people to submit PRs to the ledger, or would you rather avoid people doing so (and have only the core maintainers do the work that's agreed on CIPs)?

Something in the middle, I think. This involves perhaps a big topic that we should discuss.

Features do not get implemented in isolation in the ledger. We collect up a bunch of features, see how they interact, design a specification for a new ledger era, and then implement the era. I actually wrote up an earlier draft of this CIP that was much different than what I've submitted, where "ledger era" was its own ledger sub-type. In the end, I talked myself out of this since it could be argued that this is more of the release process. The ledger team would of course love help with PRs, but it would need to match our existing workflow, which is to build out the code once the era is settled. I'm not sure how the bundling should be involved in the CIP process.

[JaredCorduan]

short section about what it takes for a ledger CIP to become active

I could probably use some guidance from the CIP committee as to how to organize this. These are the stages in my mind:

• everyone agrees it is a good change
• it is selected for inclusion in some ledger era
• it is added to the formal spec
• it is implemented
• the hard fork happens and it is live

[rphair]

@JaredCorduan this is up for Last Check at our CIP meeting today & I'm going to recommend that this be merged... one issue perhaps to discuss at the meeting is if the text still requires any specific mention of the RSS proposals, other than your suggested language to cover them, which I think is currently adequate (#456 (comment)).

[rphair]

OK counter to my last comments, from meeting (@KtorZ) it seems that, although clarification is still needed, it should be literal clarification that RSS proposals are out of scope for Ledger because proposals categorised as such should be straightforwardly applicable to the Ledger.

[JaredCorduan]

OK counter to my last comments, from meeting (@KtorZ) it seems that, although clarification is still needed, it should be literal clarification that RSS proposals are out of scope for Ledger because proposals categorised as such should be straightforwardly applicable to the Ledger.

Sorry @rphair , I'm a bit confused. I do consider RSS proposals to be within the scope of the ledger category. I was hoping that this prose would make it clear:

Some changes may require getting an expert opinion. For example, changes to the incentives model (such as changing the reward calculation) should include a proper game theoretic analysis.

[rphair]

@JaredCorduan that was also my understanding & the basis of my last couple of comments, but I had to retract it after today's meeting discussion suggested that the Ledger category should only be applied to the more traditionally ledger-oriented things (plus governance now obviously). We don't have a recording of the meeting yet so I'll leave it to @KtorZ to qualify that statement.

edit, p.s.: the discussion came up after I repeated #456 (comment) during the meeting.

[KtorZ]

Happy to move this one forward, modulo one concern:

• In the following sentence: "Some changes may require getting an expert opinion." Who qualifies as "an expert" here? Is it more of a general notion as in, someone that has invested enough efforts in studying those topics or are we expecting people with specific credentials (e.g. PhD in game theory or similar).

[rphair]

Also moved at the CIP editors' meeting that immediately after this PR is merged the editors will do one more PR to:

• apply the Ledger category to the recently merged RSS proposals and update their Inactive status to Proposed;
• also apply that category (after updating format & header structure for the revised CIP-0001) to the much older RSS proposals that were merged before the RSS restriction (there were at least 2 of them);

... so that all merged RSS proposals in the CIP repo will then have a Ledger category and a Proposed status.

[JaredCorduan]

• Who qualifies as "an expert" here? Is it more of a general notion as in, someone that has invested enough efforts in studying those topics or are we expecting people with specific credentials (e.g. PhD in game theory or similar).

This is a really great question, one that I have struggled with myself and failed to find a good answer. I definitely do not think that credentials are the answer. The best I can come up with right now is to let the CIP editors and/or the ledger category-specific reviewers make judgements calls.

[rphair]

let the CIP editors and/or the ledger category-specific reviewers make judgements calls.

@KtorZ @JaredCorduan this criteria is fine with me. Since we can't count on connections to whatever few definable experts there may be, we are about to allow (and did allow in 2021) any RSS proposal to be accepted as long as peer + Ledger team review agree that it establishes rationally why it could be desirable. There is no point at which an "expert" can be expected or required by the CIP process, so it would be logical to strike out that whole bullet point.

An alternative would be to write it differently in an a way that is both bureaucratically comfortable and true, like:

• Though proposals can be accepted solely on the basis of peer and Ledger team review, some areas (e.g. changes to the incentives model) might only considered ready for implementation if accompanied by an opinion from an expert designated by the implementor (e.g. with a proper game theoretic analysis).

[rphair]

@WhatisRT @lehins (as the remaining ledger reviewers): to help ensure this proposal is ready to be merged, would you please respond to #456 (comment) adding your preferred language regarding RSS proposals... or give your permission to the CIP editors to modify the language to qualify what is meant by an "expert" as suggested above? cc @JaredCorduan @KtorZ @Ryun1

[WhatisRT]

I think having some language in the CIP about expert opinions is good, so I'd rather keep that line & amend it as you've proposed.

[rphair]

@WhatisRT would this be acceptable?