stylised flowchart representing digital relationships

Cardano Improvement Proposals (CIPs) — Challenges: Past, Present & Future

Part 3: How have we overcome the difficulties so far? How might we improve?

9 min readAug 18, 2023

This 3-part guide is a refinement of our funded proposal for CIP Editing in Cardano’s Project Catalyst Fund 10.

Especially if new with the Cardano community or CIPs and the progress we have made with them so far, please see the previous instalments:

What do CIP Editors watch for in spotting difficulties & potential improvements?

Beyond the particular threads that emerge in developer discussion online over particular CIP-related issues, there will always be an overall, subjective sentiment of the Cardano community about whether the CIP process is “working”.

Therefore, in addition to the goals that we have met cleanly and predictably, it can be equally instructive to look at the perceived difficulties, criticisms, and failures that might interfere with community satisfaction with the CIP process as a whole:

“Does the CIP process work?” Some challenges & responses:

About straightforward standards issues we can always clearly demonstrate that user and editor feedback has been used to build a complete proposal. For these CIPs, the objective will simply be to progress all proposals within the time available to us as editors. Therefore the main measurement for developer satisfaction for the bulk of proposals will be a lack of visible complaints online: the simple metric of “no news is good news” (though there will often be “thank you” messages posted which can be seen in many GitHub PR threads).

But ultimately the feasibility of the CIP process, and of my own “community” role in this process, will be whether the Cardano community is also satisfied by handling of CIP issues which border on external issues beyond our control, especially “political” factors. Even after all the routine work is done, communities will often remember only the most dramatic issues: so the ultimate test of performance will be whether we have handled these pathological cases with enough objectivity and sensitivity.

Two examples are helpful to illustrate this subjective criteria: one to show how it’s been applied successfully in the past, and another illustrating our ongoing efforts at this time:

Handling of RSS (Reward Sharing Scheme) proposals

From 2020 through 2022, members of the Cardano community submitted CIPs to argue for commonly requested updates to blockchain proof-of-stake rewards: particularly to change the minimum pool fees, the K factor (number of ideal stake pools), and/or the effects of pool pledge. These CIP authors understood that CIP editors had no control over whether these policies were implemented by IOG core developers, but still became disappointed with the CIP process as a whole because this lack of institutional support led to older RSS proposals being apparently ignored and newer ones accepted with an “Inactive” status.

Eventually a combination of internal communications and editor & community advocacy led to a resolution of this problem by defining these proposals as under the purview of the Ledger team at IOG (see key factor in resolution). Once this opportunity became available, although the long exclusion was “not our fault” we were still responsible for maintaining faith in the CIP process by promoting the acceptance of these proposals (see Tweet & its link). The positive response from the community suggests that our attention to community satisfaction has been successful after a long period of unavoidable difficulty on this sensitive issue.

Governance and CIP-1694

During nearly 8 months of posted updates to the Ledger based governance mechanism of CIP-1694, the sensitive “political” nature of this proposal invited unprecedented amounts of feedback about issues of constitution, representation and ethics which were literally outside the scope of the proposal. This presented a difficult moderation problem (over 700 comments) and some routine commenters complained bitterly when we were ultimately obligated to merge this proposal (see announcement) so that further edits and focus-group (“governance workshop”) refinement could then be applied regularly.

For this realm of proposals, CIP editors will face criticism that we have somehow appointed ourselves “gatekeepers” who only represent the interests of the Cardano established companies. On my own part I definitely have no such obligation (I am unpaid by and unaffiliated with these companies) and am confident that all other editors consider only technical factors in whether or not proposals are merged.

Still we have to accommodate this negative feedback (seen in response to the announcement above) which may be inevitable from the community. Although we may not respond when this presents criticisms that are beyond the scope of a particular CIP or the CIP process, the community will have to observe that we remain impartial and keep taking this criticism into account. As in the case of the successful RSS proposal issue above, there is often a resolution that can come from somewhere, even if we ourselves cannot provide it… so we need to remain attentive to the most intractable issues in the meantime.

Why doesn’t the Cardano Foundation (or IOG) just insource, govern & fund the whole CIP process directly?

From the number of currently active editors multiplied by their estimated average workload, the total amount of work in the CIP process could likely be done in a 40-hour work week: i.e. by a single employee doing nothing else. So why all this fuss about division of labour and arranging Catalyst funding for independent, part-time editors? Because of the factor that makes blockchain possible in the first place: consensus.

A solitary bureaucrat would be the direct opposite of a team with finely distributed skills and dispositions: such a diverse team makes broad community outreach possible as defined above. Internal connections to IOG and the Cardano Foundation are impossible to arrange from my own position (and have been vital to validate implementation paths for CIPs, as other editors have), while my own diverse work focus allows me to cover more ground in community, communication, and documentation.

More importantly, CIP editors from multiple organisations provide consensus about CIP decisions and processes from multiple points of view and a mixture of technical, social, and academic dispositions. Since we cannot make everyone in the Cardano community a CIP editor, we can achieve the next best thing by diversifying the CIP Editor group to cover large segments of that community through practical understanding as well as empathy.

As hoped in mid-2021 when I was tentatively invited to be a CIP editor, last year’s precedent of funding editors on Catalyst has more fully decoupled the CIP process from bureaucratic constraints including potential or perceived conflicts of interest that might result from the Cardano Foundation paying editors directly to manage its own repository.

What would it take to future-proof the CIP process?

First consideration: Time requirements

Relative to my own workload, regular time spent on CIP issues has approximately doubled in the last year, mainly because of the following demands (while our CIP team remained the same size; though at the time of this writing we’ve been happy to approve a recent addition… but that’s another story):

The range of subjects that CIPs cover is broadening steadily.
The number of CIPs pending review (open “pull requests”) has increased linearly.
The potential relationships between CIPs are increasing combinatorially.
Community discussion of CIPs has increased steadily, with new channels being added periodically.
CIPs are achieving more commercial impact (particularly NFTs, token metadata, voting & governance) and therefore have greater demand for usability and precision.
Many CIPs, especially the oldest and most popular ones, are now evolving multiple versions.
CIPs will be used in a wider variety of contexts (translations into other languages, repurposing onto web sites beyond GitHub).

Second consideration: Incoming and outgoing CIP editors must be balanced.

An important part of this effort would be also to ensure an equitable and well-documented division of labour between editors: a crucial aspect of retention within any organisation.

This brings us back to a proposed “long term project” mentioned earlier — to document CIP Editor workflow loosely yet practically — with the blog series you are reading now being my own first attempt to do this (other than my Catalyst Fund 10 proposal itself: see link at beginning of this article).

I am proposing we build a body of CIP Editing & Review documentation which will include, organise, and impersonalise the workflow description already written this series of articles. Content will be, roughly:

Workflow of CIP editors: both regular and irregular components
Meetings: why we have them, what we do, how to prepare & follow up
Community engagement: groups and suggestions for interacting with them
Ethos: what the CIP process does and doesn’t do (expanding on this from CIP-0001)
Reviews: how specifically a GitHub review is processed when preparing to accept a CIP draft… this will also be directly useful to developer and community members engaging with the CIP process, including but not limited to those who may be “training themselves” for an eventual application as a CIP editor.

I realised how important this was going to be in the course of this conversation on the “Cardano Professional Society” Matrix MBO (courtesy of Christophe Garant):

Question that’s been bugging me. For CIP reviewers and others, is there a “Constitution” guiding light to make decisions again? Is it the Cardano whitepaper or ethos on the website? What to a CIP Editor do you use for a source of truth to vet decisions against? …
Would you have a “getting started” or “recommended reading list” for CIP training? Obviously the Whitepaper and CIP-0001.
Maybe it already exists in Cardano.org Docs or Developer Portal? …
I think it would be wise at some point to start planning for training material (maybe just a few slides or a 2-pager best practice and lessons learned).

My response, which explains why we originally took a very sparse approach to documentation, but also admits we might now have to drop that reservation:

… the “unwritten constitution” of CIP editing is that any defensible technical idea that is well-conceived and -documented enough to be accepted as rational by editors, peers and potential implementors has a place as a merged CIP. Anything emotional or political that surrounds this strictly logical process can and should be ignored. …
From the beginning the first CIP editing committee intended (and I supported, being the first person to arrive beyond the original group) that there should be complete visibility on the GitHub repo alone. That has the maybe unfortunate consequence that the entire body of CIPs, reviews, comments, tags, commits & merges (plus meeting agenda & notes now) is the training material.
It has the advantage that this steep learning curve reinforces the idea to ourselves & others that subjectivity cannot be part of the process, as there would be if there were “training materials” in writing or video to narrow reviewers’ and editors’ focus to the most “essential” things. Without that, we can only continue the precedent of attention to every detail about how any CIP might be technically contested or poorly documented.
… but in writing the Catalyst proposal for my next years’ funding you are making me want to include something about a time budget to expand the CIP committee… so I might include a statement about training, documentation & maybe even mentorship.

Based on this dialogue I realised the importance and timeliness of having this documentation element as part of our overall CIP process (though not in the strictly defined CIP-0001): both to “future proof” its evolution and to assure that we would have a pool of potential new CIP editors based on assurances that the consequences & responsibilities of their participation could be well understood in advance: something we cannot officially offer them today, but will certainly be able to in the coming year.

“That’s awesome! How can I help?”

Please share, tweet and discuss (you can leave comments on) all 3 parts of this series:

To support our work in CIPs and other cryptocurrency standards for security and governance, please delegate to COSD stake pool in your Cardano wallet and follow us on Twitter for related announcements: 🙏