A gentle reminder to the Advisory Committee that our governance outlines its responsibilities in the following way:
The Advisory Committee runs the administrative aspect of the project on a day to day basis with the support of Code for Science and Society (CS&S).
Please keep discussions about things like development, user surveys, governance, etc in the open among community members and contributors instead of behind closed doors.
Absolutely agree that any Governance DECISIONS will always be held in public meetings. The Advisory Committee can certainly discuss anything they want in their meetings including Governance to give us advice. But when it comes to DECISIONS about Governance, and many other decisions about the OpenRefine project itself... those decisions are to be made by the community and the OpenRefine project leadership.
I see it this way:
The Advisory Committee is there to take care of legal doings through CS&S's capacity to handle law and finance issues. Anything beyond those 2 areas is where OpenRefine community and it's current project leaders make the decisions.
I think what is happening is that the Advisory Committee might have forgotten the roles slightly.
Or it just might be that @Martin is not typing up these meeting minutes with his real hat on, as the OpenRefine Project Leader, and instead wearing multiple hats... simply through the loose usage of the word "we" at times. I think @Martin should take steps to not do that anymore, and when "we" is typed out, think about which party that is and replace it with "The Advisory Committee informed me that...". In other words, @Martin should step in and out of the Advisory Committee as always wearing the same hat, that of the OpenRefine project leader who is representing our entire community inside those meetings and out.
@Martin do you agree that you could improve on this in the future where you always maintain your OpenRefine Project Leader hat which represents the community and never the Advisory Committee (and use better wording in your communications)?
Do others agree with me and see things the same way in the above areas of responsibility?
Absolutely agree that any Governance DECISIONS will always be held in public meetings. The Advisory Committee can certainly discuss anything they want in their meetings including Governance to give us advice.
Advisory Committee does not have an advisory role, it has an administrate role per our governance.
Discussions about governance, development, website, and other non admin tasks should be brought to the community(which include the AD members!) not held in closed meetings purposed for administration.
Of course anyone is allowed to discuss any topic in any forum, but we shouldn't see things related to development, documentation, and governance coming out of the AC meetings.
I can see how the lack of a Steering Committee replacement and the need for a project leader to have fixed group to discuss things with could have contributed to this drift away from the governance model.
I know @Martin has worked on increasing transparency and I appreciate those efforts, I do, however, think that those efforts has been countered this drift of responsibilities and discussions into a closed group/forum.
Thanks for shifting this conversation away from GitHub where it originated.
I'm confused by this governance page - OpenRefine/GOVERNANCE.md at master · OpenRefine/OpenRefine · GitHub
I ask because there are differences between that page and this public page from last year: Join OpenRefine's Advisory Committee | OpenRefine
I'd like to understand more about this process. Thanks for explaining more.
Here is the issue and PR related to the latest update on the governance document. This date back from 2020 and 2021 when the project joined CS&S.
Thanks, @Martin. Based on those two links it looks like the governance model was decided in two discussions on GitHub between four individuals, and never shared outside of GitHub. Is that an accurate assessment, or is there more documentation?
I guess if it's the case that there were only four people weighing in before, one possible solution would be to take up a more open process for building consensus over the governance structure as the first step, and then going from there?
If there was a more open process not represented here, I'd love to learn about it.
I think our governance is, broadly speaking, a big mess, and I have been wanting to work on fixing that for some time, for instance with this attempt:
It feels like maybe there is enough interest in this topic to draft something new there?
I would love to move towards more openness in the Advisory Committee, and am very happy that the work with Vision, Mission and Values is giving a good foundation for that.
I would also love to clarify the Governance document in itself because right now it is possible to read it with too wide interpretations. Specifically, I am reading the second point as it is the role of the AC to guide the staff in all manners, not only administrative ones. This also makes sense as it is tightly connected to being the ones having to approve hiring them (point 3) or fire them if needed (point 4). Even if interpreted in a narrow way, ensuring that the staff is doing at least what is needed for the contracts they have and the tasks they have been assigned through grant agreements will naturally span over a wide variety of topics.
And however wide or narrow the responsibilities for the AC are clarified to include, the way we work can be made more open, only very few topics are usually needing privacy for legal or other ethical reasons and those could be handled separately. Let's move in that direction!
This is in my opinion not really a fair point. These people are also the main contributors, the people who license their work for others to use, they should be the ones deciding governance, and it should be done on Github.
Do the AC members actually think it makes sense if things like governance, development, website, etc were a part of their role? Rather than with the committers? I'm rather surprised by the responses here to be honest.
To me things like governance, grants, website, development, etc should be the responsibility of committers, anything else is to me very weird and not something I have seen in open source.
To me, the advisory committee has been designed from the start as an executive body, which is therefore concerned governance, development and all other operational aspects of the project. I don't see how the advisory committee could "Provide guidance and oversight of the Project’s staff and operations" in a software project if it is not entitled to talk about governance or development.
The committee was set up from CS&S' template for fiscally sponsored projects, which we adopted probably too eagerly, because we didn't have enough experience with it, and we trusted their model to be broadly appropriate. There are many improvements we could make, first with the term "advisory" which is a really poor fit (swapping the names of the advisory and steering committees would already have been an improvement…).
@abbe98 do you have any examples of open source projects which use a governance model that would be a good fit for OpenRefine in your opinion? What do you think of the changes I proposed earlier?
To me, the advisory committee has been designed from the start as an executive body, which is therefore concerned governance, development and all other operational aspects of the project. I don't see how the advisory committee could "Provide guidance and oversight of the Project’s staff and operations" in a software project if it is not entitled to talk about governance or development.
This to me is very concerning as you are the only major committer on the AC and while I find your work very interesting most of your effort is spent on your own thing on your own branch with your own funding.
To me an open source project should be governed by those who contribute and transparent in its decision making and financials, the setup with CS&S and the AC is the opposite and results in less transparency than that of a public company.
Reality is that OpenRefine has very few active contributors, and I question if complex governance models would even be beneficial for the ~5 people who contribute actively or if consensus among these are a better model.
That said I'm quite fond of GNOMEs model, but but I think the effort of enforcing such models is to much for a project like OpenRefine.
There have been a number of discussions about governance over the years. In addition to the conversation that Antonin highlighted, there are, at least:
The original governance model was an Apache-style consensus based meritocracy and I still like that.
The ASF's successful collaborative, meritocratic technology and community development process known as "The Apache Way" is both highly emulated by other open source foundations and the subject of numerous industry case studies and business school curricula. Each Apache project is overseen by a Project Management Committee (PMC), a self-selected team of active contributors that guides the project's day-to-day operations, including community development and product releases.
The Eclipse Foundation has a similar, but more heavyweight and more corporate-focused, model. Many of the important attributes such as committer elections are identical or very similar to Apache's. Obviously, OpenRefine, as a single project, doesn't need some of the higher level multi-project infrastructure.
Because the Apache Software Foundation's primary goal is to produce software, the focus is on software development and the roles (developer, committer, lead) reflect that. Other communities have different focuses. For example, Wikipedia is focused on producing (curating) knowledge, so editor is a primary role and tool developers exist principally to support editors. Antonin's proposal includes a variety of additional roles, including trainer, translator, etc (called "teams" in that proposal).
I don't think contributors are limited to those who write software, but I do think that those who contribute should drive the project. I agree with Albin that the governance processes (and everything we do) should be scaled appropriately for the size of our (small) project. It's great to have ambitions to grow to the scale of Wikipedia, Apache, or Eclipse, but that's not where we are today.
One of the main reasons that governance is important is for sustainability, but I'd argue that growing the contributor community is even more important for sustainability. Solid governance is a tool to encourage and support that growth.
Tom
Sorry for the long answer; I have been iterating on it since Friday and kept enriching it from more recent conversations. I will break down my answer in
From the current draft mission statement is (see latest proposed version here it may be slightly revised based on the survey results) OpenRefine primary goals is about Empowering everyone to meaningfully engage with data by providing an accessible open source tool and nurturing a diverse, supportive community., which include a software and community support part.
I agree. The issue today is that many non-code contributors do not feel like they are part of the community or have a voice in governance. As reported by Bocoup, currently: "Several interviews also mentioned wanting it to feel like a proper community, rather than a collection of users. There was a theme around wanting community members to feel like they’re part of a project where developers are listening to them about what to prioritize, and where there is clarity on how to interact and exchange ideas."
As part of my work on the contributor handbook, I aim to provide them with clear pathways so they understand how they can get involved and we have more diverse voices (see point 2. below).
The OpenRefine community is dispersed (see my attempt to map it) but not small. The recent Barcamp brought together 29 contributors at various levels; this count does not include many current contributors who were absent. I estimate the total number of contributors to be closer to 50 than 5.
As far as I know, there is no ambition to become as large as Wikipedia, Apache, or Eclipse.
@thadguidry in your this GitHub comment you mentionned
I wish now for the old days when we had a very good form of simple meritocracy.
I do not agree that our previous model was better. I think our old model was deeply broken by being nontransparent regarding its process and empowering a few vocal individuals. I am resharing here feedback I received last year in User Interviews Results Part 3: Cultivating a Thriving Developer and Trainer Community
Three experienced contributors to the project have noticed that some communication between team members can be off-putting or aggressive at times. They also believe that to contribute to the project in the long term, one needs to be persistent and tenacious. They are also concerned about how newer contributors perceive those exchanges.
I think the usage of the terms persistent and tenacious show that we are failing somewhere in our onboarding process.
I believe members of the Advisory Committee (all volunteers except for Antonin) are also OpenRefine contributors and that they have a voice in the governance, development, grant applications (for which they will be responsible for fund administration), and the website like other contributors.
@antonin_d I am not sure which document you are referring to. When we joined CS&S we only added the Advisory and Steering Committee to our governance document. We started to work on the GOVERNANCE in 2014, as described in the history below.
However, I agree that we are far from having perfect governance, and there is plenty of things to improve. For me, the top three issues are that
Since September 2023, I have been exploring the creation of contributor pathways. I understand that I've taken the long road, and things take time to take shape. I hope my approach addresses the root cause and ensures we listen to all voices and explore different options.
This post Proposition for the Ambassador Council - #8 by Martin highlights what I expect from the pathways.
For the following three months (October to December 2023) I worked on how we turn this plan into action (see Reflecting on 2023 and Looking Ahead to 2024 as OpenRefine Project Manager)
At the beginning of 2024, I focused on building the foundation of that approach via
I recently shared my progress here.. So far, I have received limited but constructive and supportive feedback to improve the different components.
For the rest of the year, I plan to continue working with the community on updating the roadmap, project roles, and governance.
I welcome this conversation, appreciate the engagement on this topic, and welcome feedback or comments on the existing draft documents. I want to better leverage our OpenRefine community meetup to have an hybrid conversation (call with minutes + forum discussion and PR) to progress on those important topics.
Here are links to the different conversations and updates to our governance documents:
April, 2014:
Apr 2015:
Nov 2017: Fwd: Inviting existing contributor to become committers.
Dec 2017: When Google News Initiative donated USD 100,000 OpenRefine was not a legal entity. @thadguidry created an LLC in the US so we can have a dedicated bank account for the project before the Dec 31, 2027 deadline to receive the money. We knew that was short term and went to seek for a long term solution by joining a fiscal sponsor with SF Conservancy (which did not work out) and then CS&S.
Jan 2018: Create GOVERNANCE.md by magdmartin · Pull Request #1435 · OpenRefine/OpenRefine · GitHub
Nov 2018:
Nov 2019:
Jan 2020:
@Martin has replied in far more detail than I could hope to dig into (thank you for that!), so I just want to add a few points.
@ej2432
1 . I didn't see your request as a critique. I think it is important that full history is available to everyone in an easy-to-access way. @thadguidry @thadguidry, let me know if I forgot important documents in my previous message.
2 and 3. I agree with you. I invite you and anyone interested in those questions to continue the conversation here Improving the onboarding process for new contributors - #5 by antonin_d
4 . The project is already financed via The EOSS-Diversity grant, already includes improving the project governance, and a significant part of my time for the rest of 2024 will be dedicated to it. I am unsure if we need additional funding and how new funding would be used.
Thank you, @Martin ! How fortunate that there is already funding to support your doing this work.
@ej2432 I think we might be agreeing or maybe using different ways to express the same thought? But one slight concern is when I see any wording that gives the Advisory Committee any power to chart a course, or set a direction of where OpenRefine will go (It's just a slight concern, because I know the community will ensure the governance is worded well once the draft rewrite is ready for 1st review by the community)...
That doesn't mean every aspect of the project direction and management falls under the Committee,
I don't think project direction (OpenRefine being the output of contributors and their agreement to add, review, merge what they agree with the community as being beneficial) will ever be directed or dictated by a single organization other than the community.
Do you agree with this kind of phrasing instead? -
The "project direction" of OpenRefine is set forth by the community alone in a process of meritocracy, but where the community and contributors take into account guidance from other organizations and partners, such as the Advisory Committee.
@thadguidry we are in agreement that the Advisory Committee should never be the only group of people directing the OpenRefine project. I'd probably want to iterate over that language collectively. The Advisory Committee is currently (and may still be) tasked with responsibility for some specific things related to this project. They do not advise on those things, they direct. Much of this will - I trust - be sorted out between the larger community. I think that your strong vision of the governance can only be helpful as we move forward, keeping open minds.