Power-sharing in online communities
A lone protestor draws her sign in front of the Supreme Court in Washington DC, the sign is on its head but reads 'this is not normal'[1]
In some ways, there has been little news in the iNaturalist 'GenAI' story[1]: Beyond some vague posturing towards _"going to work on improving ways to get feedback from the community"_, saying (but not demonstrating) that they are 'listening', and somehow doing some 'webinar' at some point in the future, there's been very little commitment from any iNaturalist staff or leaders.
Of particular note to me, there's been no mention of actually involving the community in the decision-making process beyond the mentioned asking for feedback (and my question regarding any plans to improve this haven't been answered at time of writing).
For many years I've been _'joking-but-not-joking'_ how crowdsourced citizen science often devolves into a _'Amazon Mechanical Turk'_-style micro-tasking – just with worse pay.
Of course, that's a bit hyperbolic.
But it remains a strange disconnect that a discipline like citizen science, that makes claims towards _'democratizing science'_, as a rule typically stops at sharing real decision-making power with participants.
Which is particularly the case for large, open-ended online citizen science platforms like _iNaturalist_, that want to work with large groups of people over a long period of time.
These tendencies are not unique to citizen science efforts.
After all, many large free/open source software projects have long been run under the _Benevolent Dictator for Life_[1]-model (BDFL), which could be seen as continued trend towards an _inverted Conway's law_, where governance is shaped by the technical affordances of a admin/user dichotomy [^1].
1: _Benevolent Dictator for Life_
Similar to the situation with citizen science projects, like in the case of iNaturalist, disputes in open source can erupt over the direction of a project and often come down to who can make which decisions.
In those cases, conflicts are then ultimately resolved through making forks[^2] Recent(-ish) examples of that include Forgejo[1] and just a few weeks ago CoMaps[2].
[^1]: Conway's law[1] states that technical structures tend to reflect the social structures of an organization (like a website that is designed to follow along the internal organizational structure, e.g. across departments, regardless of whether this fits user needs. But I think increasingly we tend to see the inverse: The types of decision-making/governance/power-sharing that we can envision is being limited by what technological platforms show-case to us or make easy to put into practice – not by what we would actually want or need.
[^2]: Often these conflicts might also escalate around licensing. But while licenses are one tool in the larger governance tool kit, those can only be necessary but not sufficient for governing a project, be that in open source or citizen science.
Forking a project is one thing, but that only begs the question: What approaches to sharing power with a community could a project take, in order to not stand at the same 'fork in the road' a few years down the line, just with a different set of BDFLs calling the shots now? While there's no one-size-fits-all answer, there are different options that have been tried and tested in different settings.
Before we can dive into that, we should look at different organizational setups though: _iNaturalist_ is by now a US-registered, 501(c)3 non-profit, a type of organization that can give itself its own bylaws.
A common setup for this type of structure is to have what's called a "self-perpetuating board", which is just a fancy way of saying the board of directors of an organization keep voting for their own replacements, without any external votes cast.
This lack of democratic influence can be useful to avoid hostile take-overs, but also means that non-board contributors have zero power to shape an organization.
For some settings this can be the right setup, but it means that if an organization does have members, this membership is without any teeth (see Bradley Kuhn's write-up on what goes wrong if one tries to pretend otherwise, in that case with the _Open Source Initiative_[1].
If one hunts down the iNaturalist bylaws[1] [^3], we find a quite similar type of self-perpetuating board, with the exception that staff members of the organization also seem to have voting rights.
1: hunts down the iNaturalist bylaws
Again, and just to be clear, there's nothing wrong _per se_ with having a self-perpetuating board, but it opens an organization that tries to center community up to the type of power-sharing breakdowns I'm discussing here.
Especially if one claims a community of thousands or even millions of 'members' who are effectively not given any influence [^4] So what's the alternative? An easy one is giving members a vote in electing the board of directors, which is after all the highest organ of decision-making for a non-profit like that.
There's different ways of implementing such a thing, depending on the type of 'members' one has: The bylaws of Wikimedia[1] (also a 501c3) make clear that the organization doesn't have any traditional 'members' in non-profit speak, but it does recognize it's millions of contributors.
Half of the board-seats are reserved for _Community- and Affiliate-selected Trustees_ which are nominated and selected by active contributors to Wikimedia projects.[^5] While the final approval of those trustees is subject to oversight by the final board, it still gives a large weight to the community in directly shaping the organization.
Inspired by this, we have implemented a similar model for Open Humans[1], where a certain number of board sets are elected directly by the community.
In that case, no 'traditional' membership to the 501c3 was taken as definition, but instead a 'member' is anyone who has an active (as in: not banned) account.
These two examples give an idea how organizations that run large online platforms with many active participants/users can give their contributors an active decision-making role, which can help serve as a corrective by sharing power.
And another adjacent example comes from the Open Bioinformatics Foundation[1] (OBF), where I am a board member. Unlike the previous examples, the OBF does have an explicit member status for people aligned with the mission and goals of the organization.
1: Open Bioinformatics Foundation
While the board of the OBF is self-perpetuating as well (the membership can nominate candidates but not vote), the organization also includes a way to hold membership referendums, for example for adopting a new code of conduct[1].
1: membership referendums, for example for adopting a new code of conduct
In that way, the membership has a more direct, voting function to act on their decision-making power.
Outside the US, there's a few more examples of community-based organizations that are doing good work with involving members in decision-making: Both OpenStreetMap[1] and Codeberg[2] have formal membership roles as well.
Paying members[^6] are in both cases invited to the general assemblies to vote on concrete questions and elect the board of directors, handing ultimately all power over to the membership.
Importantly, in both cases not every user or contributor to the respective platforms needs to become a member, but it's an option that's open for those who want to engage with decision-making.
And maybe closer to home for citizen scientists: The _European Citizen Science Association[1]_, while mostly made up out of academic, uses the same approach for its governance.
1: European Citizen Science Association
There are many ways with a wide variety of options for sharing power with community.
And while iNaturalist just celebrated it's 250 millionth validated observation and by extension it's community, I'd say there's still lots of room to improve its own sharing of power. These examples hopefully show that it can be done, time will tell if there's any willingness. But regardless of that, it's maybe worth to look at the organizations you're part of and see if and how community decision-making is implemented – and how it could be improved.
[^6]: In case of OpenStreetMap there's also a way for active members to become voting members without paying membership dues
[^3]: They unfortunately aren't linked anywhere from what I can tell.
[^4]: This also applies to non-profits who serve millions of 'users', thinking of the Mozilla Foundation that seems to need a lesson in that too!
[^5]: Active membership is defined by making enough contributions[1], e.g. by editing Wikimedia wikis such as Wikipedia, helping translate, contributing code.