My first encounter with the open movement was being involved in standards development at the World Wide Web Consortium (W3C) while completing my Ph.D. almost thirty years ago. Since then, I’ve produced open content, written open code, and published open data. I’ve attended Open Knowledge conferences, shaped open government data initiatives, led the Open Data Institute, become a board member of Creative Commons and gained a fellowship at the Shuttleworth Foundation. Cut me and I bleed openness.
But, like the authors of the Paradox of Open essay, I find the digital world we live in is very different from the one I imagined we were building. The thousand flowers have bloomed into democracy-warping mis- and dis-information. Instead of being empowered by the open web, mega-corporations exploit communities whose billionaire leaders amass sickening levels of private wealth. Instead of feeling free, we limit our own self-expression in the face of mass digital surveillance, striving to please the algorithm in pursuit of likes and follows.
Is this because openness is the right thing to aim for, but we’ve failed at gaining adoption – are we not open enough? Or, conversely, has openness actively contributed to the problems we see – were we wrong to aim for it in the first place?
In this response to the Paradox of Open, I argue that opening assets is just one part of a broader goal to enable and nurture creative communities. And that this perspective gives a different way of viewing the largest digital platforms, and a more targeted response to their dominance.
Openness is about relationships
Why do we care about openness? The reasons I hear given for being open can be grouped into three categories:
- openness and transparency support accountability – it enables us to delve inside the workings of an organisation or a piece of code to understand it better, detect where things are going wrong, and do something about it;
- openness removes friction and reduces duplication – this is a more self-interested purpose, especially relevant in large organisations where there’s a need to break down silos, or in pre-competitive spaces where joint activity leads to joint benefits;
- openness enables creativity and innovation – with the idea that people can reuse and build on existing work to create new things.
In all three cases, openness is really about strengthening our relationships with each other and our sense of community and shared purpose. We’re transparent about what we’re doing to build trust with each other. We remove barriers to our resources to make collaboration easier. We share our work in the hope of recognition and kudos. Openness helps us work together.
This link between openness and our relationships works in both directions: being more open strengthens our relationships, but stronger relationships and communities – ones built on trust, recognition and shared purpose – also help us to be more open. It is a positively reinforcing loop, but one that is hard to enter. Whether in personal relationships with each other or in organisational collaborations, taking the first steps towards openness makes us vulnerable and feels risky.
I think this explains why openness has been successful where it has been, as identified within the Paradox of Open essay. Commons projects such as OpenStreetMap bind people and organisations together with a common purpose without much risk to those who take part. External incentives help to counteract the risks of openness with the promise of reward (or avoidance of punishment).
But if our fundamental goal is to create more relationships with each other, to build trust, enable collaboration and unlock creativity, we need to think more widely about the kinds of openness we need. We also need to consider the kinds of technological platforms, legal mechanisms and organisational structures that are necessary to support it.
Openness is about building community
Building relationships requires a different focus on our openness, and recognition of new kinds of challenges they give rise to.
Shallow models of openness are distant, uni-directional, and based on generosity. A creator chooses to make a resource they own and control open to others to reuse. A reuser takes that resource and does something with it. The creator may not even be aware that their resource has been used by someone else.
Deeper models of openness are close, bi-directional and based on mutual interest and purpose. Creators are not just open with the resources they hold, but also open to those who build on them. They might encourage pull requests on open source code; integrate improvements to open datasets; or reference and build on derived content in their own future work. Creators build relationships with reusers, and the lines become blurred between the two, such that resources become more like community assets.
As Leigh Dodds has explored in his blog posts on the topic, community building needs more than open licensing. Alongside technical tools, we need ways of defining and explaining social norms and expectations about contribution and value sharing. We need collective mechanisms for decision-making about shared assets, that make people feel heard while recognising the limitations, constraints and realities of their maintenance. We need people and organisations to see themselves as conveners rather than publishers, as contributors rather than reusers, and for everyone to take responsibility for growing and strengthening their relationships with each other.
Openness on the big platforms
At a surface level, many platforms seem to have done a reasonable job at furthering this cause of relationship and community building. Whether it’s collaborating on code through GitHub, on documents with Google Docs, or even just talking on Twitter – these platforms enable far greater opportunities for collaboration and community than we would have without them.
This is even more true for my kids’ generation. I am fascinated by the way in which TikTok enables creativity and community building by supporting the reuse of soundtracks, stitching, and dueting, leading to cultural phenomena founded on collaboration, such as the Ratatouille musical or Sea Shanty TikTok. While we characterise digital platforms as walled gardens, the Minecraft YouTuber (mcyt) and other fandoms are not limited by those walls: they stream on Twitch, create animatics on YouTube, post fan art on Twitter and Instagram, make reaction videos on TikTok, discuss them on Reddit and Discord.
But below the surface lurk two challenges where I believe the open movement needs to place more sustained attention.
First, there is the ownership and control over the use of assets created on and by these platforms. In most cases, the terms and conditions that we agree to when using digital platforms – or indeed the licences that we adopt – grant both platform owners and third parties permission to reuse our assets in ways that may technically meet licensing conditions, but don’t feel fair. A recent example is GitHub’s CoPilot service – an AI-generated from code hosted on GitHub that can predict the code you want to write, which has prompted complaints from developers about the reuse of their code. Other examples include the use of openly licensed images to create machine-learning datasets, the commercialisation of games created by children on Roblox, and, more broadly, the exploitation of indigenous knowledge.
These examples don’t feel fair to creators because, in them, organisations are extracting value from, rather than being part of and bringing value to, a mutually sharing community.
Our commons are being enclosed and privatised. We wanted openness to provide raw materials – content, data, code – to those who lacked them, enabling them to create and do things previously out of reach. Instead, it has enabled those who already have wealth, power, and capability to gain even more. We wanted open resources to be used for justice and the public good, but they have also been exploited for oppression and private gain. Our openness has been used to create private assets – including data about us – which are then exploited by platform controllers. The rising tide has lifted all boats, but particularly the superyachts.
I can see four responses to this, all of which will need work:
- formally encoding our community norms into the policies and processes around gaining access to the content we create, for example by including ethics clauses in licences (of which restricting to non-commercial use is the most prevalent), or by putting in place governance gates that assess requests for access to content against frameworks such as the Five Safes;
- layering on and enforcing separate sets of norms around the reuse of open content, whether that’s softer social norms such as naming and shaming, or harder mechanisms such as regulation (data protection laws being one existing example);
- requiring organisations to return value to the commons by gaining more access to and control over privately held data and content, such as by promoting share-alike licences more, or through regulation that enables, encourages and enforces public access and governance, particularly when it is used for the public good and towards justice; proposals for B2G data sharing are part of this effort;
- directing more resources towards bolstering the capacity for those without them to take better advantage of open resources.
But as I have argued, openness is about more than access to and use of assets. It is about how we build and maintain our communities. The second challenge, then, and one that is just as important, is to re-examine how we create a community in the digital world.
The internet has always been a place where people can find each other and form communities, from bulletin boards and mailing lists to Black Twitter, Queertok and 4chan. Early platforms were neutral message carriers. Modern platforms have intermediated our relationships, reducing their depth and complexity into engagement metrics; setting themselves up as the universal arbiters of community norms through algorithmic amplification and content monitoring; and failing to distinguish between people and communities that counter oppression and those that promote it.
While we might use these spaces, open communities have particular needs. They are formed not from shared identity but from entwined attention and collaborative work on shared resources. They thrive on diversity and so require the creation of an environment that is proactively inclusive and welcoming. They need fluidity, iterativity, and a self-organising, distributed, emergent activity that draws on theories of social innovation and new power.
So while we continue to develop our ability to access and use common resources, we must also create the spaces – and the technologies, agreements and governance that support them – to build communities around them. These might be local communities stewarding sensor data from their neighbourhoods to lobby for better environmental regulation. They might be worker communities like those enabled by Worker Info Exchange using data to negotiate better working conditions. Or they might simply be creator communities like the SCP Foundation building shared worlds and stories.
When I first worked within the W3C, I was a twenty-something woman, working from my bedroom, contributing to global standards alongside established engineers in international software companies. I found a creative community there, on mailing lists and IRC chats, one that did not just talk but also made something, together. And it has always been that combination of shared purpose and collaboration around a common good – from up-to-date legislation to electoral candidate datasets to board games – that has sparked invention and brought me joy.
Jeni Tennison is the founder and Executive Director of CONNECTED BY DATA, an initiative that aims to put community at the heart of data narratives, practices and policies. She is the co-chair of the Data Governance Working Group at the Global Partnership on AI and a Shuttleworth Foundation Fellow. She is an adjunct Professor at Southampton’s Web Science Institute and an associated researcher at the Bennett Institute for Public Policy where she examines the practice and public policy behind collective data governance.
Jeni was CEO at the Open Data Institute, where she held leadership roles for nine years and worked with companies and governments to build an open, trustworthy data ecosystem. She sits on the Boards of Creative Commons, the Global Partnership for Sustainable Development Data and the Information Law and Policy Centre.
She has a Ph.D. in AI and an OBE for services to technology and open data. She loves Lego and board games and is the proud co-creator of the open data board game, Datopolis.
We asked leaders and experts from the broad open movement what the Paradox of Open means to them and how to address the challenges that it poses. This essay is one of the responses.
