Collaborations Across Borders and Disciplines: Insights from Open-Source Investigations

[Isabella Regan is PhD researcher at the department of Law, Society and Crime of Erasmus School of Law, Rotterdam, focusing on public and private open-source investigations and atrocity crimes.

Alexa Koenig, PhD, MA, JD, is a research professor at UC Berkeley School of Law, faculty director of UC Berkeley’s Human Rights Center, and director and co-founder of the center’s Investigations Lab.

Brianne McGonigle Leyh, PhD, MA, JD, is professor of global justice studies at Utrecht University’s School of Law, director of masters education at the School of Law, and project lead of the Global Justice Investigations Lab at Utrecht University]

In June this year, Utrecht University hosted a conference on open-source investigations with dedicated panels made up of a wide spectrum of practitioners and researchers. The diversity of organizations and disciplines represented at the conference underscores the inherently hybrid nature of open-source investigations, where journalists, human rights NGOs, legal experts, data specialists, policy analysts, students and educators all occupy the same ecosystem. During the conference, discussions on the ‘multidisciplinary’ nature of the open-source investigation (OSI) field raised a fundamental question: As OSI develops into a space of intersecting disciplines and practices, how can we cultivate collaboration that strengthens, rather than fragments, the field?

This blog reflects on how multidisciplinary collaborations are unfolding in the OSI space. It examines the key challenges practitioners face when bringing together diverse groups and offers a set of considerations for effective collaboration. The reflections build on discussions from the recent conference, our ongoing research into OSI practice, and our experiences within the Investigation Labs at Utrecht University and the University of California, Berkeley. As the field becomes increasingly professionalized and institutionalized, we argue that thoughtful, well-structured collaboration is more important than ever. While our recommendations grow out of our own experiences, they are offered as observations for others to critique, adopt, adapt or expand upon as appropriate to their specific context.

A Multidisciplinary Field of Practice(s)

As noted by scholars of OSI (here and here), open-source investigations are by nature multidisciplinary. Unlike more traditional legal or journalistic practice, there is no single universal ‘open-source investigation’ educational ladder. People can enter this work with educational or professional backgrounds in journalism, computer science, law, conflict studies, human rights and architecture, or a different field altogether. Student cohorts within university labs reflect this diversity. Teaching teams are similarly diverse, often spanning multiple departments and faculties. The types of external partners with which labs work are also highly diverse, ranging from investigative journalists to human rights researchers, and from policy think-tanks to grassroots organizations.

This heterogeneity reflects, to some degree, the broader global open-source community. Many organizations have expanded their staffing profiles to better accommodate the diverse skills needed to do this work well, or have created partnerships with external experts. Collaboration in the OSI space thus often happens across different levels and types of expertise. The mix of disciplines within the OSI space is a clear strength; the influx of diverse backgrounds has helped shape the field into one that pushes on the boundaries of existing practices in ways that can be incredibly exciting, and can feel quite innovative. But at this crossroads, we are also seeing a clear identity and distinct norms emerge: Journalists bring traditions of transparency and storytelling, while legal professionals bring rigor in terms of evidentiary quality and documentation of chain-of-custody. Technologists bring new tools, as well as new forms of automation and visualization, while academics contribute to methodological standards (including by underscoring the importance of multiple working hypotheses and peer review), ethical reflections and conceptual clarities. Increasingly, feminist and neocolonial influence (see also here) have introduced community-centered and solidarity-oriented approaches, raising important ethical questions about access and equity.

As the OSI field continues to grow, new organizations from diverse regions are adopting open-source techniques, creating fresh opportunities for collaboration across geographies and disciplines. At the same time, shrinking civil-society funding  makes collaboration not just valuable but increasingly essential. Yet, while partnerships are vital to the OSI ecosystem, they also bring challenges. When organizational cultures or disciplinary logics fail to align, the very diversity that makes collaboration so powerful can result in friction, as detailed below.

Key Collaboration Challenges

In reflecting on the strengths and weaknesses of multidisciplinarity in OSI, we’ve identified three sets of challenges that seem especially salient: (1) speaking different ‘languages’; (2) unclear or mismatched expectations; and (3) hierarchies, including resource disparity.

1. Speaking different ‘languages’

Misunderstandings of each other’s professional and disciplinary working cultures can cause confusion and frustrations if they are not identified, made explicit, and aligned from the beginning. One of the first stumbling blocks is simply not speaking the same language. Not only in the literal sense of different native languages, but more importantly in the sense of disciplinary ‘language’ and culture. A lawyer, a journalist and a data-scientist may each use the term ‘evidence’ but mean different things by it. A journalist may prioritize ‘information sources’ and ‘leads’; a legal practitioner may emphasize chain of custody, verifiability, admissibility; a data-scientist may focus on replicability, metadata provenance, automation. Similarly, the concept of ‘open’ in open-source investigations may be understood differently across disciplines. Conceptual understandings of these terms do not always ‘travel’ smoothly across disciplines. Communication about what is meant by various concepts or terms is important to avoid issues that delay or derail a joint project. Furthermore, communication about what is permissible in one’s given profession in terms of gaining access to certain databases, using automated tools for scraping, whether it’s appropriate to use a false identity online, and how partners view ‘grey’ or ‘semi-public’ data is equally important.

2. Unclear or mismatched expectations

The second set of challenges concerns unclear or mismatched expectations on roles, ways of working and intended outputs. Collaborations taking place in fast-paced environments or investigations unfolding in real-time risk investigators overlooking the crucial step of investigation planning, especially communicating about expectations and roles. Partners with no prior experience with open-source investigations may have unrealistic expectations of what OSI can do, which can lead to unrealistic assumptions on the scope, timeline and resources of the project. For example, some may hope that OSI will produce a single ‘smoking gun’ piece of evidence, rather than recognizing its supporting role as part of a wider, multi-method investigation.

Relatedly, working across organizations with different goals, priorities, mandates and approaches can complicate deciding on shared outcomes and impact. Many collaborations will result in co-produced outputs (a report, a story, a policy recommendation), yet partner organizations may have different missions and different metrics of success. Some heavily-hierarchical organizations may have institutionalized top-down ways of working that clash with bottom up, context-sensitive practices of local organizations. Within an education-led lab context, there is an additional variable: students are learning, educators are teaching, and that pace of academic practice may differ from advocacy or journalism environments. Partners unfamiliar with academic contexts and working with students may also (and understandably) expect expert-level outputs and may overlook the educational and learning dimensions—and power—of such collaborations.

3. Hierarchies and resource disparities

A third category of challenge concerns the socio-political dynamics of collaboration. Who leads the investigation? Who gets credit for the work? Whose voice is foregrounded in the final output? And do all partners have equitable access to the necessary resources to fulfill their part in the whole? Are some individuals or organizations being exploited? In global OSI collaborations, issues of access, representation, institutional power and resource disparity repeatedly surface. Not considering these aspects can give way to persistent hierarchies and inequitable role divisions. For example, local partners may be asked to provide translation or contextual support without being given a meaningful analytic or decision-making role, thereby undervaluing their expertise and potential contribution. Or practitioners’ contribution to research may be undervalued in traditional and hierarchy driven university contexts, for example, not resulting in a byline if a major media piece is produced, or authorship credit in a journal article. The sustainability of OSI projects and collaborations also risks being undermined by unequal access to financial, technological and other institutional resources to carry out investigations. For example, project leaders may benefit from strong organizational systems yet overlook the emotional toll that viewing violent or distressing material can take on local partners who lack similar institutional support. Collaborations that fail to acknowledge and address these disparities from the start, make the sustainability of a project—let along a partnership—especially fragile.

Considerations for OSI Collaborations

To offset these issues, we suggest the following five considerations for multidisciplinary collaboration in open-source investigations. These are intended as starting points for reflection and iteration.

1. Define roles, manage expectations and align objectives

Taking time to discuss project goals, outputs, timeline, roles and working culture early on can help build mutual understanding. What is the purpose of the collaboration? Who is the end-user of this investigation (advocacy, policy, academic publication, educational output)? What are each partner’s priorities, constraints and definitions of success? Partners may find it useful to explore questions such as: “What is an ‘open-source investigation’ in this context and what is not?” It is also useful to establish what methods are acceptable (e.g., what counts as open-source, what is off-limits), how verification will happen, how outputs will be published, and which terminology will be used. Such an alignment phase could help prevent misunderstandings later in the process.

In educational settings, successful partnerships tend to be those in which external partners recognize and value the educational context. Similarly, it’s helpful for academic leads to be explicit about what students bring to the partnership, including enthusiasm and capacity for simple-yet-mission critical tasks, such as video classification or verification, rather than tasks that require time-sensitive or advanced investigative judgement.

2. Tailor the methodology and outputs

Ideally, the investigation’s methodology, analytical depth, tools and outputs align with its end goals. A highly technical data-visualization may impress, but if the audience is a local advocacy group or policy-maker unfamiliar with reviewing content in that type of format, the output may fail in its intended impact. In some cases, a concise or community-focused output (for example, a well-written essay or policy brief, in-person community outreach, a workshop) may be more effective than the visual presentation of complex technical work. Regardless of output, maintaining verification and credibility standards remains essential, even when facing time pressure. Multidisciplinary teams will ideally optimize trade-offs between speed, investigative depth and feasibility.

3. Build in feedback loops

Teams often benefit from periodic checkpoints (“are we still on track?”, “what’s working?”, “what needs adjusting?”) and excel in environments where feedback is welcomed. In early phases of collaboration, forms of structured support, feedback and mentorship are crucial. Creating welcoming spaces for less-experienced partners to raise concerns can strengthen trust and learning, especially in multidisciplinary teams where different cultures and practices reconcile. Flexibility tends to support more sustainable collaboration. If the collaborative dynamic or the project scope shifts, it can be useful for partners to revisit the expectations rather than soldier on with a rigid fidelity to a now-outdated plan.

4. Recognize limits and leverage collective strengths

It can be helpful to identify early on what an organization is good at, and where challenges may arise, as well as how strengths differ across members of a team. Where expertise is lacking in a given domain (for example, relevant cultural or linguistic knowledge, or social-media use in a region, or legal admissibility considerations), additional partners with complementary skill sets may be brought in. Openness about the boundaries of one’s own expertise often strengthens trust and, by extension, collaboration. This also helps in distributing credit equitably. Highlighting the expertise of local partners, students, and others helps to ensure they are doing meaningful work, beyond the often mission-critical but frequently undervalued work of translation or data-entry.

5. Be intentional about equity, credit and institutionalization

Attention to power dynamics, representation and credit at an early stage can foster more equitable collaboration and prevent tensions downstream. From the beginning, collaborators should agree how and on what basis partners will be acknowledged, including how authorship or similar credit will work, how data or findings will be shared, and how resource-imbalances will be managed. This includes thinking about ethics in crediting, how local partners or students are named (or not), and how the work may contribute to institutional capacity-building. Considering sustainability – how partnerships and local capacities can be maintained over time, including through joint funding opportunities – may also support longer-term impact.

Looking forward

The field of OSI is at an inflection point, moving towards higher degrees of professionalization, institutionalization and, as a result, reflection. As the field advances, participants should recognize that the OSI researcher does not have to (and should not be expected to) do everything, everywhere, all at once. Combining complementary skills, knowledge, and perspectives is key to effective collaboration.

Yet, this can be especially challenging in multidisciplinary contexts. Meaningful and impactful partnerships require clear and honest communication and attentiveness to the needs of the various organizations and individuals involved. Moreover, the realities of increasingly-competitive funding landscapes may make collaborations feel like adding extra work when time and staffing are precious. As priorities shift and internal pressures mount, collaborations may fail, especially when seen as optional rather than strategic. A structural approach to OSI collaborations may enhance rigorous, credible and meaningful investigations that leverage the richness of multidisciplinary input, rather than being hampered by it. As we look ahead, these reflections are offered as an invitation for continued dialogue on how to ensure the future is as innovative, equitable and effective as possible.