During the second half of the workshop, we held so-called “data sessions” for each dataset selected from our open call for datasetsLink opens in a new window. This is a collaborative format for data interpretation, which was developed by Harold Garfinkel and colleagues in the 1970s (Heritage, 1984), and allowed us during the workshop to explore with participants how open interpretative research could be advanced using these datasets. To facilitate this, we provided the working groups with a set of questions to provoke and guide the exploration of their date (Figure 1). Alongside this “question sheet," we also provided participants with an index of each dataset, listing its contents, as well as examples or ideas for opening it up to encourage discussion (Figure 2).

Figure 1. Question sheet for data sessions. Visualisation by Ginevra Terenghi.

Discussions during the data sessions explored challenges in archiving, sharing, and ethically reusing diverse datasets, highlighting issues of accessibility, standardization, and consent.

The session on the ecopoetics dataset surfaced the issue of inconsistencies in metadata for data collections that have been developed organically over time. We also discussed the trade-off between audio fidelity and ease of sharing, and the need to better connect different components of a mixed media data set—recordings, spectrograms, and poems—to enhance public accessibility. The session on the Data and Displacement dataset problematised the concept of Open Interpretative Research, and discussed broader accessibility and ethical concerns like consent, misinterpretation, as well as structural inequalities arising from efforts to “open data” for differently positioned data subjects in the Global South and the North. What started as a conversation about open interpretation thus focused attention on researcher responsibility, the requirements of proper contextualization for qualitative datasets and alternative formats as solutions. The session on the Stochastic Parrots Twitter dataset debated the ethical complexities of re-purposing social media data, particularly de-anonymization, consent, and evolving platform policies, emphasizing the need for sustainable infrastructures. A recurring theme across discussions was the need for clear guidelines and standardized practices to balance accessibility and reuse with ethical responsibility in data sharing.

The explorations and discussions during our data session informed the “Case studies” that the project team subsequently created for each of the three open datasets.

.