It’s a chilly Tuesday in Frankfurt, the Matildas just advanced to the final of the 2026 Women’s Asian Cup, and I’m at the opening of the Social Media Access Days at the German National Library, co-organised by my dear friend Katrin Weller from GESIS, the Leibniz Institute for the Social Sciences. The programme begins with a day in German, and opens with a paper by Pascal Siegers, who introduces the AVERA project. This emerged from a federal ministry project supporting the collection and sharing of data from research projects on racism and far-right extremism, and a first need it identified was the establishment of a list of actors in far-right communication networks.
Such lists do exist, but are linked to individual projects and rarely shared publicly; the solution developed for such sharing is the establishment of a data trustee structure that operates on a sound legal basis and incentivises researchers to share their data. This also implies a collective responsibility for this scientific data ecosystem, and perhaps also serves to improve the overall data quality.
This turned into a data portal for racism and right-wing extremism research, supported by a data trustee model, onboarding programmes for researchers, and added server security measures, operating under the name AVERA. It provides a list of actors and accounts required for researching the spread of far-right online activities, radicalisation processes, narratives, and discourses.
Data and their quality are constantly verified and updated, using the VrAN collaborative social media account editor which enables researchers and data trustee curators to search and manage the datasets. This work also required specialised legal clearance: while personal data are generally protected under the EU’s GDPR, a Legitimate Interest Assessment (LIA) that affirms the societal relevance of these accounts (since they represent right-wing extremists) enables the collection and sharing of this information under a carefully designed, secure vetting and onboarding process.
That process was designed using a co-creation process with the relevant research community: from this emerged the key principle of reciprocity amongst participants (requiring users to give back by verifying and updating some of the actor lists accessed by participants); the visibilisation of participating researchers by making AVERA a citable resource and enhancing the reputation of key participants by acknowledging their contributions; the development of trust through participatory governance; and technical support for data management processes.
The current dataset contains entries for some 2,100 entities (individuals, organisations, accounts) across key mainstream and niche platforms (Telegram, X, TikTok, etc.). It does not collect data for researchers, but enables their data collection processes by providing them with a reliable list of actors whose activities can be tracked across social media platforms. This is a new data infrastructure approach, compared to projects that provide a central dataset and limit how such data may be used by accredited researchers. It does not prevent such further data sharing projects, of course, and the sharing of social media activity data collected about these actors could in fact be a further extension of the infrastructure.
