The participants of the workshop that took place within the framework of the ROSiE project, formulated general recommendations for the development of responsible Open Science Infrastructures (OSI), as presented below:
- Training should be developed (both for users and for platform makers);
“Yeah, on helping researchers integrating this whole idea of FAIR data and OS into their real activities when they perform research, beyond a single experiment.”[1]
- Particular attention has to be paid to the plurality of disciplines and issues related to OS, i.e., OS should be thought of in a “cross-disciplinary manner”;
- More resources should be dedicated to OS;
- The CARE principles should be added to the FAIR principles in order to better take into account the human that generated the data[2];
- When guidelines on responsible OS are developed, they should target specific aspects (e.g. data collection, consent forms, legal aspects, etc.), and they should be operational, up-to-date and ready to use materials.
- Responsibilities should be distributed and infrastructures should collaborate.
Participants recognised that responsible OS is only possible on the basis of a shared responsibility. They stressed that platforms and infrastructures do not have the same roles and therefore not the same responsibilities. For example, regarding data quality and integrity, while infrastructures are responsible for being transparent about their processes in order to allow users to assess themselves the data quality, platforms would have a role to play in managing data or guide data management. Platforms participate in the responsibility of data integrity, but they are not the only one responsible: owner(s) of the data, but also scientists and citizens who collect it, have their role to play. A participant mentioned that it is not possible for a single infrastructure to meet all the challenges of responsible OS: this requires the collaborative work of several infrastructures. The DMP was recognized as a good way of determining who is responsible for what, and to do so upstream of the data collection.
It should be noticed that the participants did not mention technical issues or solutions either spontaneously. When they were specifically questioned on these matters, technical options did not seem to be a priority in addressing the above-mentioned challenges.
[1] All quotes provide from the discussions in the workshops.
[2] The CARE principles (Collective benefits, Authority to control, Responsibility, Ethics) were developed by the Global Indigenous Data Alliance in order to ensure better inclusivity and better consideration to indigenous communities in open science and data management. See: https://www.gida-global.org/care
This passage is part of D6.2: Final analysis and mapping of existing European and national OS infrastructures with regard to promoting responsible OS written by Carole Chapin, Nathalie Voarino.