Daniel-Mietchen
commented
2 years ago Reframing of the talk:
Title
The ecology of FAIR Digital Objects, with special attention to roundtripping and benchmarking across the research ecosystem
Abstract
The more FAIR a Digital Object is, the more likely it is to interact with other entities in the research ecosystem and beyond. As long as the interoperability of these entities is not perfect (and it rarely is), a variety of interactions with a given Digital Object will mean a variety of representations of it, with some closer to the original than others. Some systems will even gather multiple representations of a given original (or aspects of it), creating the need to assess similarities, differences and relationships and to include them in curation, management, education, dissemination and preservation workflows.
In this presentation, we will look at the suitability of sets of FAIR Digital Objects to serve as indicators for several aspects of FAIRness across different elements of the research ecosystem. These sets could involve existing FAIR Digital Objects as well as new or hypothetical ones, and inclusion or exclusion with respect to a given set could be defined using a wide range of criteria pertaining to the ecosystem elements of interest. Taking inspiration from tracing, monitoring, benchmarking and roundtripping activities in various research fields, we will then explore how far, how well and how quickly such sets can travel through multiple elements of the research ecosystem (e.g. different databases or software pipelines or different stages of the research cycle) and what this means in terms of potential improvements to the FAIR Digital Objects themselves, to the sets, to the way the assessments are run or to relevant elements of the research ecosystem.
I had initially made four submissions to the 1st International Conference on FAIR Digital Objects, which now need to be restructured into one presentation and up to two posters. In order not to loose information, I am thus posting the original content of the submissions here, which will serve as a basis for the rearrangement.
Challenges and opportunities of roundtripping and benchmarking FAIR Digital Objects across the research ecosystem
The more FAIR a Digital Object is, the more likely it is to interact with other entities in the research ecosystem and beyond. As long as the interoperability of these entities is not perfect (and it rarely is), a variety of interactions with a given Digital Object will mean a variety of representations of it, with some closer to the original than others. Some systems will even gather multiple representations of a given original (or aspects of it), creating the need to assess similarities, differences and relationships and to include them in curation, management, education, dissemination and preservation workflows.
In this presentation, we will look at the suitability of sets of FAIR Digital Objects to serve as indicators for several aspects of FAIRness across different elements of the research ecosystem. These sets could involve existing FAIR Digital Objects as well as new or hypothetical ones, and inclusion or exclusion with respect to a given set could be defined using a wide range of criteria pertaining to the ecosystem elements of interest. Taking inspiration from tracing, monitoring, benchmarking and roundtripping activities in various research fields, we will then explore how far, how well and how quickly such sets can travel through multiple elements of the research ecosystem (e.g. different databases or software pipelines) and what this means in terms of potential improvements to the FAIR Digital Objects themselves, to the sets, to the way the assessments are run or to relevant elements of the research ecosystem.
Integrating research ethics workflows with FAIR Digital Objects
The process of reviewing research - whether finished, ongoing or planned - has a number of components that would benefit from closer integration with the FAIR research landscape. These components include the circumstances suggesting or even requiring an ethical review, the types of information that need to be exchanged during the process, the types of communications set up to convey said information, the stakeholders involved in any part of the process, the ways in which metadata about the process is stored and shared, and rules that govern any of these aspects and related matters.
In this presentation, we will look into some of the key components of the ethics process and discuss what the benefits and risks would be of making more use of FAIR Digital Objects in these contexts. These questions will be discussed from the perspectives of several stakeholder groups, e.g. researchers, research subjects, research administrators, reviewers (on ethics committees or during manuscript or grant proposal review), data stewards, tool developers, science journalists, ethics educators and others.
In particular, we will consider the potential of a more FAIR ethics process to reduce the burden on the stakeholders involved and to make their participation more meaningful, while raising compliance with applicable regulations, increasing the speed and transparency of the process and improving documentation and standardization.
Research data policies as prime candidates to become FAIR Digital Objects
As the FAIR Principles reach farther and deeper into the research ecosystem, they are increasingly reflected in policies and other regulations. Yet most of these regulations are themselves not FAIR, thereby hindering uptake, arguably causing friction in alignment with current practices, missing educational and community engagement opportunities and hampering efficient monitoring of compliance or systematic identification of potential policy improvements.
In this presentation, the focus will be on identifying some key elements of policies and regulations pertaining to FAIR Digital Objects and on exploring how the role of these policies and regulations would change if their key elements would increasingly involve FAIR Digital Objects. These explorations will cover both technical and social aspects: what mechanisms are available and already used to increase the FAIRness of policies? Does it help or hinder if certain aspects of the transition to a FAIRer ecosystem are shared in a more or less FAIR way or with shorter or longer delays? Does having more FAIR policies themselves provide funders, institutions, publishers or other organizations with more of an edge or a handicap in terms of assisting their respective communities in the transition towards more FAIRness in their respective corner of the research ecosystem? How can the design of FAIR policy elements be tailored to optimize learning opportunities for specific stakeholder groups pertaining to specific types or collections of FAIR Digital Objects?
Connecting FAIR Digital Objects with communities and across FAIR silos
The last few years have seen considerable progress in terms of integrating individual elements of the research ecosystem with the wider FAIR landscape. This integration process has lots of technical as well as social components and ramifications.
As the volume, breadth and depth of FAIR data and the variety of FAIR Digital Objects as well as their use and reuse continue to grow, there is ample opportunity for multi-dimensional interactions between generators, managers, curators, users and reusers of data, and the scope of data quality issues is diversifying accordingly.
In this presentation, we will look at two ways in which individual collections of FAIR Digital Objects interact with the wider FAIR research landscape. First, we will consider how individual FAIR Digital Object collections interact with communities that curate, generate or use data, metadata or other resources pertaining to said collection: which aspects of a given community's interactions are affected by different degrees of FAIRness of parameters of the collection? Second, we will consider the case of communities that overlap across FAIR collections and what this means in terms of challenges and opportunities for enhancing interactions between and across FAIR silos.