DOI: 10.3897/rio.8.e96082

Authors: Carlos Utrilla Guerrero, Maria Vivas Romero

Abstract: Arguments for the FAIR (Findable, Accesible, Inter-operable and Reusable) principles of science have mostly been based on appeals to values. However, the work of onboarding diverse researchers to make efficient and effective implementations of FAIR requires different appeals. In our recent effort to transform the institution into a FAIR University by 2025, here we report on the experiences of the Community of Data Driven Insights (CDDI), a interfaculty initiative where all university-wide research data service providers are joined together to support researchers and research groups (e.g. see research showcase example here) with all aspects concerning research data management. CDDI aims to turn all digital objects within Maastricht University (UM) into FAIR Digital Objects (FDO) and by disclosing the progress and challenges of implementing FDOs (e.g. see CDDI OSF repo: https://osf.io/398cz/), we hope to shed light on the process in a way that might be useful for other institutions in Europe and elsewhere. We initially identified 5 challenges for FDO implementation. These challenges were first a matter of reshaping the culture of science making practices to fit the FAIR principles. Additionally, it required an educational awareness within the scientific communities, and finally financial and technical tools to actually facilitate the transition to FAIR practices of science making. These perspectives show the complex dimensions of FAIR principles and FDO implementation to researchers across disciplines in a single university.

HTML

XML

PDF

DOI: 10.3897/rio.8.e95975

Authors: Mark Wharton

Abstract: Semantically-defined Digital Twins (DTs) and Fair Digital Objects (FDOs) are similar in concept. They both adopt the "Find first" philosphy and they both point to data about some entity in the real world (where "entity" is a very loose definition of any asset, real or imaginary). Are there any parallels that we can draw? Can we use a digital twin plaform to host FDOs?Are Fair Digital Objects and Digital Twins the same thing?The objectives of the GO-FAIR organisation, that data should be Findable, Accessible, Interoperable and Reusable were originally designed to help human beings find and understand data. There has always been a side-helping of desiring computers and machine intelligences to find and use that same data.The FDOs that people normally think of are datasets, probably time-series data collected over a considerable period, amounting to a large file or a database of records.What if you took the concept of FDOs and compared it to Digital Twins? If you define a Digital Twin as a virtual representation of a real-world object, entity or concept in that its identity, metadata and data are stored “in the cloud”, then it’s not a big jump from that to an FDO. If you approach it from the opposite direction - reduce the size of an FDO until it’s an individual entity, not a collection, then you arrive at the same destination.In our proposed talk we will outline the ways that FDOs and semantically-defined digital twins overlap. We will show real-world examples of Digital Twins in action and argue that an FDO of sufficient granularity would serve a very similar purpose.The Venn diagram of FDOs and Digital Twins is not a circle. Digital twins have attributes that FDOs do not have. For example, they might have real-time feeds of data (such as current temperature, fuel consumption, heart-rate… ), or they might have a control interface to change their state (turn on or off, raise a draw-bridge, etc). These dynamic activities hint at one big difference between the two - digital twins have behaviour. They do things in real time. There’s nothing in the FDO specification that suggests that’s important or required, but we would argue that’s from lack of imagination, rather than lack of desirability.The other lobe of the Venn diagram - the properties that FDOs have that Digital Twins do not - is mainly concerned with historical data. Digital twins are about the “now” of an entity, while an FDO is a historical collection of data, possibly about many entities over a period of time. This mismatch can be overcome by using the Digital Twin concept to define the dataset and using the “data bypass” pattern to allow the dataset to be transferred “out of band” to the requestor, given sufficient access permissions.Perhaps the strict equivalence that FDOs are Digital Twins is going too far. They certainly share many common features and posit a world where machines and humans can cooperate using the same data. Maybe the answer is to have an admixture of the two?

HTML

XML

PDF