Indicators Structure¶
Welcome to the comprehensive guide on FAIRsoft indicators, designed to systematically assess the FAIRness of research software. Our indicators are split into two categories: high-level and low-level, each tailored to provide insights at different granularity levels.
High-level Indicators¶
High-level indicators represent overarching aspects of FAIRness based on the FAIR principles, offering a broad evaluation of software adherence to these principles. These indicators are categorized by the specific FAIR principle they assess:
- F (Findability)
- A (Accessibility)
- I (Interoperability)
- R (Reusability)
Each indicator is denoted by a letter representing the principle followed by a number indicating its position within that category, e.g., F3 for the third Findability indicator.
Examples of High-level Indicators¶
- F3: Discoverability - Assessing the ease with which software can be found by both humans and machines.
- I1: Data Format Standards and Practices - Evaluating adherence to industry standards in data formatting.
- R1: Usage Documentation - Examining the availability and comprehensiveness of documentation.
Low-level Indicators¶
Low-level indicators provide detailed, actionable criteria supporting the high-level indicators. These indicators delve deeper into specific aspects of FAIRness, enabling a thorough assessment process.
Each low-level indicator is denoted by extending the high-level code with a sub-number, indicating its detailed focus under the broader category, e.g., F3.1 for the first detailed aspect under the third Findability indicator.
Examples of Low-level Indicators¶
- F3.1: Discoverability in Software Registries
- F3.2: Discoverability in Software Repositories
- F3.3: Discoverability in Literature
Applicability¶
The relevance of each indicator may vary depending on the software type—whether it's "web" or "non-web": - Web software: Applications accessed through the web, such as online tools and APIs. - Non-web software: Applications run locally, like desktop apps and command-line tools.
Examples of Applicability¶
- A1.1: Existence of an API or Web Interface - Applicable to Web.
- A1.2: Existence of Downloadable and Buildable Software Version - Applicable to Non-web.
- I1.1: Use of Standard Data Formats - Applicable to All.
Scores and Scoring Method¶
Indicators are scored on a scale from 0 to 1, with 1 representing optimal FAIRness. Each indicator's weight influences the overall FAIRness score, highlighting its importance in the comprehensive assessment.
Scoring Insights¶
Scores are instrumental in generating statistics that help evaluate the collective FAIRness of software portfolios, guiding improvements and strategic decision-making. Scores should be understood as a general indication of the software's FAIRness, rather than a definitive assessment
Overview of Indicators¶
The indicators are visualized in a structured diagram that illustrates the connection between high-level indicators and the underlying FAIR principles. The diagram also highlights the weights assigned to each indicator, underscoring their significance in the overall evaluation process.
Key Points in the Overview Diagram¶
- Indicator Weights: Displayed on the connecting arrows in the diagram, influencing the total FAIRness score. The weights of the low-level indicators are not shown in the diagram.
- Implementation Status: Unimplemented indicators are shown in gray to indicate their development status.
- Software Type Applicability: Not all indicators apply universally; specific applicabilities are noted in the diagram legend.
