Table 6 eScience-compliant data and metadata standards
Standard | Guiding question | Requirements | Solution |
|---|---|---|---|
Terminology | |||
 Concept | What is the meaning of a concept? What do we know of the corresponding kind? | Must provide a human-readable and machine-actionable ontological definition and thus unambiguous meaning (i.e., semantic value) to a term. Must contribute to a person’s semantic inferential competence. | ontology terms, controlled vocabularies, Aristotelian definitions, essentialistic classes, cluster classes, fuzzy sets, exemplars |
 Nomenclatural | Which words or symbols are used for referring to a specific kind? | Must provide an unambiguous link between term (i.e., word/symbol) and concept, giving meaning to terms. Must contribute to a person’s output inferential competence. | unique identifiers such as URIs, preferred labels, synonyms |
 Diagnostic | How can I recognize and identify instances of a kind? What does an instance of a kind look like? | Must provide human-readable and machine-actionable empirical recognition criteria (i.e., operational definitions) that enable identifying instances of a kind. Must contribute to a person’s naming and application referential competence. | ontology terms, controlled vocabularies, Aristotelian definitions, essentialistic classes, cluster classes, fuzzy sets, exemplars |
Assertion | |||
 Format | Which syntax and file format must be used? | Must provide a format and a syntax that enables machine- actionability and that guarantees the comparability of data and metadata. The standard also requires applications that can translate data and metadata into a human-readable version. | RDF/OWL and a semantic data model, combined with tools for human-readability |
 Content | Which information is relevant? | Must provide a basic categorization and classification of contents and corresponding schemata for documenting them. | domain-specific semantic data model |