Mauro Data Mapper / Oxford Metadata Catalogue

The Mauro Data Mapper (a.k.a. the Oxford Metadata Catalogue) is used to develop and maintain linked, versioned descriptions of data standards, datasets, and questionnaires. These descriptions capture essential structure and context together with a detailed account of each variable, comprising: name, natural language definition, data type, and multiplicity. A data type may be a primitive type with units (such a height in metres) or an enumeration, complete with an explanation of the intended meaning of each value (such as 1 = male, 2 = female).

The explicit treatment of structure means that common aspects of variable definitions can be factorized for scalability: that is, introduced at the level of an enclosing data class, table, or form section. Descriptions of variables, and descriptions of these component structures, can be linked to indicate semantic relationships between them: most often, the existence of mappings and/or conversions that allow one variable, or one component to be used in place of another (semantic interoperability).

Descriptions and links are managed as components of data models. Each model in the catalogue will correspond to a particular artefact: for example, a data standard, a data extract, or a form design. There may be more than one model of the same artefact, reflecting different perspectives and/or purposes: for example, a link indicating semantic interoperability might be valid for some purposes but not for others. The catalogue supports contextualised (even apparently inconsistent) and evolving viewpoints.

This support for multiple perspectives adds to the scalability of the approach: descriptions and definitions can be contributed by many parties, without the need for prior reconciliation. Users of the catalogue can choose which models, and which descriptions and links, to build upon. In most cases, they will make this choice based upon model authorship or provenance, and choices can be automated. Other tools may be used to automate the production of models from artefacts and - conversely - the generation of artefacts (for example, forms, or queries for data extracts) from models already in the catalogue.

For more details, please contact us via email: