The Practical Policy working group of the Research Data Alliance is developing examples of policies that can be used to control data management applications. The policies are expressed as computer actionable rules. The rules can be used to enforce management policies, automate administrative functions, validate assessment criteria, and automate data analysis pipelines.

Collection of policies

Data Management

Data management policies are intended to automate administrative tasks. Examples include managing access controls, managing storage caches, managing ingestion of files, and managing migration of files.

[Freelinking: unknown plugin indicator "path"]

Preservation

Preservation policies are used to enforce assertions about the records in an archive, including authenticity, chain of custody, integrity, and original arrangement.

[Freelinking: unknown plugin indicator "path"]

Replication

Replication policies are used to simplify disaster backup by managing distributed copies of each file. The replication policies may interact with policies for managing backups and policies for versioning of files.

[Freelinking: unknown plugin indicator "path"]

Sharing

Sharing policies are used to manage distribution of data between collaborators and time-dependent access controls.

Federation

Federation policies are used to manage interactions between independent data management systems, including automated synchronization of holdings, replication of data, and sharing of data.

Description

Description policies are used to automate extraction of descriptive metadata from files.

[Freelinking: unknown plugin indicator "path"]

Trustworthiness

Trustworthiness assessment criteria have been defined in the ISO 16363 standard, and include criteria for monitoring repository management functions.

[Freelinking: unknown plugin indicator "path"]

Caching

Caching policies are used for managing interactions with object stores and tape archives. A standard example is the use of a cache to aggregate files into a container before writing to tape.

Processing

Processing workflows are used to capture the knowledge needed to execute a research analysis. The workflow could include access to remote data repositories, transformation of the retrieved data, and execution of an analysis code.