Warning: These documents are under active
development and subject to change (version 2.1.0-beta).
The latest release documents are at:
https://purl.dataone.org/architecture
Summary
The standards to which science metadata are developed evolve over time, and it should be possible to perform lossless update of science metadata to ensure that content is kept up-to-date with current use standards.
Since Coordinating Nodes hold original copies of all science metadata, it should be reasonably straight forward for a privileged service on a CN to create new versions of science metadata given appropriate tools for performing the transforms. The original copies of the metadata should be preserved and the associated system metadata references updated to indicate the newer version of the science metadata.
Note that while technically feasible, such a service may not be desirable as it implies a level of control exerted by the Coordinating Nodes that may be in conflict with data sharing and participation agreements.
Notes