Strategy and principles for versioning

As the guiding principle for versioning of data models and interfaces semantic versioning is used. Semantic versioning (https://semver.org ) defines 3 levels of versions:

Major.Minor.Patch

• MAJOR version, use for breaking changes

• MINOR version when you add functionality in a backward compatible manner

• PATCH version when you make backward compatible bug fixes and patches

When a datamodel or interface has a version 1.0.0 og higher, the principle of versioning is used strictly.

Version 1.0.0 for data models is normally the version where comments from the public review are incorporated in the model and documentation and is the first version of a data model that is deployed to production at the Orchestration component. Version 1.0.0 of the related interface is also the first version that are deployed to production.

Derived or related artefact from a data model, e.g. schemas or mapping documents, follows the same version as the data model. Meaning, that if a data model has version 2.0.0, the derived or related artefact has version 2.0.0 also.

Working version og data models and interfaces

For version of data models and interfaces, before they have reached version 1.0.0, the versioning principles is only followed partly and with some amendments (see below). Draft versions of data models and interfaces are very often incompatible due the many changes and add-ons. For draft versions of data models the following applies: 

• For changes to a draft version, the version numre is normally only changed at the patch level.

• Minor level is used to capture more stable draft versions of data models and interface, used for creating UX mockup or other design activities and review activitties.

• Version 0.8.0 is the version that are used for end-user test of the UX-mockup

• Version 0.9.0 is the version that are sued in public review

• Interfaces in test environments follows the version as the data model