View Source

Main principles for the versioning of data models in the Data Vocabulary Tool are as follows:

When you create a new data model, it obtains Draft status.
- Only the members of the data modellers' organisation can see a data model with Draft status, and only if they are logged in to the tool. Other users cannot see the data model.
The draft of the data model can be edited freely.
The draft of the data model can be published as public versions. They will then get a version number.
The published draft versions will have either Valid or Suggested status.
The published draft versions are frozen (snapshot) and the actual contents cannot be edited anymore. This is for maintaining the data integrity of the published data models.
- This way the other users of the data model can make references without worrying about any changes to the data model, as its classes, attributes and associations will not change after their publication; hence, after the time the reference to them was made.
- When the data model is published as a new version, the users of the data model can choose to maintain the references made to the older version or change the references of their own model to target the new version.
Only the metadata of a published data model can be changed; for example, the content contributor can be changed to another organisation.
The published data models cannot be totally erased from the tool when, for example, the content has expired. The data models not in use will obtain another status: either Retired or Superseded. The non-used data models will not be shown in the data model list on the home page. When needed, they can be found with search functions.

Main principles for the semantic version numbering are as follows:

The version numbering scheme is not enforced, i.e., you are able to give any numbers you wish to new versions, as long as they follow the semantic versioning format: X.Y.Z. In the format, X denotes a major release number, Y a minor one and Z is a patch.
Major versions are expected to be breaking, i.e., not backwards compatible. Minor versions are meant to be backwards compatible, and patches are meant to be small fixes with no changes to the actual substance. A more detailed explanation of the principles: https://semver.org/.

The Statuses used in The Data Vocabulary Tool

The contents of the data model can be transferred to another status with the following rules:

Current Status	The new status
Draft	Suggestion, Valid
Suggestion	Valid, Retired Before the status of the data model is changed to Valid, the modeller is asked if the version number will remain the same. If the version number is to remain the same, the tool only changes the status of the data model. If the modeller wants to give the data model a new version number, the modeller will be asked to provide a new version number. The tool then makes a copy of the data model with a new version number. The original data model in Suggestion status will have the new status Retired, so the possible references made to it will remain unchanged.
Valid	Retired, Superseded
Retired	Superseded, Valid
Superseded	Retired, Valid

More detailed information

Name	Description	Code value
Draft	The creation of the content has not yet been finished, and the current data model is not ready to be published. The contents can be seen only by the other users of the same organisation than the editor when they have the editing rights to the content. The draft data model does not have a version number. It is a working copy, from which the other versions can be published in the tool as publicly shown snapshots. These published snapshots cannot be edited. However, it is possible to change the status of the published data models.	DRAFT
Suggested	The data content has been published for comments and it has a version number. The data model with Suggested status also has a public IRI address, i.e. an IRI identifier. However, it is not recommended to add references or links to the data model in Suggested status, because it is likely that the data model will be changed later, and it may never be published as valid. The content creator must decide when the commenting period will end and after that assign the data model either Valid status (changes were not needed) or Retired status (the recommended content was not accepted). The IRI identifier of the data model is reserved for this data model and cannot be further given to any other data model.	SUGGESTED
Valid	The data model has been published and has a version number. The content is generally used as, for example, a reference by the other data models or linked to the contents in the interoperability platform tools. The IRI identifier of the data model is reserved for this data model and cannot be further given to any other data model. When the content has been published and has the Valid status, it cannot totally be removed from the Data Vocabulary Tool. This will ensure that the links and references will remain valid (the content will remain resolvable). The content can only be changed to Retired status to remove its contents from active use.	VALID
Superseded	This data content is no longer recommended for use, as it has been superseded with other data content. The IRI identifier of the data model is reserved for this data model and cannot be further given to any other data model.	SUPERSEDED
Retired	The data content is no longer recommended for use. The data content has not been replaced with other data content. The IRI identifier of the data model is reserved for this data model and cannot be further given to any other data model.	RETIRED