Recommendation:
We recommend that dataset release (version) identifiers can at least represent the following information:
- Changes in the file content that only improve its compliance with the already published documentation ("patch" release).
- Changes in the file content that reflect a different scientific methodology ("major" release).
- Any changes in the file content not falling in the above categories. For example, changes that improve interoperability of the content with certain convention or software tool ("minor" release).
Furthermore, for any dataset identifier scheme the ordering of its fields and their formatting should allow for lexical ASCII sort order of granule file names from the oldest to the most recent dataset release.
Recommendation Details: The dataset release (version) identifier is usually the final field that can assure uniqueness of a granule file name if all other information for that granule remains the same. This recommendation neither prescribes nor endorses any existing scheme for dataset release identifiers. Further, it does not prohibit providing any additional (more specific) release information. What constitutes a major, minor, or patch dataset release is left up to the data producer to decide based on the recommendation’s broad descriptions.
