Introduction
Record
The central storage object in the MBDB is the record. The record not only stores the metadata and files, it also contain links, owner information, requests, versioning information and of particular importance is the state.
The state determines both who can interact with the record and how they can do so. Below is an overview is given of the state from the perspective of the record owner (you, the depositor).
Record states
| State | Publicly visible | Editable by owner | Has DOI |
|---|---|---|---|
| draft | No | Yes | No |
| submitted | No | No | No |
| accepted | No | No | No |
| published | Yes | Yes(1) | Yes |
| retracting | No(2) | No | Yes |
| deleted | No(2) | No | Yes |
- Editing a published record will normally result in a new version that has to go through the approval process again.
- Most information will not be visible, however the record page will continue to exist and have minimal metadata see Deleted records.
Publication workflow
The publication of your dataset follows a similar workflow as when publishing scientific literature:
For a worked example of how this looks like in practice, have a look at the Tutorial (opens in a new tab).
Retraction workflow
Records that as not been published can always be fully deleted (see Deleting unpublished records for further details).
It's possible to retract published records from the MBDB and it can be be initiated by the owner at any point. However it comes with a few important caveats:
- The raw data and metadata has been placed in the public domain (see Licensing) so people are free to use and distribute existing copies.
- The record landing page will still exist and have minimal data associated with it. This is a requirement in order to have persistent identifiers.
See Retracting published records for further details.