Release Candidate
SIP structure
The meemoo SIP consists of a hierarchical directory structure with 3 levels:
- the root directory
/or the bag level; - the
/datadirectory inside the root directory or package level; - the
/representationsdirectory inside the/datadirectory or representation level.
The bag level is a transport layer and contains essential information for checking the integrity of all files in the SIP. It uses a manifest file listing every file in the SIP together with its checksum.
The package level contains descriptive and preservation information about the SIP’s main subject, namely the different IE(s) of which digital representations are being delivered, and preservation information about the SIP as a whole (e.g. the software used to create the SIP). In addition, a mets.xml file supplies information about the SIP’s structure (in particular that of the package level) and administrative information about the SIP’s submission (e.g. the organization that submits the SIP).
The representation level contains the media files, grouped in representation folders. Each representation folder also contains its own mets.xml file, together with descriptive and preservation information about a specific representation of the IE(s) of the SIP situated at the package level and preservation information about the media files.
Running example
This section introduces a fictional scenario as a running example and is to be considered informative. The examples in the upcoming sections are based on this scenario. A complete sample (including all examples) can be downloaded here.
One of meemoo’s content partners, the (fictional) Flemish Cat Museum (henceforth FCM), wishes to archive a number of rare digitised pictures of the Felis Catus Flamens (a cat species originating in Flanders and threatened with extinction). More specifically, the FCM owns two pictures of the Felis Catus Flamens lying on a sofa and one picture of the Felis Catus Flamens on its cat tree.
The FCM views the Felis Catus Flamens itself as the IE of the SIP. They consider the two different environments in which the cat appears, as two separate representations of the main IE and decide to nest that main IE into two separate IEs (i.e. lying on a sofa and sitting on a cat tree) since they have different metadata about both environments available. This approach results in one main IE (i.e. the Felis Catus Flamens), further subdivided into two other IEs (i.e. the cat lying on a sofa and sitting on a cat tree) and two representations, each representing one of the IEs. One representation then consists of two pictures, while the other consists of one picture.
Continue to bag level.