METS provides a way of encoding the descriptive, administrative, and structural metadata associated with objects in a digital library. It is a hot topic in the metadata and digital library world right now. David Ruddy will lead a discussion of METS: a general introduction, its strengths and limitations, and possible applications within CUL.
Here is a definition of METS from the official website: "The METS schema is a standard for encoding descriptive, administrative, and structural metadata regarding objects within a digital library, expressed using the XML schema language of the World Wide Web Consortium."
David Ruddy explained that METS grew out of the MOA II Project. Originally designed for metadata associated with digitized archival objects, it has now been generalized and can be used for any digital object. METS specifies a standard format in XML for encoding the metadata needed to manage and exchange digital objects. It provides a framework, or container, for metadata. METS objects can satisfy the needs of the OAIS model (as information packages) and could be used in digital projects that are underway at CUL.
A METS object, an XML file that encodes the metadata associated with a digital object, has six possible sub-elements:
The descriptive metadata section has two possible sub-elements. One (mdRef) references external data (for example, a MARC record, or another METS object). The other (mdWrap) provides the framework into which you can directly plug data encoded according to various metadata element sets (extension schemas).
The administrative element has four sub-elements that group types of administrative metadata together. Within each section, the same structure as in the descriptive metadata element applies: you can reference external metadata or wrap internally specified data. The four administrative metadata sections are: technical metadata, intellectual property rights, original source information (for example, the analog book on which the metadata is based), and data regarding digital provenance (how the files were created and how they evolved).
The file section is an inventory which groups together all files that constitute one version of the digital object. For example, all TIFF images that represent one book can be grouped together here. If there are several versions of the digital object, there can be several file groups. For example, besides all master TIFF images, each page may have a thumbnail version, or an OCR text file, associated with it. These would constitute a separate file group. This is primarily an inventory but also provides some information about how files relate to each other.
The structure map is the heart of the METS object. Its many pointers may be external (to another METS object) or internal (to the file section). It is possible to have multiple structure maps for a given object representing different "views" of it (e.g., a flat, physical view of just the pages, or a more hierarchical view of its logical structure: preface, chapters, index, etc.). The structure map determines how an object can to be viewed and navigated. The structure map is the only required element in the METS schema.
METS provides a simple framework, but, because of the amount of metadata it can contain and the interrelationships among all that metadata, in practice METS objects can get very complex. An individual record can be dizzying. There need to be tools for constructing records; hand-coding is prohibitively tedious and time-consuming. METS also depends heavily on extension schema, so there is a need for community developed digital standards. And because METS can be implemented in such a variety of ways, application profiles for its use are needed before exchange and sharing of METS objects is feasible.
METS has advantages: it provides a common metadata format, it is flexible, modular, extensible, expressive, and is an open (non-proprietary) standard. Currently it is under review by the Digital Library Federation. After their comments, it will become METS Version One and RLG will take administrative control of it.