Workbook on Digital Private Papers > Administrative and preservation metadata > Using METS for the preservation and dissemination of digital archives

Using METS for the preservation and dissemination of digital archives

Strengths of METS

  • METS is a standard maintained by the Network Development and MARC Standards Office of the Library of Congress, and is non-proprietary. Any system capable of handling XML documents can be used to create, store and deliver a METS file, thereby mitigating problems of software obsolescence.
  • It is written in XML, which is robust as an archival medium and is readily interchangeable because it uses standard ASCII code rather than a binary format to encode its data.
  • As an XML Schema it has certain advantages over a DTD. Schemas are generally richer and more powerful than DTDs: they provide greater control of the content and usage of elements and attributes; they are extensible to future additions, meaning that METS can accommodate new standards as they are developed in the future; and they support the use of multiple XML namespaces, which allows different kinds of metadata to be encoded in the same document.
  • METS has the ability to deal with a wide variety of materials; this is particularly useful in the context of personal archives which typically contain a wide range of object types - including text documents, images, email directories, websites and blogs.
  • METS was specifically designed to act as an OAIS Information Package: it can deal with all categories of OAIS metadata; it packages this metadata with the digital object it describes, ensuring that the object is self-documenting over time; and it intellectually links together all the categories of metadata for an object, even if they are stored in separate locations.
  • It was developed by the library community and has been widely adopted in preservation repositories, although its use in the context of personal digital archives has been little explored to date.
  • It effectively expresses the hierarchical structure of digital objects. This is very useful in the case of personal archives, as it enables the creator’s original structure of directories and folders to be maintained.
  • Whilst METS has the capability to deal with large and complex digital objects which might be comprised of many files, it is also useful for dealing with individual files (e.g. a single word-processed document). The latter is useful for digital archives, where long-term digital preservation requires that extensive metadata be recorded for each individual digital object.
  • The possibility of creating multiple structural maps in a METS document means that whilst it is invaluable for preserving original hierarchical relationships between objects, archivists can also take advantage of its capacity for sorting and reordering records in varied ways for researcher access.
  • It is relatively easy for archivists familiar with XML to pick up: it copes well with the kind of hierarchical structures archivists are used to dealing with, and the mark-up will be relatively familiar to archivists who are used to encoding archival finding aids in EAD.
  • It is extensible: new versions of metadata may be incorporated alongside older versions of metadata, thus providing an audit trail.

Weaknesses of METS

  • The very flexibility of METS can raise interoperability problems. It does not ensure standardisation because it does not operate as a metadata standard, rather as a framework within which metadata can be stored.
  • Whilst using METS Profiles can mitigate these problems to an extent and facilitate manual cross-mapping, this still does not allow the automatic transfer of files between systems.
  • At present, and certainly in the case of personal digital archives (which are not only extremely varied but also require large amounts of preservation metadata), METS documents largely have to be generated manually; this is unfeasible in the long-term, and solutions will have to be found.
  • METS relies on the effective use of unique identifiers. This can be difficult to administer.