Clever Recordkeeping Metadata Project Outcome Area 6

Metadata registry functionality and general robustness of technology infrastructure

Instantiation of the demonstrator in different technical environments revealed the lack of robustness of open source services oriented architecture components. Similarly, the research revealed different requirements for metadata registry functionality operating at different levels of granularity to support recordkeeping reuse and interoperability.

Brief discussion

The research deliberately focussed on emerging computing environments and particularly those that were open and standards based. This led to technology frustrations as the maturity of those technologies proved less than claimed. However, in the process, the conceptual understandings of the role of metadata registries within that environment was explored leading to the development of potentially productive new fields of influence to inject recordkeeping understandings and concepts into the service registries. Similarly, a much improved articulation of the multiple roles of metadata registries in implementing recordkeeping interoperability was achieved.

Summary findings

  • The research was frequently frustrated by constraints imposed by the existing status of technologies:-
    • Despite the hype, Web services technologies are still relatively immature. As a result they may lack robustness, reliability and a full complement of capabilities. The prototyping within the CRKM Project was constrained by these issues.
    • The technology available cannot appropriately undertake authentication and validation.
    • Schema construction is not robust.
  • Metadata registries need to be specified to undertake different functionality at different levels of operation. Elements and semantics from multiple systems need to be progressively registered on an ongoing basis. This provides an accumulating but not onerous mechanism to enable metadata element capture and re-use. Such metadata structuring needs to be done in schemes sourced to systems as well as to schemes relevant to industry for context issues.
  • Multiple and repeated use of metadata translation services in different ways in all registries is envisioned.
  • Registration of elements/schemas should not be hierarchical, but relational, to enable as much re-use as possible, and this may affect the metadata semantics inherent in inherited meaning where hierarchy is a crucial element in schema design.