Login required to access the wiki. Please register to create your login credentials We apologize for any inconvenience this may cause, but please note that this step is necessary to protect your privacy and ensure a safer browsing experience. Thank you for your cooperation. Documents available for download: GAMSO , GSBPM , GSIM |
(Feedback from Australia)
Beyond review of the specific changes in GSIM 1.5, a number of broader observations were offered by implementers within the ABS
- Implementing organisations should establish a firm view of the role they expect Concepts to play and the granularity they want Concepts created at. Being too fine grained in definition of Concepts may add proliferation while adding little value, while being too coarse grained leads to Concepts that are more like Topics/Subjects rather than a Unit of thought differentiated by characteristics.
The relationship between Nodes and Concepts and between Variables and Concepts could be optional to ensure Concepts are only defined where they will value beyond the definition of the Node or Variable itself. (In practice some Nodes in "non standard" code lists consist of one or two indicative words - eg from a prompt card - rather than being underpinned by a fully and formally defined concept. Similarly defining a Concept for an intermediate derived Variable may add little value.)
- In retrospect, ABS implementation based on GSIM 1.1 would have benefitted from establishing in advance
- a clearer, more detailed, sense of how data and metadata was expected to flow through the business processes, and
- how authoring of metadata would be made efficient and sustainable for statistical staff.
This would then be coupled with small scale, well selected but realistic proofs of concept from a business perspective to test and iterate various aspects of the implementation approach (eg in regard to conceptual metadata, in regard to structures, in regard to exchange channels).
A "top down" approach from GSIM to implementation, and/or an approach focused on shaping system architectures in advance of engaging with business, is unlikely to be sufficient.
- Generalising the above points, while reviewing the updates to the modelling between GSIM 1.1 and GSIM 1.5 is useful, sharing experiences (including rationales and lessons learned) from implementations across agencies is likely to make the model more approachable and understandable in practice.