Abstract Modelling and Model Driven Architecture

Abstract modelling of concepts in any system of more than trivial complexity is really essential in order to understand the system behavior and anatomy.  This modelling generally falls under the topic of enterprise architecture but really relates to understanding any system design without having to understand the detail of the implementation. These days we often want to describe systems with diagrams. In the past these have suffered from a lack of formality with the risk that everyone interprets them in their own way.

UML offers a significant step forward in describing a wide range of elements of a system in a way which has a common language. and it can be used, through the various notations it includes, the use cases of a system, the data structure, both static and dynamic and the processes involved in enacting the system.  It can also be used to describe both abstract and implementation specific models, allowing us to describe a system at all levels.

In particular in recent times in the geospatial world we have used UML as the ‘gold copy’ description of geospatial information models which describe how we capture a model of real world geographic data. Add to this model which is abstract, a set of tools to realise that model in a database, a file encoding or a web service and you can generate consistent information structures across multiple environments, multiple user domains and vendor or OSS technologies which can inter-operate freely.  The automated process of model generation is known as a model driven architecture (MDA) and is critical to many large spatial data infrastructure implementations.

The question most people ask is ‘why should i worry about this?’ or ‘does it really affect my organisation?’ If your organisation is going to operate in a completely closed environment, probably not. But if the information you need comes from external sources, or if you produce data needed by others, then it will become more and more important.

20-20 Geospatial specialises in providing advice and consultancy in this area, and can help insure best practice or leasing edge activity is taken into account in your design, derisking your implementation by learning the lessons from other implementors.You can contact us today and we can provide impartial, system independent advice to help guide your exploitation of the next generation of spatial data infrastructure.