Magma Core is a lightweight set of Java classes to enable HQDM data objects to be created and used as RDF Linked Data through Apache Jena.
Representing things that are of common interest in data presents a consistency challenge. How can we use the apparent logic of computer based systems to store, process and query without suffering the losses due to the use of different models, structures and requirements on which interconnected systems are based? There is a lot to be gained from achieving consistency of representation, from addressing the core challenges to do with data quality and information management through to enabling emerging topics like Digital Twins & Artificial Intelligence. Inconsistent representations in data also undermines the ability to achieve improved information security, policy controls and ensuring that information is fit for its intended purpose.
This code release is a contribution to support the adoption of data models that have properties that enable consistency of representation, particularly in the face of emerging and changing requirements. The theory on which the model that this repo adopts is based upon the entity-relationship model published by Dr. Matthew West* as the High Quality Data Model Framework. At its core are some ontological commitments drawn upon the identity of actual things being based on distinct existence in space-time. A collection of arguments, including Set Theoretic commitments, provide a powerful model framework that can be extended to (almost) any application area of interest. This approach is often summarised as 4-dimensionalism and is being incorporated into the UK's Information Management Framework (part of the National Digital Twin), as the Foundation Data Model - a backbone of distributed information systems that are sufficiently integrated to address system-wide data quality. One of the challenges in adopting such models is simultaneously addressing the analytic & technological challenges. This release is provided to help those interested in learning about and adopting such models by lowering the technical barriers.
* Dr. West was not involved in the creation of this project. GCHQ are grateful for his open contributions to the field of ontology and data quality management.
HQDM contains the replication of an openly available data model based on key ontological foundations to enable the consistent integration of data. The HQDM Java package comprises a set of Java classes and respective interfaces, 230 in total, to replicate the entity-relationship model published by Dr Matthew West as the High Quality Data Model Framework. This class model can be used to create extensions of the entity types, based on the founding ontological commitments and logical restrictions (such as cardinalities), and instances of those types all in Java application code. This, in theory at least, provides a framework for the consistent representation of almost anything that is, or could be, real*. All the data model patterns published in the HQDM framework are supported by the HQDM package. The object properties are constructed around a top-level Java HQDM Object class with some root attributes to enable class-instances to be managed in a database. The choice of database can be left to the user but the structure of the attributes is optimised for the use of Linked Data IRIs and RDF triples to represent HQDM object relationships and other object properties as predicates. All of the HQDM objects can be created and searched using the HQDMObject methods and collections can be handled using the Object Factory. To complement this there is an OWL version of the HQDM data model that is a close match for the original EXPRESS model and the HQDM Java package.
* This is a gross simplification, but it characterises the goal of the model and in use it has proved to be very capable. The UK's National Digital Twin programme is developing a model that aims to address this goal with even more rigour, called the Foundation Data Model (FDM). Data created using HQDM is likely to be mappable to the FDM with low mapping (due to similar ontological commitments).
An introduction to Magma Core is provided in the Magma Core Wiki.
Magma Core can be incorporated into other maven projects using the following dependency:
<dependency>
<groupId>uk.gov.gchq.magma-core</groupId>
<artifactId>core</artifactId>
<version>4.0.0</version>
</dependency>Magma Core is not currently hosted on Maven Central, so a local install of this repository will be required.
We welcome contributions to the project. Detailed information on our ways of working can be found here.
In brief:
Magma Core is released under the Apache 2.0 License and is covered by Crown Copyright.