Mission Critical’s approach to IT
Mission Critical’s approach to IT
At Mission Critical IT, we believe in separating, as much as possible, the business logic of an application from the technology and “plumbing” used to build the application. This approach has key benefits for business as well as IT.
IT does not have to grapple with the inherent business complexity as well as the inevitable technical complexity. Instead IT focuses on consuming, in a general way, the model developed and maintained by the business experts to produce a working application, while the business experts focus on modelling the concepts, rules and processes that make up the business domain.
Mission Critical IT uses W3C endorsed semantic web standards for modelling the business domain. Once the business model is in this format it can not only be reused in other in-house projects, but also by the increasing number of external tools and services that are becoming semantic web aware.
Automatic code generation
Mission Critical IT makes it easier to separate the business knowledge and technical IT knowledge by providing tools that can generate large portions of the application code automatically from the business model. This makes integrating complex business functionality into working applications easier.
The automatically generated code is fully compliant with the Java and .NET platforms. On the Java platform it is documented using Javadoc and follows the JavaBean convention, allowing the code to be easily integrated into larger enterprise developments. On the .NET platform the generated code follows .NET best practice.
Performance and scalability
The business model is interpreted by high performance reasoning and rule execution engines. Mission Critical IT provides a range of different reasoning and rule execution engines which can be tuned to trade off expressivity for performance.
The business model and facts about the model are stored in a RDF store. Mission Critical IT provides a range of RDF stores: in-memory, berkeley-db, SQL and sesame accessible stores. This allows the appropriate store for the application to be chosen.
All of the Mission Critical IT components are fully stateless so can be used in highly scalable architectures.
Living breathing specifications
When the business model is in a form that is understandable by business experts and can be directly consumed by computer programs then the business requirements cease to be a heap of documents subject to misinterpretation and a burden to maintain, but instead become a living, breathing specification that directly drives the application. Mission Critical IT can help IT to deliver the applications that the business really wants by directly involving business experts in the modelling process and allowing IT to consume that model directly. This avoids IT frustration and potentially expensive reworking because of unclear or incomplete specs. It also makes development truly agile, because it is driven by the business requirements in a very direct and manageable way.
The business model is vendor neutral. This means that the business data can be housed on a wide range of database backends and moved between different platforms with minimal effort. For example data can be stored in a traditional relational database (such as Oracle, SQL Server and MySQL), or stored in one of the many distributed high-performance stores designed specifically for semantic web data. Mission Critical IT has also successfully deployed applications in the cloud on Amazon EC2 and Google App Engine. Because the data is described by a vendor neutral, standards compliant model, it can be moved between any of these platforms trivially. The same model can also be used to expose data stored in legacy systems using open, accessible standards.
Mission Critical IT are experts at data integration and aggregation. We have integrated numerous exotic and disparate systems. Our approach is always the same. First we model the domain using semantic web standards. Once we have a model we have a clear frame of reference by which systems can be integrated. Because the model can be consumed directly by computer programs, once the model is defined a lot of the integration work can be automated.
ODASE™ from an IT perspective
Well established software development combines modeling using UML, proprietary Business Rules Engines (BRE), some form of Business Process Management (BPM), each with their own language. As they do not share a common model, a lot of Java or .NET code is needed to “glue” it together.
Combining model, rules and processes
The ODASE™ ontology-driven approach unifies the model, the rules and the processes into one common knowledge asset (an ontology) exploiting the power of reasoning and inferencing. Thanks to the logic foundations, very advanced tools such as the ODASE™ Workbench can be used. Based on W3C Semantic Web standards, the model and the code are Web friendly and Web 3.0 ready.