Aspect
|
Conventional application
|
Semantic web application
|
Business link
|
Translation of functional analysis to technical analysis. Translation of technical analysis into programming language.
|
Functional analysis = data model. Technical analysis = data model. Program = data model.
The data model serves the 3 purposes.
|
Control of operations: has to be developed via KPI's (Key Performance Indicators), dashboards, software development,...
|
Operational control is demonstrated straight through the implementation of the model.
|
|
Documentation is not uniform, seldom up-to-date, not 'just-in-time' accessible.
|
Uniform documentation straight from the data model, accessible through Java-docs.
|
|
Messages
|
Manual mapping of each message field. At each message modification there is a program modification
|
XML compliant. The finance ontology is ISO 20022 compliant.
|
User interfaces
|
1 for the customer via internet
1 for the front office
1 for the back office
Each with different access rights and permissions.
|
Single interface with different access rights and permissions according to the internal policy.
|
Languages
|
Many programming languages, specialized persons and infrastructure needed.
|
No programming, modelling is performed.
|
Functions
|
The functions are nested in servers, separate software components.
|
The functions form an integral part of the model. They can be called through web services.
|
Database
|
Entities/relationships with codification of the values: translation, documentation, access to the documentation and updates necessary.
Same codes can have different meanings when systems get connected.
|
Ontology with plain text description: readable to men and machines.
|
Design and optimization with keys, indexes, inner and outer joins whereby finally the structure remains frozen because of the complexity and unknown impact of changes.
|
No database design: we promote the use of triple stores. The design for optimization is left to the database provider.
|
|
Inflexible by nature: adding a column (property) to an entity is dramatic in view of interface and server adaptations, testing, data warehousing...
|
Ontologies are by definition generic.
|
Project
|
Finance
|
Module
|
Parties involved in securities handling and financial transactions.
|
Presentation ('28)
|
|
Tutorial (+/- '30)
|
|
Documentation
|
|
Module
|
Financial instruments subject to securities handling and object of financial transactions.
|
Presentation ('15)
|
|
Documentation
|
