| Scope | Explanation | Applications |
This section describes the business applications of FIBO.
Please refer to the separate Coverage section for details of the coverage of FIBO and the current status of each section.
The potential applications of FIBO fall into two broad categories:
These are outlined in more detail below.
The Financial Industry Business Ontology is a business ontology, that is a representation of the industry terms as understood by the business. This is known in some development methods as a "Conceptual Model". Other kinds of conceptual model include business process models, formal "Use Case" description of program behavior requirements, and business logic. Together, these methods provide a means for the business to formally specify aspects of what they do and what new technology applications are intended to do in the business. FIBO, as a formal business ontology, represents the things and facts in the business domain in such a way that database and message applications may be developed with reference to this.
Business applications of a formal ontology include:
In each of the above applications, the business needs to communicate a view of the requirements for data, or existing data sources need to be compared. Any situation in which someone has to ask "What does this data element mean?" points to the need for a common repository of business meaning.
In many firms this role is fulfilled by spreadsheets. These may often be very complex in their structure, and often are not subject to any formal project control. A simple application of FIBO is to simply take the spreadsheets produced from this repository, extend them as required with internal business meanings, and use these in place of informal, uncontrolled spreadsheets.
More sophisticated firms may use the underlying model repository itself. This has been built using the same kind of tooling as is normally used for software development (Computer Aided Design or CAD tools), so the repository content can be put straight to work. In these tools, linkages or "trace" relationships are used to show the relationship between some design element (data table, class, element etc.) and the business concept or concepts to which it relates.
The Semantic Web is a new approach to technology, in which formal ontologies may be put to work to process data in new ways. Information is stored in a new kind of data structure called a "Triple Store" in which every assertion about the world is framed in terms of some "Subject-verb-Object" relationship called a "Triple". This method of storing and accessing data is flexible, extensible and may be applied across multiple sets of data in a common data language.
The opportunities for semantic technology applications arise as a by-product of the way in which the common business semantics in FIBO have been framed. From the available methods for capturing formal business semantics to satisfy the above applications, we chose the "Semantic Web" approach. This means that each of the terms in the repository is framed in a formal manner, using underlying syntax defined in the "Semantic Web", that is the Web Ontology Language (OWL).
As part of the formal standardization process being undertaken with the OMG, the FIBO content is made available as physical OWL files. These are not available for the parts of the repository which have not been submitted into the OMG process.
Important Note: In order to put these OWL files to work further technical changes are required. These changes will vary according to the application requirements. Please refer to the Technical section for more detail on this.
Semantic Web applications include:
In order to make use of these new opportunities, the data to be processed needs to be in the new "Triple Store" format. However, some applications such as semantic search may also be applied across conventional data stores using commercially available "adaptors" designed for this purpose, provided that a suitable formal ontology is available to define what the terms mean.
The advantages of this style of data storage, processing and retrieval include flexibility and extensibility. This however requires that the ontology be developed along truly semantic lines, so that for example each kind of item in the business domain is formally framed in terms of "what kind of thing it is" and "what facts distinguish it from other things" (please refer to the technical section for more information on this). Provided that the semantics is applied adequately, any new type of term may be added at any stage in the future, without having to "break" the data model. This is in contrast to conventional data models, which are typically brittle and which therefore require careful change management.
Users should be aware however that this benefit only follows if the ontology has been structured semantically - this is not a simple matter of framing the data elements in the Semantic Web syntax. Needless to say, syntax is not semantics.