Essential Components for
the Business/IT Lexicographer
Today’s organizations often struggle with how poorly their information systems perform. Sources of this poor performance can be many, including functions and components in the conceptual, structural, organizational, and operational domains, among others.
One recurring source of this poor performance is linguistic in origin. The root of this lies in the multiple vocabularies that are frequently present and that lack correlation between them. In short, for effective communications and shared understanding to be achieved between participants, the meaning of a message must be preserved between the source and the recipient. This is known from basic information theory and practical experience. Without this preservation of meaning, information systems can be, and often are, a metaphoric ‘Tower of Babel’ of the Information Age.
One of the most notable challenges of preserving meaning can occur between business vocabularies and technical data vocabularies (aka ‘data dictionaries’). For example in Figure 1, the concept ‘Order’ in a business context is likely to have a different description (as characterized by its metadata), than the equivalent technical description. In healthy information systems, a correlation or linkage between the two is required to assure the business meaning is accurately translated into technical properties. A lack of correlation has several potential negative effects on an organization, such as:
- Misunderstanding and ambiguities leading to mistrust and uncoordinated decisions and actions.
- Loss of accuracy in the representation of business requirements leading to loss of trust in the veracity of the organization’s data.
- Degradation of decision quality resulting from data inaccuracies. Increased rework and remediation as a result.
- Inability to maintain control over the organization’s information. Today’s data universe is big, diverse, and complex. Without correlation and traceability/lineage, change management is impossible.
A solution to taming this chaotic situation is to integrate three core functions and supporting technologies. These functions are Business Term Management (aka Business Glossary), Technical Metadata Management (aka Data Dictionary), and Mapping (aka Data Catalog).
Business Glossary Role
A business glossary is a compendium of business terms that serves as a uniform basis of reference of the business’ knowledge, Figure 2. The business glossary is controlled and managed by the organization’s business function.
Organizations that have multiple business glossaries can have conflicts and ambiguities in the interpretation of business requirements. The concept of providing a common understanding of business terminology is a guiding architectural principle for the business glossary. Where multiple glossaries exist, establishing and preserving reference-ability is key to developing a common usage across the organization’s context, both internally and with customers and suppliers.
Data Dictionary Role
Generically, a data dictionary contains the descriptions and details involved in the storage of data. The data dictionary includes details about the properties of a database (metadata) such as definition, data type, length, lineage, transformations, and other pertinent details, Figure 3. The Data Dictionary is managed by the technology organization.
Especially in the development and management of databases, data dictionaries help provide consistency, clarity, reusability, and completeness. They also aid the sharing of information about the data architecture and integration projects.
Data Catalog Role
A data catalog provides the required mapping between a business glossary and a data dictionary, Figure 4. The data catalog provides a single reference source for the location of any data set required for varying needs such as Operational, BI, Analytics, Data Science, etc. Assembling the data catalog is a collaborative effort requiring involvement from both business and technical sources of knowledge.
This mapping is the primary and most efficient way currently available for preserving meaning between business and IT vocabularies.
How the Three Functions work Together
Ideally, a business glossary and a data dictionary are importable into a catalog where mapping, searching, referencing (lineage), publishing, and management functions can be performed. If multiple versions of business glossaries and/or data dictionaries are to be mapped, an optimal approach is to harmonize, i.e resolve gaps and conflicts, in these prior to their integration in a catalog, as in Figure 5.
What is Gained by this Functional Collaboration?
This process enables organizations to optimize communication and understanding of business and technical terminology across the organization’s domains, both internally and externally. Relevant and timely information becomes more readily available to decision makers when this is done. Additionally, the organization’s responses to change can cost less and be more accurate and timely through improved control and management.
Additional benefits of using a structured approach as described here is gaining tremendous savings in time and cost. Organizations that make ad hoc attempts to align today’s large and complex business and technical vocabularies face difficult technical, organizational and management obstacles.