The Next Big Thing
Posts about next generation technologies and their effect on business.

The Need for Multilingual Modeling Support

The time for multi-lingual modeling capabilities is past due. Many enterprises are global, with operations in different countries speaking different languages. Some countries have significantly different dialects or multiple languages. Within business, different functional areas have different vernaculars. Engineering and financial people may have different terms for the same concept; why should the information technology organization (IT) require that they use the same terms? IT activities need to be able to capture and express models in different languages so that business people as well as IT people can contribute to the models and understand them.

When models were only used and viewed by IT people, it was reasonable for them to agree on terms in a single language. Programming languages and database schema are designed to expect consistent terminology to identify entities and attributes. In addition, companies were seldom multi-national and solutions were typically focused on a single department. However, enterprise integration, globalization, MDA (Model Driven Architecture) and business models have changed all of that. Models of business processes, business rules and other business aspects are intended to be used by business people from across the enterprise. Acceptance of these models and realization of the benefits requires that they be multi-lingual.

It should be possible to name each concept in a model with terms from multiple languages. The user of the modeling tool should be able to select his or her preferred language for interaction with the model concepts.

As a starting point, multi-lingual capability is important for development of enterprise conceptual data models. Endless hours are spent on developing consensus on names.  This often gets confused with developing agreement on the concepts, and discussion shifts back and forth between debates about the concept and debates about the name. With a multi-lingual data modeling tool, each functional area, and each language community involved could use names that are most appropriate for them. The concepts that appear in data models will also appear in other models. This is important, not only for international business enterprises, but for international standards as well. Standards will be more readily adopted if the specifications can be expressed in different languages.

In the final analysis, it may still be necessary to use a single set of terms to describe the concepts for XML tags and for application and database implementations within an enterprise. However, these are IT artifacts, and IT is just another functional area that needs a consistent language.

This capability is not unprecedented. The OMG specification for SBVR (Semantics of Business Vocabulary and Rules) separates concepts from terminology so that users can interact using different vocabularies, and business rules can be expressed in different languages. MEGA, a French company, has built multi-lingual capabilities into their MEGA Modeling Suite. MEGA provides business consulting and business modeling tools to support clients in over 40 countries. I'd be interested in knowing if there are other modeling tools with such capabilities.

Search
Follow Us
About the Author(s)
  • Steve Simske is an HP Fellow and Director in the Printing and Content Delivery Lab in Hewlett-Packard Labs, and is the Director and Chief Technologist for the HP Labs Security Printing and Imaging program.
Labels