Use a terminology database
How do you minimize terminology problems? Start by creating a robust terminology database that records both homonyms and synonyms and can mark new terms and other usage information such as product use and subject fields. Then, make this information widely accessible to your employees, preferably through a Web site, and integrate the data into your globalization process and tools. To do this, your database will need to be well-structured and granular.
To manage homonyms, the database must be "word-oriented." Word orientation assembles all the meanings of a word into one record, the way a traditional dictionary does. Most off-the-shelf terminology management software programs are word-oriented. The following diagram illustrates a word-oriented view:
To manage synonyms, the database must also be "concept-oriented." Concept orientation connects terms that have the same meaning to one record. This is a fundamental principle of terminology management that some off-the-shelf systems do not support. The following diagram illustrates a concept-oriented view.
Finally, to record subject fields, product identifiers, usage labels (to indicate whether the term can be used or not), and other information, the system needs to support a high level of data granularity.
This leads to another important point. Glossaries only need to contain terms that are actually used in the product or application. Incorrect and unused terms are normally excluded. A terminology database is different. It serves a wide range of purposes and users and therefore should include as much information as possible to guide usage. You should not exclude incorrect or non-preferred terms from the database; rather, you should mark them as incorrect or non-preferred and provide other information such as usage notes. Otherwise, who knows whether terms not found in the database are missing because they are purposely omitted or simply haven't been included yet?