Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
ISO 26162-3:2023 — Terminology Databases — Part 3: Content
Management of Terminology Resources — Terminology Databases — Content
1. Terminology Content Structure and Quality
ISO 26162-3:2023 completes the ISO 26162 series by addressing the content dimension of terminology databases. While Part 1 covers database design and Part 2 covers software functionality, Part 3 focuses on what goes into the database the terminology content itself including definitions, usage contexts, grammatical information, and comprehensive quality criteria. This standard is essential for terminologists, technical writers, and domain experts who create and maintain terminological entries, providing detailed guidance on content creation practices that ensure consistency, accuracy, and usability across all supported languages and applications.
Part 3 introduces a maturity model for terminology content quality, ranging from basic term lists at the lowest level to fully validated, richly annotated terminology assets suitable for mission-critical applications such as medical device documentation, aviation safety manuals, and pharmaceutical regulatory submissions.
The standard establishes clear requirements for definition writing definitions must be concept-oriented and define the concept not the term, use controlled vocabulary within the definition text to avoid circularity, avoid circular definitions that reference each other, and be fully self-contained without requiring external references for comprehension. Each definition shall include only the essential characteristics that distinguish the concept from all related concepts within its subject field, following the classical principles of Aristotelian definition using genus proximum (nearest generic concept) and differentiae specificae (specific distinguishing characteristics).
| Content Element | Requirement Level | Quality Criteria |
|---|---|---|
| Definition | Mandatory | Concept-oriented, precise, non-circular, genus-differentia structure with clear wording |
| Context / Usage example | Mandatory | Authentic source citation, representative of actual usage in technical documentation |
| Grammatical information | Mandatory | Part of speech, gender where applicable, number, inflectional forms for morphologically rich languages |
| Subject field classification | Mandatory | Mapped to standard classification system with consistent granularity across all entries |
| Concept relations | Recommended | Generic-specific, part-whole, associative with explicit relation type identification |
| Usage notes | Recommended | Regional preference indicators, register information, stylistic and domain usage restrictions |
2. Multilingual Terminology Management
A significant portion of ISO 26162-3 addresses the complex challenges of managing terminology across multiple languages. Each concept entry must clearly identify the language of each term using standard language codes and support language-specific grammatical information, usage contexts, and regional variations. The concept identifier serves as the cross-lingual linking mechanism, ensuring that translators and technical writers can find equivalent terms across all supported languages with confidence. The standard requires that each language-specific term entry include grammatical information relevant to that language such as gender for French and German, aspect for Slavic languages, and measure words for East Asian languages.
False friends terms that appear similar across languages but have different meanings must be explicitly flagged in the terminology database with cross-linguistic notes warning users about potential mistranslations. This is particularly important for technical domains where incorrect term selection could lead to safety hazards or regulatory non-compliance.
For languages with significant regional variation such as European Portuguese versus Brazilian Portuguese, French as used in France versus Canada, or English variants across the United States, United Kingdom, and Australia, the standard recommends using geographical usage labels to indicate where each term variant is preferred. The database should also track language status (official language, minority language, technical community usage) to provide appropriate guidance for term selection in different communication contexts and regulatory environments.
3. Domain Modeling and Concept Systems
ISO 26162-3 emphasizes the critical importance of embedding terminology within a well-structured concept system that reflects the actual knowledge structure of the domain. Each terminological entry should be linked to a subject field classification, and concepts should be organized in hierarchical and associative networks that mirror how domain experts think about their field. This approach enables users to navigate from broader to narrower concepts through systematic browsing and discover related terminology through associative connections, significantly improving the usability and discoverability of terminology assets.
A well-constructed concept system not only helps users find terms faster but also reveals gaps in terminology coverage, identifies redundant entries that should be merged, and supports the systematic development of new terminology for emerging technologies and evolving domain concepts.
The standard provides comprehensive guidelines for concept system diagramming, recommending graphical representations such as concept maps, tree diagrams, and network graphs that show generic-specific (hierarchical), part-whole (meronymic), and sequential (temporal or procedural) relationships. These diagrams should be stored as part of the terminology database metadata and made accessible to users for navigation, quality review, and training purposes, providing a visual map of the domain knowledge structure that complements the textual term entries.
Populating a terminology database without first establishing a concept system leads inevitably to inconsistent classification, overlapping and contradictory definitions, and missed synonym identifications. Organizations planning large-scale terminology projects should invest adequate resources in concept system development before beginning data entry, as retrofitting a concept system to an existing populated database is significantly more difficult and costly than building it upfront.
4. Frequently Asked Questions
Q1: How long should a terminological definition be?
The standard does not prescribe a specific length but emphasizes conciseness and precision. A good terminological definition typically ranges from 15 to 40 words and contains only the essential characteristics needed to distinguish the concept from all others within its subject field, avoiding encyclopedic detail.
The standard does not prescribe a specific length but emphasizes conciseness and precision. A good terminological definition typically ranges from 15 to 40 words and contains only the essential characteristics needed to distinguish the concept from all others within its subject field, avoiding encyclopedic detail.
Q2: Can definitions include images, diagrams, or multimedia elements?
Yes, the standard fully supports multimedia annotations including images, diagrams, audio pronunciations, and video demonstrations. These are considered valuable supplementary content that enhances understanding, but they should never replace textual definitions which remain the primary content element.
Yes, the standard fully supports multimedia annotations including images, diagrams, audio pronunciations, and video demonstrations. These are considered valuable supplementary content that enhances understanding, but they should never replace textual definitions which remain the primary content element.
Q3: How should deprecated or obsolete terms be handled in the terminology database?
Deprecated terms should remain in the database with a status label of deprecated and a clear reference to the preferred replacement term. They should never be deleted because users may encounter them in legacy documents and need to understand their meaning and find the current preferred term.
Deprecated terms should remain in the database with a status label of deprecated and a clear reference to the preferred replacement term. They should never be deleted because users may encounter them in legacy documents and need to understand their meaning and find the current preferred term.
Q4: What is the recommended review cycle for terminology content quality assurance?
The standard recommends a risk-based approach core terminology for safety-critical or regulatory domains should be reviewed annually, while general technical terminology can follow a two to three year review cycle. Each entry should carry a review date and next review due date for the complete termbase. The standard emphasizes that regular quality audits are essential for maintaining the value and reliability of terminology assets over time.
The standard recommends a risk-based approach core terminology for safety-critical or regulatory domains should be reviewed annually, while general technical terminology can follow a two to three year review cycle. Each entry should carry a review date and next review due date for the complete termbase. The standard emphasizes that regular quality audits are essential for maintaining the value and reliability of terminology assets over time.
