Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
CAN/CSA ISO/IEC 11179-4-05: A Technical Guide to Formulating Data Definitions for Metadata Registries
Understanding the rules and best practices for creating precise, unambiguous data definitions as specified by the international standard
Introduction
The quality of data definitions is critical to the interoperability, clarity, and reusability of data within any organization or across systems. The international standard ISO/IEC 11179-4:2004, adopted in Canada as CAN/CSA ISO/IEC 11179-4-05, provides rigorous rules and guidelines for the formulation of data definitions. This standard is part of the ISO/IEC 11179 series on Information technology — Metadata registries (MDR), which addresses the structure and management of metadata for data elements. This article examines the scope, core technical requirements, implementation considerations, and compliance aspects of this essential metadata standard.
Standard Identifier Note: The designation CSA ISO/IEC 11179-4-05 refers to the Canadian adoption by the Canadian Standards Association of ISO/IEC 11179-4:2004, with no technical deviations.
Scope of the Standard
CAN/CSA ISO/IEC 11179-4-05 specifically addresses the formulation of data definitions for data elements within a metadata registry. Its primary aim is to ensure that definitions are expressed in a consistent, unambiguous, and computable manner. The standard applies to anyone involved in creating, maintaining, or using data element definitions in a metadata registry environment, including data administrators, metadata managers, database designers, and information modelers.
The scope includes rules for constructing definitions from both a semantic (meaning-based) and lexical (word and phrase-based) perspective. It does not cover the content of specific data elements or the processes for managing the metadata registry itself, which are addressed in other parts of the 11179 series.
Technical Requirements
The standard defines a set of mandatory and recommended rules that govern the structure and wording of data definitions. These rules are designed to eliminate ambiguity, ensure consistency, and support automated processing. The requirements are organized into several categories:
Semantic Rules
- Uniqueness of meaning: Each data definition must express a single, clear concept without overlap with other definitions.
- Essential characteristics: The definition must identify the essential characteristics of the data element and distinguish it from similar concepts.
- Context independence: A definition should not rely on a specific application or system context to convey its meaning.
Lexical Rules
- Use of finite verbs: Definitions should avoid the use of finite verbs (e.g., “is,” “are”) and instead use the present tense verb form (e.g., “represents,” “indicates”) or be expressed as a noun phrase.
- Single term resolution: Each definition must be a single, grammatically coherent sentence or phrase.
- Circularity prohibition: A definition must not refer back to itself or to other definitions in a circular manner.
- Standard terminology: Use terms that are already defined within the registration environment and avoid unfamiliar jargon.
Important: Definitions must not contain definitions of other data elements, nor should they include implementation-specific details such as data type, representation, or value domains. These are captured in other parts of the registry.
Structural Constraints
Each data definition should follow a consistent structural template: “A [supertype] that [differentiae].” The supertype places the concept in a broader class, while the differentiae specify the characteristics that distinguish it from other members of the same supertype.
Key Compliance Criteria
Compliance with CAN/CSA ISO/IEC 11179-4-05 is determined by evaluating whether a set of data definitions meets the required rules. The following table summarizes the essential compliance criteria:
| Rule | Requirement | Compliance Level |
|---|---|---|
| Uniqueness of meaning | Each definition must have a distinct meaning that does not overlap with other definitions in the same registry. | Mandatory |
| Non-circularity | Definitions must not reference themselves or other definitions in a way that creates a logical loop. | Mandatory |
| Essential characteristics | The definition must include the necessary and sufficient characteristics to identify the concept. | Mandatory |
| Use of noun phrase | Definitions should be expressed as a noun phrase or a statement beginning with a present tense verb, not as a full sentence with a finite verb. | Recommended |
| No implementation details | Definitions must not include details about data types, length, or representation. | Mandatory |
| Use of standard terminology | Terms used in the definition should come from a controlled vocabulary or previously defined data elements. | Recommended |
Implementation Tip: Organizations adopting this standard should establish a centralized review process for new or updated data definitions to ensure consistent compliance. Automated tools can help detect non-circularity and uniqueness violations.
Implementation Highlights
Effective implementation of CAN/CSA ISO/IEC 11179-4-05 requires an organizational commitment to metadata governance. Key steps include:
- Training: Data stewards and definition writers must be trained on the semantic and lexical rules outlined in the standard.
- Template adoption: Use a consistent template as recommended by the standard (supertype + differentiae) to structure all definitions.
- Tool support: Deploy metadata registry software that enforces the rules during entry, or develop validation scripts to check definitions against the rules.
- Periodic audit: Regularly review existing definitions for compliance and update those that violate the rules.
The standard also encourages the use of examples and notes within a registry to supplement the formal definition, as long as these do not confuse the core definition itself.
Common Pitfall: A frequently observed error is the use of the word “used to” in definitions (e.g., “A field used to store customer name”). This violates the lexical rules by introducing an implied purpose rather than defining the essential nature. Correct form: “Customer name assigned to an account.”
Compliance Notes
Adoption of CAN/CSA ISO/IEC 11179-4-05 is voluntary but strongly recommended for organizations aiming to achieve high levels of data interoperability, especially those involved in data sharing across departments or with external partners. It is important to note:
- The standard provides both mandatory and recommended rules, but a compliant registry should strive to meet all recommendations as well.
- Compliance is assessed by examining a representative sample of data definitions from the registry; e.g., regulatory bodies may require evidence of rule adherence.
- Organizations may also choose to adopt the entire ISO/IEC 11179 series (parts 1–6) to build a complete metadata management framework.
Non-compliance typically leads to ambiguous definitions, misinterpretation of data, and reduced ability to integrate across systems. Since the standard is identical to the international version, compliance with CAN/CSA ISO/IEC 11179-4-05 also aligns with ISO/IEC 11179-4:2004.
Frequently Asked Questions
Q: What is the relationship between CAN/CSA ISO/IEC 11179-4-05 and ISO/IEC 11179-4:2004?
A: CAN/CSA ISO/IEC 11179-4-05 is the Canadian adoption of ISO/IEC 11179-4:2004. It is identical in technical content. The “05” in the identifier indicates the year of adoption by the Canadian Standards Association, not the original publication year.
A: CAN/CSA ISO/IEC 11179-4-05 is the Canadian adoption of ISO/IEC 11179-4:2004. It is identical in technical content. The “05” in the identifier indicates the year of adoption by the Canadian Standards Association, not the original publication year.
Q: Does the standard apply only to data definitions in English?
A: The rules are language-neutral in principle, but the lexical rules are primarily based on English grammar. When applying the standard in other languages, users must adapt the lexical rules while preserving the semantic intent. The Canadian standard is published in English and French.
A: The rules are language-neutral in principle, but the lexical rules are primarily based on English grammar. When applying the standard in other languages, users must adapt the lexical rules while preserving the semantic intent. The Canadian standard is published in English and French.
Q: Can a data definition include examples?
A: Yes, the standard allows for examples to be attached to a data definition as supplementary material (e.g., in a “example” field in the registry), but examples should never be part of the definition sentence itself, as they could introduce ambiguity or limit the scope.
A: Yes, the standard allows for examples to be attached to a data definition as supplementary material (e.g., in a “example” field in the registry), but examples should never be part of the definition sentence itself, as they could introduce ambiguity or limit the scope.
Q: How does this standard interact with other parts of the ISO/IEC 11179 series?
A: Part 4 works closely with other parts: Part 3 (Registry metamodel and basic attributes) defines the attributes of a data element, while Part 4 defines the rules for the “definition” attribute. Part 5 (Naming and identification principles) and Part 6 (Registration) provide complementary guidance.
A: Part 4 works closely with other parts: Part 3 (Registry metamodel and basic attributes) defines the attributes of a data element, while Part 4 defines the rules for the “definition” attribute. Part 5 (Naming and identification principles) and Part 6 (Registration) provide complementary guidance.
— Article updated 2026. Based on CAN/CSA ISO/IEC 11179-4-05 and ISO/IEC 11179-4:2004.