The term “reuse” appears very often in the Semantic Web literature, encouraging the adoption of existing data specifications into new ones rather than redefining the same/similar concepts once again.

But the word “reuse” has different meanings depending on the context.

Therefore, this ambiguity creates challenges for knowledge engineers, developers, and organizations aiming to implement reusable data models effectively.

This blog explores what “reuse” means, how it is approached within the SEMIC framework, and why clarifying its nuances is the first for developing robust and interoperable Semantic Data Specifications (SDS).

1. What is “Reuse,” and Why is it not Self-Evident?

In semantic modeling, “reuse” is not a one-size-fits-all term. Its meaning often depends on what is being reused and how. Within SEMIC, reuse is addressed systematically to avoid misinterpretation and ensure consistent implementations.

🟢 The Problem of 𝗼𝘄𝗹:𝗶𝗺𝗽𝗼𝗿𝘁

In the Semantic Web, reuse is commonly defined as incorporating elements from one specification into another. While this sounds logical, the implementation reveals significant hurdles, particularly when using the OWL (Web Ontology Language) mechanism for importing.

🟢 The Intended Simplicity of 𝗼𝘄𝗹:𝗶𝗺𝗽𝗼𝗿𝘁

The 𝗼𝘄𝗹:𝗶𝗺𝗽𝗼𝗿𝘁 construct is designed to allow one ontology to include the definitions of another directly. This means a knowledge engineer could theoretically import an established specification and instantly leverage its concepts, properties, and constraints. However, the reality is far more complex.

🟢 Why 𝗼𝘄𝗹:𝗶𝗺𝗽𝗼𝗿𝘁 Proves Insuficient

→ Lack of Context Sensitivity:

When importing an ontology via 𝗼𝘄𝗹:𝗶𝗺𝗽𝗼𝗿𝘁, all definitions, including properties and constraints, are adopted wholesale. This can introduce irrelevant or conflicting elements into the new specification, especially when only a subset of the original ontology is needed.

→ No Control Over Granularity:

𝗼𝘄𝗹:𝗶𝗺𝗽𝗼𝗿𝘁 doesn’t allow selective reuse. If you need one class or property from an ontology, you must import the entire specification, often leading to unnecessary complexity.

→ Versioning and Maintenance Challenges:

Imported ontologies evolve over time. If the source ontology is updated, the dependent SDS must also be revised to maintain alignment, creating a continuous synchronization burden.

→ Semantic Misalignment:

Definitions in the imported ontology may conflict with the semantic nuances required in the new context. For example, the original ontology’s class 𝗽𝗲𝗿𝘀𝗼𝗻 may not fully align with a specific domain like “Public Sector Citizen.”

🟢 Beyond 𝗼𝘄𝗹:𝗶𝗺𝗽𝗼𝗿𝘁: The Need for Adaptation

To address these limitations, SEMIC emphasizes careful adaptation over direct import. This involves either:

→ Reusing classes or properties “as-is” when they align perfectly with the new context.

→ Adapting reused elements through specialization (e.g., creating a subclass) or terminological adjustments to fit specific domain requirements.

These approaches ensure that reuse enhances interoperability without introducing semantic inconsistencies.

2. An Example of Un-Reusable SDS in the EU

The European Union has been a leader in promoting semantic interoperability, particularly by creating Core Vocabulary and Application Profiles. However, achieving reusability in such specifications has proven challenging, as seen in the Core Public Organisation Vocabulary (CPOV) example.

🟢 The CPOV Challenge

The CPOV was designed to standardize how public organizations are described and referenced across EU Member States. While the vocabulary offers a robust framework, its re-usability in diverse national contexts has faced obstacles:

→ Conflicting Definitions Across Member States:
Public organizations differ significantly in structure and legal status across countries. For instance, the term “public organization” may encompass non-profits in one Member State but exclude them in another. Re-using the CPOV “as-is” often leads to semantic misalignment.

→ Rigid Property Constraints:
Some properties in the CPOV are defined with strict domain and range constraints, making them incompatible with certain national implementations. For example, the property hasJurisdiction might not map directly to the governance structures in decentralized administrations.

→ Terminological Gaps:
The labels and definitions provided in the CPOV sometimes lack the nuance needed for local contexts. Translating or adapting these terms without semantic shifts requires careful effort, which is not always feasible.

→ Dependency on 𝗼𝘄𝗹:𝗶𝗺𝗽𝗼𝗿𝘁:
In some cases, efforts to reuse the CPOV relied heavily on 𝗼𝘄𝗹:𝗶𝗺𝗽𝗼𝗿𝘁, introducing all constraints and properties, including irrelevant ones, into national implementations. This often necessitated manual adjustments to align the vocabulary with local requirements.

🟢 Moving Toward Re-Usable SDS in the EU

The lessons from the CPOV highlight the importance of:

→ Selective Reuse: Reusing only relevant classes or properties rather than entire specifications.

→ Semantic Adaptation: Creating specialized subclasses or sub-properties to address local nuances.

→ Clear Documentation: Providing explicit notes and references to clarify the context of reused elements.

→ Tool Support: Using tools like model2owl to automate adaptation and synchronization processes, reducing the manual effort involved.

3. Conclusion: Reuse as a Strategic Process

The word “reuse” is deceptively simple, but its implications in semantic specifications are far-reaching. “Reuse in semantic specifications is a strategic process that requires thoughtful planning, precise adaptation, and ongoing maintenance. The challenges of 𝗼𝘄𝗹:𝗶𝗺𝗽𝗼𝗿𝘁 and the CPOV example underscore the need for careful consideration in implementing reuse.

Using SEMIC’s principles of selective adaptation and advanced automation tools, organizations can create SDS that are both reusable, meaningful, and interoperable in diverse contexts.

We invite you to explore this article for additional insights about reuse.

Meaningfy continues to support the European Commission’s initiatives, leading the charge toward a transparent, efficient, and interconnected European public sector. If you represent a European Institution or a public company that needs to implement an interoperability solution, contact us, and we’ll help you implement it effectively.