Linked Open Usable Data (LOUD)
A summary page about LOUD, page in construction…
Linked Open Data (LOD) has been pivotal in fostering a web of openly connected datasets. The progression to Linked Open Usable Data (LOUD) underscores an evolution towards making data not just accessible and linked but also readily usable for a broader audience, particularly software developers. This shift is propelled by the understanding that data’s real value is unlocked through its usage. LOUD aims to enhance scalability, both from technical perspectives and in terms of data production and adoption, making data more interconnected and thus more useful.
JSON-LD is a modern web standard designed to simplify the encoding of linked data using the widely known JSON format. As a compact and easy-to-manage data format, JSON-LD enhances the ability to share structured data across disparate systems on the Web, while embedding rich semantic capabilities. It achieves this by allowing data to be serialised in a way that is compatible with traditional JSON, yet interpretable by RDF-based semantic processing tools.
At its core, JSON-LD introduces a context (
@context) that maps terms in a JSON document to Internationalised Resource Identifiers (IRIs) used in RDF vocabularies. This mapping allows JSON documents to be interpreted as RDF graphs, facilitating the integration of structured data into the Semantic Web without requiring developers to depart from familiar JSON syntax. The design of JSON-LD aims to bridge the gap between the ease of use of JSON and the rich data connectivity and interoperability offered by RDF. The development of JSON-LD was motivated by the need to make linked data more accessible to Web developers and to encourage wider adoption of Semantic Web technologies. By embedding semantic annotations directly into JSON, JSON-LD enables developers to contribute to the linked data cloud with a minimal learning curve, accelerating the growth of a semantically rich, interconnected web of data.
Per the JSON-LD 1.1. specification:
JSON-LD is a concrete RDF syntax as described in RDF11-CONCEPTS. Hence, a JSON-LD document is both an RDF document and a JSON document and correspondingly represents an instance of an RDF data model. However, JSON-LD also extends the RDF data model to optionally allow JSON-LD to serialize generalized RDF Datasets.
The strategic adoption of JSON-LD as a means to reconcile the simplicity of JSON with the descriptive power of RDF has sparked an insightful debate about its practical implications and its classification as a polyglot format. A blog post by Tess O’Connor outlines several challenges that emerge from this approach, including issues related to dual data models, algorithmic complexity, and the potential brittleness and performance impacts of JSON-LD’s
@context mechanism. To address these challenges, a strategic focus on simplicity and clarity is advocated. Adopting a straightforward JSON format that doesn’t require knowledge of RDF can simplify interactions and improve the appeal and usability of the data. In addition, creating canonical mappings from JSON to RDF addresses the needs of RDF users while maintaining the accessibility of the base format. In addition, the adoption of stable sets of
@context is critical for consistent data exchange.
However, this polyglot characterisation of JSON-LD has been met with counter-arguments, such as those from Pierre-Antoine Champin, that offer a different perspective. Critics argue that JSON-LD should not be considered a polyglot format in the strictest sense because it operates on a “JSON then JSON-LD” basis rather than an “either/or” scenario. In this view, any JSON-LD processor first interprets the document as JSON before applying the JSON-LD layer, similar to how other JSON formats, such as GeoJSON, encode specific types of information beyond the basic JSON structures. This process does not make these formats polyglots, but rather extensions of JSON that provide mappings to more complex data models, thus emphasising JSON-LD’s role in simplifying the transition to semantically enriched data on the web.
Building upon this understanding, the LOUD design principles emerge as guiding forces in the quest to make data not only accessible and interconnected but fundamentally usable and intuitive for developers and end-users alike.
LOUD Design Principles
One of the main purposes of LOUD is to make the data more easily accessible to software developers, who play a key role in interacting with the data and building software and services on top of it, and to some extent to academics. As such, striking a delicate balance between the dual imperatives of data completeness and accuracy, which depend on the underlying ontological construct, and the pragmatic considerations of scalability and usability, becomes imperative.
Similar to Tim-Berners Lee’s Five Star Open Data Deployment Scheme, Robert Sanderson listed five design principles that underpin LOUD:
- The right Abstraction for the audience
- Few Barriers to entry
- Comprehensible by introspection
- Documentation with working examples
- Few Exceptions, instead many consistent patterns
A. The right Abstraction for the audience
Developers do not need the same level of access to data as ontologists, in the same way that a driver does not need the same level of access to the inner workings of their car as a mechanic. Use cases and requirements should drive the interoperability layer between systems, not ontological purity.
B. Few Barriers to entry
It should be easy to get started with the data and build something. If it takes a long time to understand the model, ontology, sparql query syntax and so forth, then developers will look for easier targets. Conversely, if it is easy to start and incrementally improve, then more people will use the data.
C. Comprehensible by introspection
The data should be understandable to a large degree simply by looking at it, rather than requiring the developer to read the ontology and vocabularies. Using JSON-LD lets us to talk to the developer in their language, which they already understand.
D. Documentation with working examples
You can never intuit all of the rules for the data. Documentation clarifies the patterns that the developer can expect to encounter, such that they can implement robustly. Example use cases allow contextualization for when the pattern will be encountered, and working examples let you drop the data into the system to see if it implements that pattern correctly.
E. Few Exceptions, instead many consistent patterns
Every exception that you have in an API (and hence ontology) is another rule that the developer needs to learn in order to use the system. Every exception is jarring, and requires additional code to manage. While not everything is homogenous, a set of patterns that manage exceptions well is better than many custom fields.
Systems embodying LOUD principles include the International Image Interoperability Framework (IIIF), the Web Annotation Data Model, and Linked Art, demonstrating the ethos of making data more accessible, connected, and usable:
- International Image Interoperability Framework (IIIF), especially the IIIF Presentation API 3.0 which describes how the structure and layout of a digital object can be made available in a standard manner, defining their appearance in viewers and players through the
Manifest, a JSON-LD file bundling all elements of a IIIF resource with essential metadata and structural information.
- Web Annotation Data Model: It offers a standardised framework for creating annotations on web resources, promoting semantic data integration and use across the web.
- Linked Art: A community effort to define a metadata application profile and API for describing and interacting with cultural heritage resources.
IIIF and Linked Art are both community-driven initiatives.
IIIF is both a model for presenting and annotating content as well as a global community that develops shared application programming interfaces (APIs), implements them in software, and exposes interoperable content.
Currently, IIIF has introduced six specifications, with the Image and Presentation APIs being the most notable, both updated to version 3 in June 2020 and often considered the core IIIF APIs. Additionally, the Content Search and Authorization Flow APIs, both at version 2 and released in 2022 and 2023 respectively, are expected to receive updates to match the core APIs’ standards. The Change Discovery and Content State APIs, both in version 1.0, play essential roles in discovering, aggregating, and sharing IIIF resources.
Despite its strengths in facilitating the sharing and interoperability of digital objects, IIIF can sometimes appear disconnected, especially in contexts where discovery and linking to semantic data is not fully realised. However, this perceived disconnect provides an opportunity for improvement through the implementation of robust discovery mechanisms.
Web Annotation Data Model
The Web Annotation Data Model is a World Wide Web Consortium (W3C) standard that provides an extensible and interoperable framework for creating and sharing annotations across various platforms. It defines relationships between resources using an RDF graph, which includes the
annotation, a web resource, the
body, and the
target. This model allows a single comment to be associated with multiple resources.
Here is an example of machine-generated annotations in a IIIF setting. The JSON-LD snippet represents an
AnnotationPage that contains one or more annotations related to a particular IIIF resource.
"value": "Object Detection (vitrivr)",
"value": "<br><small>Detection score: 0.9574</small>",
body of an annotation is where the content of the annotation is defined. In this example, there are three textual bodies:
- A “TextualBody” with the value “person” for
- Another “TextualBody” with the value “Object Detection (vitrivr)” for
- A third “TextualBody” with HTML content indicating a detection score, also for
These bodies represent the content of the annotation, including comments and tags related to the annotated resource.
target specifies where the annotation applies. In this setting, it points to a specific area of a IIIF Canvas. Key details include:
- The source URL, identifying the specific Canvas within this IIIF Presentation API 3.0 resource.
- A selector of type “FragmentSelector”, using the Media Fragments specification (with a value indicating the specific rectangular area on the canvas targeted by the annotation).
- A link (
dcterms:isPartOf) to the IIIF Manifest that the Canvas is part of.
Linked Art is a community and a CIDOC (ICOM International Committee for Documentation) Working Group collaborating to define a metadata application profile for describing cultural heritage, and the technical means for conveniently interacting with it.
|CIDOC Conceptual Reference Model (CRM)
|RDF encoding of CRM 7.1, plus extensions
|Getty Vocabularies, mainly the Art & Architecture Thesaurus (AAT), as well as the Thesaurus of Geographic Names (TGN) and the Union List of Artist Names (ULAN)
|Object-based cultural heritage (mainly art museum oriented)
|JSON-LD 1.1, following REST (representational state transfer) and web patterns
These three systems are complementary and can be used either separately or in conjunction.
A LOUD ecosystem is characterised by an emphasis on interoperability between independently developed systems. This approach to design and implementation promotes semantic interoperability, ensuring that data can be understood and used across platforms and applications without the need for centralised coordination. By adhering to the LOUD principles, systems can communicate more effectively and share data in a way that is meaningful and useful to a wide range of users. This level of interoperability supports the creation of a more integrated and accessible digital environment, where data from different sources can be seamlessly connected and used for a variety of purposes.