Writing the Web: Formal Edition
As a researcher, I publish peer-reviewed publications on the topics I explore. You’ll find those publications listed below. If any of them interest you or you’d like to discuss the research further, feel free to get in touch.
2025
In Proceedings of the 24rd International Semantic Web Conference: Posters and Demos
- Ieben Smessaert0
- Julián Rojas1
- Pieter Colpaert2
Working with temporal data on the Semantic Web remains challenging due to SPARQL’s limited support for comparing time literals of different data types and handling floating times without explicit time zones.
These issues are especially problematic when dealing with partial time literals (such as xsd:date, xsd:gYearMonth, or xsd:gYear) and floating times, both of which are common in real-world knowledge graphs like Wikidata.
To showcase the relevance and urgency of the problem, we gathered and reviewed existing discussions, specifications, draft proposals, and examples from deployed knowledge graphs, providing a consolidated starting point for further community dialogue.
We then proposed a solution in the form of a set of SPARQL extension functions—Time Functions—designed to reinterpret time literals as time intervals, enabling consistent and type-agnostic temporal comparisons.
These functions are formally described using the Function Ontology (FnO), and implemented in the Comunica query engine, with a publicly available demo application that allows users to interactively explore and test the functions.
The demo includes curated example queries that highlight both the limitations of existing SPARQL behavior and how the Time Functions enable more accurate filtering and sorting of temporal data.
In addition to providing a technical proposal, we advocate for improved temporal data publishing practices, urging data providers to use accurate data types and explicit time zones to support reliable temporal reasoning in the open-world context of RDF.
2024
In Proceedings of the 23rd International Semantic Web Conference: Posters and Demos
- Jesse Wright0
- Jos De Roo1
- Ieben Smessaert2
The Web is transitioning away from centralised services to a re-emergent decentralised platform. This
movement generates demand for infrastructure that hides the complexities of decentralisation so that
Web developers can easily create rich applications for the next generation of the internet.
This paper introduces EYE JS, an RDFJS-compliant TypeScript library that supports reasoning using
Notation3 and RDF Surfaces from browsers and NodeJS.
By developing EYE JS, we fill a gap in existing research and infrastructure, creating a reasoning
engine for the Resource Description Framework (RDF) that can reason over decentralised documents in
a Web client.
In Proceedings of the 2nd International Workshop on Semantics in Dataspaces
- Ieben Smessaert0
- Patrick Hochstenbach1
- Ben De Meester2
- Ruben Taelman3
- Ruben Verborgh4
Forms are key to bidirectional communication on the Web: without them, end-users would be unable to place online orders or file support tickets. Organizations often need multiple, highly similar forms, which currently require multiple implementations. Moreover, the data is tightly coupled to the application, restricting the end-user from reusing it with other applications, or storing the data somewhere else. Organizations and end-users have a need for a technique to create forms that are more controllable, reusable, and decentralized. To address this problem, we introduce the Declarative Form Description Pipeline (DFDP) that meets these requirements. DFDP achieves controllability through end-users’ editable declarative form descriptions. Reusability for organizations is ensured through descriptions of the form fields and associated actions. Finally, by leveraging a decentralized environment like Solid, the application is decoupled from the storage, preserving end-user control over their data. In this paper, we introduce and explain how such a declarative form description can be created and used without assumptions about the viewing environment or data storage. We show how separate applications can interoperate and be interchanged by using a description that contains details for form rendering and data submission decisions using a form, policy, and rule ontology. Furthermore, we prove how this approach solves the shortcomings of traditional Web forms. Our proposed pipeline enables organizations to save time by building similar forms without starting from scratch. Similarly, end-users can save time by letting machines prefill the form with existing data. Additionally, DFDP empowers end-users to be in control of the application they use to manage their data in a data store. User study results provide insights to further improve usability by providing automatic suggestions based on field labels entered.
