Master theses

A recurring cost driver for data intermediaries is that generic visualization and form creation for missing/invalid data is still largely handcrafted per dataset and per application. DiSHACLed frames this as the third major hurdle: when datasets are incomplete (or need validation), organizations fall back to building custom forms, even though much of the structure is already present in SHACL application profiles and dataset shapes. The project calls out that existing prototypes can generate forms from shapes, but the missing piece is formalization and interoperability: a shared approach that works across development contexts (server-side forms, PWAs, native apps) and across vendor toolchains, while still allowing limited, controlled customization.

Identify (or define) a widely adoptable web form specification that can serve as a target for translating SHACL shapes into UI definitions, then implement a generic library/service that performs this translation and supports both data viewing and editing (including prompting for missing fields). The thesis should include a validation framework that compares generated forms against high-quality manually built forms, quantifies where generation is sufficient versus where human customization remains necessary, and produces concrete guidance (and test cases) for interoperable “shape-to-form” generation that project partners can integrate into their products.