Linked Data equals Linked Data? A comparison of technical implementations across CLARIN and beyond

Linked Data is a standard widely used to help provide FAIR data and metadata, particularly in research infrastructures like CLARIN. In this poster a detailed analysis of Linked Data requirements and their technical implementation across various CLARIN repositories and connected authority services is presented. The comparison results reveal notable technical differences in the implementation approaches for Linked Data within the CLARIN service ecosystem and beyond. This leads to a limitation of (meta)data accessibility and interoperability, which poses challenges for cross-repository interoperability and complicates the development of tools that work across the entire CLARIN infrastructure. We conclude with a call for the formulation of an updated set of best practices for Linked Data. These should address the ambiguities observed, particularly concerning standardised content negotiation mechanisms, recommended RDF serialisation formats, a consistent method for identifying the requested object within the RDF graph with an agreed-upon common predicate, and promoting the provision of the same set of metadata regardless of the requested RDF serialisation format.