Nine-tailed data monsters? Lessons learnt in the compilation of GlobaLID

Lead isotope data are an important tool for the reconstruction of raw material provenances of non-ferrous archaeological materials. The quality of the provenance reconstruction depends, among other factors, on the comprehensiveness of the reference data the archaeological samples can be compared to. When compiling such databases, three major challenges typically emerge: (a) Access to the respective publications, many of them scattered in conference proceedings and local journals; (b) a huge variety and often incompatibility and incompleteness of meta-information; and (c) language barriers. To overcome these obstacles, a group of colleagues started to tame this data monster in 2018 through the compilation of the global lead isotope database GlobaLID. Its aim is to create in a community-driven process a FAIR database and an associated web application which provides easy access to the database and visualisation tools for people with different backgrounds and knowledge levels. Prototypes were published in late 2021 and work is still ongoing. Many obstacles had to be overcome; many more are already looming on the horizon. After a short summary of the project, some of them will be presented, which we perceive as general challenges in taming such data monsters, and the solutions we came up with. Subsequently we will discuss where we are in urgent need from the scientific communities and related research data infrastructures for tools that allow us to fully tame the data monster into a truly FAIR database.