Résumé
For many researchers, providing open access to research outputs has become standard prac- tice. Efforts to make data Open and follow FAIR principles have strongly increased over the last decade, as has the number of data repositories offering data publication services. However, older datasets often remain inaccessible, not necessarily due to reluctance, but because of practical barriers such as time constraints or the effort required to digitize and organize legacy data. It is increasingly urgent to preserve and make these older valuable datasets accessible for future generations before they are lost.Scientific journals increasingly require making the underlying data of scientific results publicly available. Funding agencies require data produced as a result of their grants to be made available publicly. To meet this demand, a large number of repositories have been developed, including large generic and institutional repositories. As a result, data published following the FAIR prin- ciples may be hard to discover for other researchers, unless they are guided by the publication. Domain-specific repositories and databases are therefore valuable for making the data discover- able. Moreover, the comparison of different datasets requires the harmonisation of taxonomic identifications or measurement units, which can only be achieved by involving the scientific community. The practice of individualised data sharing makes harmonisation a time-consuming exercise requiring experts in the field.