Computer infrastructure for supporting the scientific community in Serbia is continuously increasing its capacity, especially in the domain of setting up institutional repositories. Different software platforms can fulfill the requirement of institutional repositories (in-house solutions, Zenodo, DSpace, Omeka, Islandora, EPrints, etc.). All of those platforms share the same functionalities that are common for intuitional repositories, but they are usually adapted to policies that are demanding for particular institutions. With customization that had to be taken, it has become too difficult to track all the changes in those repositories and monitor their progress over time. The main idea of this poster is to propose practical guidelines that can be integrated within intuitional repositories to make requirements for developing a platform capable of establishing metadata exchanging protocol based on OAI-PMH (Open Archives Initiative Protocol for Metadata Harvesting) and establishing procedures that can improve quality of metadata exchanging between multiple platforms. Those guidelines can help in increasing compatibility with multiple platforms and would support overall metadata quality. The majority of institutional repositories in Serbia provide their metadata through OAI-PMH using DC (Dublin Core) dissemination format. Unfortunately, this format isn’t capable to manage two important issues. DC format has a limited metadata schema that is insufficient for providing a complete record description. The second issue mostly affects PIDs (persistent Identifiers) that have become widely used by the science community over the years (e.g. ORCiD). Even though there are initiatives for expanding the DC format, there isn’t a standard for disseminating PIDs. The quality of metadata and the diversity of metadata types change over time. The repository is technically able to support new changes, but those metadata are locked in the local system and makes the platform less interoperable and compatible with other systems. This article is focusing on suggesting technical solutions for overcoming those issues. Adopting good practices and suggestions by the international scientific community, DC can be extended technically with additional metadata types. Those additional types wouldn’t affect the DC standard. Repositories would be still considered DC compatible, but also repositories will provide additional metadata for those platforms that can harvest additional fields. The majority of intuitional repositories in Serbia are based on the DSpace platform. The repository platform DSpace comes with its metadata schema. DIM (DSpace Intermediate metadata) is a metadata schema that is easily adaptable to all necessary changes. This is a better alternative when it is necessary to disseminate metadata collected from internal unstructured databases. If it is necessary to support a standardized format, the DIM schema can be transformed with XLST (Extensible Stylesheet Language Transformations) into more suitable formats.

This poster was presented during the 4th Open Science Fair (25-27 September, Madrid, Spain)