Transform DOI system into a giant science hub

In its current mode of identification of scientific publications, the digital object identifier (DOI) is not more than a web linking of published material to their publishing sources. When a given DOI is searched in the DOI website (doi.org), we are redirected to the publishing websites, if the material is available, or an error message (Not Found) will appear if the DOI-associated content is not available or has moved to a new location. To bestow a worthwhile value to DOI assignations, I suggest the establishment of a unique persistent DOI database (for e.g., as a DOI hub, DOI library, or DOI indexer) in which all the DOI assigned by publishers and journals will be listed in one and same place with basics bibliographic metadata and complete citation information, including the DOI link itself, authors’ names, manuscripts’ titles, publishing source, date of publication, and ideally abstracts and full text if available for free (open access). As a result, when a DOI is searched in the DOI hub, full bibliographic information should be retrievable regardless of its status in the publishing source. Basic indexation information and metadata associated with published articles will thus be always accessible and findable independently from the publishing sources. A unique, general and long-term preserved DOI hub will make it easy to search, find and cite scientific literature from the various scientific fields even if a journal or publisher ceases its publishing activity.