An investigation into digital publications recently revealed that over a quarter of scientific articles published on the internet are not adequately preserved, meaning that many published papers are lost without notice. Here are the details regarding the missing research paper data…

Are the archives of the scientific community secure enough?

Where are over 2 million research papers? According to data published on January 24th in the Journal of Librarianship and Scholarly Communication, systems aimed at storing articles online, i.e., in digital environments, were found to be insufficient. The initial stage of the study involved calculating the number and systematic nature of the content within the archives. Subsequently, the results of the stored studies were examined directly, revealing discrepancies in the content numbers.

In relation to this, Martin Eve, a researcher at Birkbeck, University of London in literature, technology, and publishing, checked whether 7,438,037 items tagged with digital object identifiers (DOIs) were stored in the archives. Before briefly explaining the study, we need to discuss the DOI technology.

DOI technology assigns a virtual “key” to the data you prepare in the virtual environment. This key can be a number or a different visual symbol. Along with these symbols, scientific articles and data are stored in archives.

Eve’s research revealed that approximately one-third of the articles, i.e., slightly more than 2 million, were not preserved. Furthermore, Mikael Laakso, who works on academic publishing at Hanken School of Economics in Helsinki, suggested that assuming articles stored with DOI data will never be lost is a mistake.

Lastly, Kate Wittenberg, the general manager of Portico, a digital archiving service in New York City, warned that smaller publishers are at a higher risk of failing to protect articles compared to larger publishers.