KEY TAKEAWAYS

• A recent study found that millions of published works are not stored in a digital archive.
• This lack of archiving means articles are at risk of being lost from the scientific record.

In an evolving publishing landscape that faces challenges ranging from paper mills to AI-generated content, protecting the integrity of the scholarly record is perhaps more important than ever. But, as Sarah Wild describes in a recent Nature News article,  current inadequacies in digital preservation risk the loss of millions of scholarly articles.  

Wild reports on a study by Martin Paul Eve published in the Journal of Librarianship and Scholarly Communication that examined almost 7.5 million publications to see if they were sufficiently archived. The publications all had digital object identifiers (DOIs), but only 58% had been preserved in a major digital archive. Twenty-eight percent of articles had not been preserved in an archive.

Twenty-eight percent of articles had not been preserved in an archive.

Eve acknowledged that the study had its limitations; for example, the authors only examined works with a DOI and did not check every available repository. Nonetheless, the analysis raises important issues. Firstly, articles stored in just one digital archive may be lost if the link were to stop working or the organisation shut down. Secondly, smaller publishers are at a greater risk of insufficiently preserving works compared with larger publishers, due to a lack of funds, expertise, or technology. Eve suggests implementing various measures to improve digital preservation, including tighter requirements from DOI registration agencies and education on the issue.

Wild highlights the need for greater awareness on the risks associated with poor digital preservation methods and calls for researchers to consider “long-term sustainability” in research archiving.

————————————————–