In archiving we trust

Abstract

If 25 specialists in preserving scholarly information had sat together in June of 1907 at the University of Berlin on Unter den Linden (Humboldt–Universität zu Berlin), they could likely have agreed that materials stored in the libraries of one of the world’s great research universities in the capitol of the richest and most powerful state in Europe could reasonably be trusted to survive long term. One hundred years later, after the events of the twentieth century had assaulted the collections with fire, water, looters, and censorship, representatives of four digital archiving systems came together to discuss the strengths and weaknesses of their systems face–to–face in front of an audience of librarians, who would have to choose whether any of these systems could be trusted to overcome the unknown events of the twenty–first century. A key conclusion was the need for interoperability and to pool efforts. An alternative to collaboration may be to let archiving systems complete on price, performance and advertising, but then as customers in that market, libraries need to think about how we can test long–term archiving, so that we have real evidence to decide whether the claims of reliability make sense.