The materials that are most heavily used by research libraries are no longer part of our collections. They are resources acquired through licenses and have arrangements for access that are governed through metadata and authentication. Few libraries that hold these licenses or subscriptions take copies of the data which are at the core of these resources. The data and metadata which enable the access that these resources depend on is created, updated, and managed by the publishers (be they commercial organizations, learned societies, institutions, or subject communities) that supply them. We should not be thinking of these materials as collections, but more in the category of ILL or document supply. Our ‘management’ or curation of these ‘collections’ has been reduced to invoicing, bulk-upload of metadata, and management of licenses. This activity will undoubtedly become more automated and delivered by higher and higher levels of aggregation and negotiation in the future.
The sense that library collections imply degrees of responsibility for stewardship or curation (acquisition, preservation, security, management, access) is one which is finally gaining traction in the digital era. There are groups of resources which are created by research institutions and for which the on-going responsibilities for stewardship and creation have been devolved to (or at least have ended up with) the library. These include research data, research outputs of all kinds (from pre-prints to dissertations and theses) as well as digitized resources (images, text, even finding aids). We are moving fast from the era of creation to the era of curation of these kinds of resources as we increase the scale, volume, and speed of creation, and as the expectation of funders increasingly requires careful consideration over the issues of management, preservation, sustainability and access.
We have performed these functions well in the print era (albeit in ‘regimes of benign neglect’ to quote Cliff Lynch), but as print becomes a smaller and smaller part of the regular intake of libraries, and occupies a smaller and smaller part of our budgets, our emphasis will (and in many institutions already is) shifting to a focus on the stewardship and creation of materials which are distinguished by the characteristic of uniqueness: these materials may be unique because they have emerged from the research activities of individuals or research groups, they may be unique because they are digital renderings of unique physical materials, or they may be unique because they are the physical materials themselves.: in particular the special collections of research libraries (Rare books, manuscripts, archives, photographs, etc). These materials often require greater efforts in their curation than the commonly held print materials that are now being replaced with digital.
Increasingly the category of special collections is becoming digital or ‘born-digital’. It can include email, spreadsheets, word-processed materials, digital images, sound, and moving images. The curatorial approach requires not only highly specialized technical knowledge (file formats, forensic analysis), but also knowledge and application of IPR and data protection. The curatorial infrastructure for managing these collections is similar to the ‘repository’ infrastructure used for research outputs etc, but this material is different in that it will often be created from outside of research institutions and often without thought for future preservation and use. The research value of personal or organizational digital collections of this type is often very significant, but the challenges of dealing with it on any scale and with any surety are equally un-trivial.
One element of this aspect of collecting is material found on the open web. Although there is significant activity in web archiving beyond the Internet Archive there are a number of domain archiving activities) nevertheless significant areas of the web lack any formal archiving. Web sites in some parts of the world can appear and disappear rapidly, and oppressive regimes and geo-political conflict can often place websites and blogs as being at risk from cyber-attack (both from regimes and terrorists). We will be increasingly interested in developing collections which might be termed the ‘ephemera’ of the web.
All of these activities imply a degree of infrastructure which enables our libraries to deal with these collections: technical digital infrastructures, policy frameworks, staff skills and competencies, in addition to the expertise in subject and content which our libraries have traditionally excelled in.
Ephemera continues to be created in the print world. In some respects the printed output of the 20th century – and even the 21st century – and this now can be considered to be under threat. The material culture of print was an exceptionally important part of the history and culture of the 20th century but aspects of it are at risk as we digitize and rationalize holdings of print (for very sensible and practical reasons). The material culture of, say, the 16th century has been saved and stewarded over centuries thanks to an appreciation of the underpinning materiality of books and manuscripts and due to the multiplicity of holdings. We are already losing much material that illuminates the 20th century because of the fragility of our understanding and appreciation of the material culture of print in the 20th century – provenance, advertising, ephemera, use and re-use of materials – this evidence is being lost as we dispose of copies and rely on digital archives to provide access.
In conclusion, or collections will become increasingly focused on unique assets, both traditional and digital. We need to adjust our organizational regimes to focus on the curation and stewardship of these unique resources, sharing and collaborating more to do so as the complexities and costs are higher than they were before.