Beyond the Pages: Early Printed Books in Digital Collections

For early printed books, somewhere in the middle ground between simple image display and multispectral imaging, between metadata searches and virtual reality experiences, there are practical, low-cost, high-impact practices available for digitization and in digital collections. They create new possibilities for digital library users to engage with those early printed books more fully, opening the books to new scholarship and new engagement.

The standard approach to digital presentation of early printed books for scholarly and other users continues to duplicate online the experience of paging through a book (a page turner interface) and reading through its descriptive record (a metadata presentation). Expanded scope brings access to higher resolution images, full text files, linked and harvestable metadata, multi image widgets, and citable permanent links. Yet none of this work enables scholars or other users to easily engage with sufficient information about books and their history - and certainly not consistently from one digital collection to the next.

The challenge is primarily one of metadata, and secondarily of digitization and presentation - for both textual and visual data about a book. In addition to the information in a high resolution image and a well researched catalog record, researchers interested in the history of books want access to diverse descriptive metadata not routinely presented, and sometimes not recorded. That metadata ranges from ownership history, multiple works bound together, and binding style to the presence of annotations, printer's devices, tipped in plates, volvelles, watermarks, and so on. Recording, presentation, and exposure of most such metadata would be a relatively straightforward extension of current practices and current technologies. For example, much of the image-level metadata could be recorded by trained undergraduate employees during digitization.

In addition to metadata, some visual information is not commonly gathered at all. Two examples are text-block edges and watermarks, neither of which is digitized routinely, and both of which are invaluable to the interested researcher. When and how do we digitize them? Where do we present them - within the set of standard images? Appended to that set? Separately? Such digitization is not technically difficult and in fact is quite cost-effective to add to routine digitization. Once digitized and noted in the descriptive metadata, such information in digital collections becomes accessible to researchers working with early printed books.

Another kind of information in digital collections is images of the same object (or page) taken at different times, perhaps years apart, or with different equipment or different lighting. This begins to look like a dataset, which can be of very high interest to researchers as well as conservators and preservation staff. In digital collections, the practice and policy could become to include multiple images of the same object along with appropriate technical and descriptive metadata.

I will present specific examples and recommendations for low-cost, high-impact practices for digitization and digital collections. To the extent such practices are adopted, digital collections can provide a great deal more information about early printed books, opening them more fully to the world in practical, straightforward ways.