Book inspection and data entry

When the library decides to digitize an item, a library specialist inspects it to ensure that it is suitable for scanning on the DL, Stanford's book-scanning robot. Fragile, one of a kind, and extraordinarily valuable objects are not scanned using the robot. These materials are scanned by hand on overhead book-scanners operated by Digital Library Systems and Services.

After the item is inspected, a record is created for it in the lab database, which contains a variety of descriptive metadata including the item's title, author, and various standard identification numbers. For items that have previously been catalogued by the Stanford Libraries, data is populated directly from Socrates, Stanford's online catalog. For other items, data is populated from other standard sources, such as The Library of Congress catalog and OCLC's WorldCat, or created from scratch.

The lab's middleware, a customized version of Image Access Bscan, allows an operator to describe the hierarchical organization of an item. Metadata templates define what metadata will be captured for a given item. For a book, this can include titles of parts, chapters, front matter and back matter. For a journal, this might include information about issues in the volume, and article titles and authors. This initial cataloguing process sets the stage for associating (or tagging) ranges of images to the different sections of a book.

When the data entry process — which can take as little as one minute — is complete, a database record that contains a unique object identification number for the item is created. This ID number, which is printed on a tracking sheet that is tucked into the book, identifies the item until a digital object for the item is delivered to the Stanford Digital Repository, where it is assigned a permanent identification number and URL.

Next »