Are university libraries disappearing?
As physical libraries experiment with new ways of using their physical spaces and attributes, they also embark on the task of becoming digital libraries in their own right.
A digital catalog was first launched in the 1970s; the days of retrospective conversion and the laborious task of noting bibliographic records, the basics that are on the spine or inside cover of the book, and physically entering them into a database. The actual digitization of books, music, and pictures began in the late 1990s and early 2000s – largely undertaken by cash-strapped undergraduates who received scanners – and, with 110,000 volumes of articles, theses and theses covering the long history of the university, chemin.
“We are the oldest public university in the United States, and therefore, taking the humanities as an example, we have a huge, and certainly the best collection, of historical documents from the southern United States here on this campus. There are millions of photographs, newspapers, magazines … so if you want to research the American South, you know you have to come here. But we want to digitize all of this and expand access.
In addition to its physical archives, North Carolina also has a vast array of data available (or in the process of being made available) that can be accessed remotely or from databases within the library itself.
“One of our biologists, for example, has two petabytes of gene sequencing data. [one petabyte is the equivalent to 1,000 terabytes or 1,000,000 gigs of data storage]. They paid for separate storage for it through our IT infrastructure, but it will ultimately be preserved by, available in and through our libraries.
But where to put and how to access all this newly acquired data – finding virtual storage space, if you will – is the challenge.
“The two things you need are bandwidth and storage. The storage is actually quite cheap, and if it’s cloud-based, the main cost becomes the bandwidth needed to transfer stuff. You have to trade the bandwidth for the compute in the library – that means finding new ways to put the compute tools with the stored items so people don’t have to transfer those large data sets. I think these are going to continue to be valuable technical needs. “
We’ve come a long way since Melvil Dewey started decimalizing. This diversification of services provided in physical libraries requires knowledgeable staff who understand the processes.
“Definitely,” said Gary, “most of our students in our master’s of library science, although not compulsory, take at least one database course, so they basically understand the principles and are able to to talk to the advanced technical people – managers, clients, patrons and editors – and be the bridge. It’s kind of a skill set that the modern librarian really wants to have in order to be successful.
Modern librarians need what Gary calls “cyber-carpentry” tools. “This is where you basically take existing tools and resources and reassemble them to solve a problem. The types of librarians we train for the future are being advised. I mean, they might not be a Python programmer per se, but they know enough that they can hack some tools together using other different tools to fix a problem in whatever library they’re in. .