You are here
Richard N. Leigh
Managing Digital Cultural Objects: Analysis, Discovery and Retrieval
edited by Allen Foster and Pauline Rafferty
Managing Digital Cultural Objects is a collection of nine pieces by 14 different authors, edited by two information professionals. The book was originally published in England, and 12 of the 15 contributors work in Europe. Both of the editors and four of the authors have ties to Aberystwyth University in Wales. The contributors are primarily professors and upper managers, supplemented by a few librarians and a software developer. The editors explain that “this book explores the analysis and interpretation, discovery and retrieval of non-textual objects, including image, music and moving image. Its purpose is to inspire prospective students to develop creative and innovative research projects at Master’s and PhD levels.” (xvii) Practitioners can also benefit from this book, though, especially considering that their institutions likely own collections that pose the specific challenges the authors describe.
Part 1 (“Analysis and retrieval of digital cultural objects”) consists of three chapters that introduce the general theoretical issues related to creating metadata for non-textual, unpublished, and unique materials.
In Chapter 1 (“Managing, searching and finding digital cultural objects : putting it in context”) editor Pauline Rafferty asks the big questions : How can we know if we really understand an object? What words should we choose to communicate our understanding? How can we remove bias, both known and unknown, from our metadata? When is it worth our resources to do things perfectly? When should we describe things “by hand”, and when should we index things by machine? Etc.
In Chapter 2 (“Data modelling for analysis, discovery and retrieval”) author Sarah Higgins explains the importance of determining one’s audiences and how they will use the materials. She touches upon the OAIS Reference Model, the 5S Model for Digital Libraries, and the Digital Library Reference Model. She also discusses software, systems interoperability, metadata standards, authority control, and the semantic web.
In Chapter 3 (“The digital traces of user-generated content : how social media data may become the historical sources of the future”) author Katrin Weller explains how user-generated content might be used as a primary source to understand contemporary reactions to elections, natural disasters, protests and civil unrest, and mass cultural events (65-66). Specific platforms discussed include Wikipedia, blogs, Twitter, social networks (ex : Facebook), photo communities (ex : Flickr), and video communities (ex : YouTube).
Part 2 (“Digitization projects in libraries, archives, and museums : case studies”) consists of three chapters that present case studies from two national libraries in Europe.
In Chapter 4 (“Visual digital humanities : using image data to derive approximate metadata”) the four authors discuss the digitization of artworks at the National Library of Wales. They focus primarily on acollection of 325 digital photographs of oil paintings by Sir John Kyffin Williams, collected from various museum websites, catalogues, and other sources. The attached metadata is often technical in nature, using mathematical concepts from the field of digital painting analysis (many of which were originally developed for authentication purposes). Researchers were able to determine approximate dates of creation for dozens of these paintings, over two-thirds of which had originally been undated!
In Chapter 5 (“Managing and preserving digital collections at the British Library”) authors Maureen Pennock and Michael Day provide an overview of digital projects at the British Library. They focus especially on their Digital Preservation Research and Development program and the various workstreams that encompasses. This chapter might have been a better fit in a different section of the book, as its “case studies” tend toward the theoretical rather than the practical; the challenges of collecting e-journals are listed, for example, but there are no “lessons learned” from working with a particular collection of specific e-journals.
In Chapter 6 (“Digital preservation of audio content”) author Will Prentice details the digital preservation of digital audio material at the British Library. He argues passionately for not only the digitization of audio materials from the original analog sources, but also for triage of at-risk items, adherence to international best practices, and the usage of state-of-the-art technology throughout the process.
Part 3 (“Social networking and digital cultural objects”) consists of three chapters that “consider digital cultural documentation discovery and retrieval within the context of Web 2.0.” (xviii)
In Chapter 7 (“Photos : Flickr, Facebook and other social networking sites”) author Corinne Jörgensen explains the benefits of both concept-based retrieval (i.e. traditional metadata assigned by humans) and content-based retrieval (i.e. technical metadata derived by computers) of images. She believes that user tagging is the future, since there are so many photographs out there and limited resources for “professionals” to describe them. She encourages researchers to attempt to understand the behaviors and motivations of users who create tags, in hopes of both improving pre-existing controlled vocabularies and refining the software used for image retrieval.
In Chapter 8 (“Searching and classifying affinities in a web music collection”) author Nicola Orio discusses “the detection and classification of affinities between music tracks in a music digital library that is the basis for an online web service of music delivery.” (184) The collection in question contained over 350,000 audio tracks, which were originally compiled by an Italian broadcaster to be used in the soundtracks of television programs. (187) The author explains how mathematical concepts were used to create graphs of each track, which could then be visually compared to identify exact duplicates, near-duplicates, and far-duplicates.
In Chapter 9 “(Film retrieval on the web : sharing, naming, access and discovery”) authors Kathryn LaBarre and Rosa Inês de Novais Cordeiro compare the film retrieval systems at the British Film Institute (BFI), the American Film Institute (AFI), IMDb, YouTube, and Netflix. The authors attempt to gauge each organization’s reliance on text-based strategies (ex: transcripts) and content-based strategies (ex: image characteristics). They conclude that the behaviors of film archives are primarily driven by preservation issues, and that the behaviors of film businesses are mostly driven by commercial concerns.
I enjoyed reading Managing Digital Cultural Objects, but it often felt more like a computer sciencejournal than a library sciences monograph. A high percentage of the content consists of charts, graphs, and references. That will be a boon for serious researchers, but this semi-casual reader (basically in search of continuing education) would have appreciated more analysis. The title led me to expect a hands-on guide to improving the organization, description, and visibility of my institution’s digital collections. Instead, this book is more like a collection of cutting-edge research on non-textual collections, interspersed with library futurist thought. Either approach is welcome, of course, as long as one knows what’s in store.
Published in 2016 by: Neal-Schuman, Chicago (xix, 227 p. ; 23 cm.) ISBN 978-0-8389-1343-7 (softcover : $88.00)
Reviewed by: Richard N. Leigh
Metadata & Digital Resources Developer
Ball State University