Professor Christine Borgman's latest book, Scholarship in the Digital Age, addresses the evolution of scholarly research and publishing, with a particular focus on the growing value of data as a scholarly product. Data present a complex set of issues for access and management, including relevance, authenticity, and interpretability. Borgman's book offers a thorough overview of scholarly practices for publishing research and explores how data have become essential scholarly products to be archived for future reference and use.
BookZone: Although the creation and distribution of digital content is a much faster process than the creation of paper publications, the nature of data presents other problems, some of which may yet be unforeseen. One of these problems lies in the task of indexing information, or building the infrastructure for scholarly data so that it is as accessible as possible. Can you describe some of the approaches involved in this daunting task?
Borgman: Once in digital form, text becomes more easily indexed and searchable, which is a distinct advantage of digital publishing. Searching for images, music, formulae, and data is much more difficult than for text, although considerable progress is being made.
Most writing, editing, and editorial production is done in digital form regardless of the final format. The only significant difference between print and digital publication is the form of dissemination. Peer review remains the coin of the realm for true publishing, as distinguished from blogs and other informal communication.
What has changed most significantly is the status of data. Data used to be considered just an interim product that could be discarded once the journal article or conference paper describing them was published. In many fields, data are now viewed as an important scholarly product in and of themselves, to be used and reused by others. Data can be combined for longitudinal or comparative analyses, and texts can be mined for new interpretations. Access to rich sources of data can enable researchers in the sciences, social sciences, and humanities to ask new questions in new ways.
BookZone: Scholars are subject to various stages of certification before they are published: peer reviews; registration, et cetera. Now that anyone can "publish" on the Internet, particularly through personal blogs, will scholars eventually have the freedom to publish research independently, or will their credibility always rely upon traditional procedures of publication?
Borgman: Online dissemination is often confused with publishing. While "publish" literally means "to make public," scholarly publishing, by definition, requires peer review. The notion of peer review, or subjecting your work to the scrutiny of your peers before it is accepted as scholarship, dates back to St. Augustine, and was first codified by Sir Francis Bacon early in the 17th century. Blogs, YouTube videos, Wikipedia, citeulike, del.icio.us, and other digital sources contain some extremely valuable content, but rarely would they be considered scholarly publications.
BookZone: Let's say there's some well-known scientist and he or she is writing a theory that he or she is posting in a blog. Would that piece of writing not be taken as seriously because the author did not go through the process of peer review?
Borgman: Let me give you a general answer and then highlight some specific issues. The general answer is that peer review continues to be the gold standard, as it has for several centuries. Peer review is an imperfect system, of course. It can be overly conservative, favor established scholars, and accept fraudulent papers occasionally. Journals are tinkering with open reviewing and other innovations, but these are still means to scrutinize work prior to public release. Journals, conference papers, and books remain the authoritative means to stake one's claim to a theory or a result. But that doesn't mean that blogs and other types of informal online communication are without value. Blogs offer very rapid dissemination of ideas, events, and opinions. They also can be used to call attention (good or bad) to published work, thus tying together the network of formal and informal communication.
One specific issue is the relationship between journal publications and digital repositories. Authors frequently deposit copies of their articles, either preprint or final form, in institutional or disciplinary repositories. Institutional repositories, such as the UC eScholarship Repository are run by universities, while disciplinary repositories are those covering individual fields, such as arXiv for math and physics or RePEc in economics. Some fields, some funding agencies, and some universities require authors to deposit their papers in these repositories, as it makes them much more accessible. Repositories also can be used to post working papers or other semi-formal documents that would not usually be submitted for peer review. These open access mechanisms greatly increase dissemination. Their purpose is not to bypass peer review, although in a few fields, some very senior people are putting things only in repositories or on their Web sites.
The best known case of scholarly dissemination without journal review is that of the Russian mathematician, Grigory Perelman, who claimed a solution to Poincaré's Conjecture. He chose to put it only in an open access repository, arXiv.org. However, by posting his work there, he was inviting public scrutiny, which he received. Upon public acceptance of his accomplishment, he was offered a Fields Medal (the "Nobel Prize of Math"), which, by the way, he turned down.
Another form of recognition is citations to one's papers. Assistant professors facing tenure would like to be able to say, 'I'm publishing in the highest citation-impact journal in this field,' or 'I am the tenth most-cited person in my field for this year.' The most widely accepted metrics for citations are those published by Thomson Scientific as part of the The Science Citation Index, Social Sciences Citation Index, and Arts and Humanities Citation Index. Their Journal Impact Factor is a ratio of citations received in one year to the number of articles published in that journal over the prior two years. That impact factor is based only on the journals indexed by Thomson, however. Other metrics based on web citation, known as "webometrics," are becoming popular, as are some usage and download metrics. A number of analyses indicate that papers placed in digital repositories receive more citations than those published only in journals or conference proceedings.
BookZone: As you note in your book, libraries have been diligently adapting to new forms of information over the years and subsequently, their archiving processes have evolved. Many have taken to creating metadata or surrogates as representations of physical documents or specimens in their collections in order to make them available via the Internet. Will libraries eventually taper off from collecting items and instead, transition into virtual sources of information through the use of Web links?
Borgman: Although the simple answer is "no" - libraries always will be much more than a set of weblinks - the future role of research libraries is much debated right now. Over spring break I was at a workshop in Italy on what the global research library in 2020 will look like. One problem for libraries is that good infrastructure is invisible; you may not see it until it starts to break down. For example, few people were aware of California' s electrical grid until the reliability decreased and the prices increased. When libraries are really doing their job well, users often don't realize how seamless are the services being provided. If you search Google Scholar, click on an article of interest, and retrieve it directly from a publisher database, you are using UCLA library services. Behind the digital scene is a set of technologies and publisher contracts that were implemented, negotiated, and paid for by your library. Certainly digital services are a growth area for academic research libraries. Journals and conference proceedings are tipping toward all-digital publishing. Data repositories are becoming the new special collections. At the same time, print is not going away. Book publishing is healthy, as are online "bookstores." Nor are we going to shred the great book and journal publications we already have. Many documents have significant value as artifacts, in addition to the intellectual or artistic value of their contents. We are not going to scan the Gutenberg Bible and then pulp it, for example. The pendulum of "access versus assets" has swung toward access, particularly for new and current materials. At the same time, libraries are putting more emphasis on their rare materials - their special collections -- because that's what makes library collections distinctive. Libraries depend on sophisticated human expertise to collect, organize, curate, and provide access to digital and print resources. Maintaining digital materials is more difficult and expensive than maintaining print, as digital resources have to be migrated to new technologies as they emerge. Libraries will have hybrid collections of digital and print, and will be both places and spaces, probably forever.
Professor Borgman is currently reading Super Crunchers: Why Thinking-By-Numbers is the New Way to Be Smart by Ian Ayres and Rethinking Expertise by Harry Collins, as well as a selection of books by Bruno Latour. She is currently researching data management and data practices in the Center for Embedded Networking Sensing (CENS), a UCLA National Science Foundation Science and Technology Center, with a "fabulous team" of graduate students, studying how data get made.
