Information science has produced a large body of literature on information use, most of which has concentrated on scientific information and communication. However, the results of science user studies can have limited applicability to other fields that differ in their modes of inquiry. While we have found that certain information practices hold constant for interdisciplinary research across the sciences and humanities, we have also identified critical variations. Moreover, many of the current science-based initiatives may not be cogent for research traditions that rely on contextual richness. For instance, important aspects of document modularity, terminology exchange, and collaborative information spaces will differ in the humanities, presenting unique problems and priorities. Recently in our field, investigators have begun to stress the social, ecological, and content-oriented nature of information and knowledge, focusing on the differing cultures, communication patterns, and vocabularies of literature-producing subject fields , , . Our approach follows this trend and is grounded in studies of knowledge creation and work practices that focus on the role of information in relation to material resources, work practices and environments, and intellectual communities , .
We are concentrating on humanities scholars in this phase of our research because we believe that their work reflects fundamental processes that should be supported in existing and future libraries. In particular, we are concerned with the ability to sustain humanists' reliance on a range of primary and secondary materials, in both physical and digital formats , as well as their deep engagement with texts, artifacts, and other forms of information. These features may ultimately distinguish working research libraries from storehouses of physical and digitized materials. By targeting a subset of interdisciplinary humanists, this study foregrounds the complications of accessing, managing, and integrating information from both heterogeneous subjects and sources.
We conducted semi-structured interviews with twenty-five center affiliates. Ten of the scholars were targeted for more in-depth case study, which included follow-up interviews and content and citation analysis of published papers, which is ongoing. The interviews covered how the scholars gather and keep up with information, communicate with colleagues, their reliance on computers and electronic information systems, and their overall scholarly work practices. The sessions averaged about 60 minutes and were audio recorded. Textual accounts of each interview were constructed based on partial transcription of the tapes and field notes. Iterative rounds of descriptive and thematic coding were performed with ATLAS.ti, software designed for qualitative data analysis and theory building. The resulting codes were brought together into conceptual networks that describe strategies, interactions, conditions, and consequences--relations considered significant for grounded theory development .
Three specific points that relate to computing and the scholarly process emerged from our data. First, digital texts are not widely used, but online bibliographic materials play a consistent yet minor role in the information routines of the scholars. Selected computer-based systems are being adopted with continued heavy reliance on conventional libraries, personal collections, and information gathered through colleague networks. Second, web-based descriptive materials such as catalogs and guides are highly valued, especially in preparation for travel to research collections. They allow scholars to explore the material that surrounds certain items in an archive or library and can provide a path to other relevant sources beyond the boundaries of the collection. Third, information technologies play a significant role in personal communication, information management, the writing process, and the generation of ideas. The interpretation and integration that results from interacting with research materials are embedded in these activities.
While it is possible that these findings are representative of humanities scholars in general, the resources and practices of these interdisciplinary researchers are best understood as part of the "dialectic process" that is the true method of interdisciplinary work . The scanning, gathering, compiling, consulting, and verifying of evidence and ideas garnered from texts, people, and other sources are instrumental to this process of exchange and synthesis. In order to work across multiple intellectual domains, these scholars develop strategies for extending the scope of their information field. For example, they are eclectic readers and active browsers who regularly probe for leads in outside domains. At the same time they place particular emphasis on what might be called "push" information sources that deliver diverse information directly into their work routines through channels such as listserves, editorial and reviewing activities, and cross-disciplinary colleague relationships sustained through e-mail and conferences. As a result, potentially fortuitous discoveries in peripheral subject areas increase, and scholars develop links with reliable sources and "locals" to assist in explaining and validating material from unfamiliar territories. Likewise, basic resources such as reference works and textbooks are essential for verification and self-education. As would be expected, complications related to language and audience are pronounced. The natural language used in the humanities is much less effective for communicating and identifying analogies ,  across domains than the more cryptic representations used in scientific discourse.