AUGMENTED PROCESSING TABLE Archival Science meets HumanComputer Interaction

AUGMENTED PROCESSING TABLE Archival Science meets Human-Computer Interaction Ciaran B. Trace, Ph. D; Luis Francisco-Revilla, Ph. D; April Norris; Jeff Crow; Shilpa Shukla The University of Texas at Austin, School of Information ABSTRACT The APT project pioneers the use of surface computing devices in the field of archival science. By combining surface computing technology with Web-based spatial hypertext, APT aims to make the processing of digitized archival collections more efficient, and enhance understanding of the best practices for archival processing. The five-person research team consists of three digital library/HCI researchers and two archival researchers. This work is partially funded by a University of Texas, School of Information Temple fellowship. For information about the first APT prototype see the upcoming paper for the Theory and Practice of Digital Libraries [TPDL] conference. [2] CONTACT Ciaran B. Trace School of Information, The University of Texas at Austin cbtrace@ischool. utexas. edu 512 -232 -3508 http: //www. ischool. utexas. edu/~cbtrace INTRODUCTION DISCUSSION Although still in the analysis phase, some expected outcomes of the research project are as follows: The APT project is studying existing best practices for archival processing and envisioning alternative practices that employ new interactive technologies such as gestural and tangible interfaces. q The APT project is documenting the tasks and tacit knowledge that archivists use to arrange collections. ØThe project is modeling what archivists actually do when they process [stages, processes, actions etc] From a practitioner perspective, the APT system supports processing as a key step in organizing records before they become part of a digital library/archive system. The system reinvents the archival workflow by taking a digitize first, process second approach. q The APT project is creating a new workflow for the process of archival work [digitize first, process second]. Ø In this new workflow digitization, processing, and access will be more tightly integrated From an educational perspective the project looks at the role of technology in supporting students’ ‘learning by doing’ – supporting decision-making, and enabling students to better document their own work and to reflect on the choices made during the processing task. APT also looks at the role of technology in enabling faculty to move beyond theoretical, lecture-based teaching such that they can better understand the nature and extent of student learning. q The APT project also suggests new ways of processing for the future. Ø Collaborative group processing [on-site] Ø Archivist working with a remote domain expert Ø Processing via large scale or small scale devices (tablet, phone, etc) q Another likely outcome of the APT project is a new paradigm for the archival finding aid. Ø Spatial Hypertext Finding Aids in 2 D RESEARCH DESIGN The APT prototype has undergone formal evaluation via a comparative usability study. The study followed a balanced experimental (“within subjects”) design with 16 participants and 8 experimental conditions (2 systems x 2 collections x 2 sessions). For each task the participant had to process an archival collection using either the APT or baseline system (paper on a table). A “think aloud” protocol was used to find out what participants were thinking, feeling, expecting etc. Data gathered in the course of the study includes: q Survey data (initial survey and exit survey) q Process data Ø Observations (researcher notes, audio & video data) Ø Task analysis (audio & video data, HTML & WARP files) Ø Participant generated data (arrangement schemes) q Bottom line data (dependent variables – time, quality of arrangement etc) Overall, over 1 TB of video data has been recorded and will be analyzed using the usability software tool, Morae (Manager). SYSTEM APT consists of a custom-made, large surface (5’x 5’ total, 47”x 28” interactive), interactive tabletop computer that runs a specialized spatial hypermedia application [WARP] for digital archival processing. Spatial Hypermedia can be thought of as “using a big 2 D space (a canvas) for sorting information. ”[1] In spatial hypermedia all objects are placed in a space and have explicit visual characteristics such as horizontal and vertical position, background color and font size. When document images are first imported into APT they are tiled across the workspace to give an overall impression of the size of a collection. In the initial state all items are located at the root level of the workspace hierarchy. Users can match, move, scale, and rotate items freely in 2 D. Users can create groups, and add items to them, which in turn can be added to higher level groups, like subgroups and series. REFERENCES 1. Grønbæk, K. , Kristensen, J. F. , Ørbæk, P. and Eriksen, M. A. (2003). “Physical hypermedia”: Organising collections of mixed physical and digital material. Proceedings of the fourteenth ACM conference on Hypertext and hypermedia (HYPERTEXT '03). 2. Jeff Crow, Luis Francisco-Revilla, April Norris, Shilpa Shukla, and Ciaran Trace, “A Unique Arrangement: Organizing Collections for Digital Archives and Libraries. ” Forthcoming in the Proceedings of the International Conference on Theory and Practice of Digital Libraries (TPDL), 2012.