Reconnecting the Science and the Humanities through Digital Research

A panel sponsored by GRENDL (the Group for Research in Electronically Networked Digital Libraries).

GRENDL is a consortium which was launched by the University of Kentucky, British Library and CIMATS (the Centre of Information Management and Technology for Scholarship) at London Guildhall University, following a symposium at the University of Kentucky in October 1995. Its aim is to promote initiatives whereby humanities scholars and scientists, particularly computer scientists, can work more closely together in support of common research objectives.

Generally humanities computing has been seen as a poor relation of mainstream computer science, and humanities research has not been seen as providing computationally challenging problems suitable for investigation by computer scientists. As humanities computing has developed, it has become increasingly apparent that this is a false assumption, and that humanities computing presents issues which are as complex as those of scientific and technical applications.

This is particularly evident from the use made by humanities scholars of imaging. Researchers in disciplines such as medieval English literature have reacted enthusiastically to the possibilities of digital imaging of key texts. The images required by such disciplines need to be of very high resolution, and therefore pose issues in relation to the best ways of storing, transmitting, compressing and manipulating very large image files without compromising their evidential integrity. Moreover, humanities scholars are interested in new ways of exploiting these image files. They may want to tag and search components of the image or provide hyperlinks between sections of images. Since the questions asked by humanities researchers are different to those posed by other disciplines, new approaches are required to the applications required to supporting humanities research.

Many of the challenges presented by humanities research to computer, information and other scientists were outlined in the symposium "Reconnecting the Sciences and the Humanities through Digital Libraries", held at the University of Kentucky on October 19-21 1995. To take some random examples, humanities scholars working with artefacts of past civilisations are interested in better means of enhancing and investigating images of such evidence as manuscripts or paintings. Those scholars working with images would like improvements in monitor and display technology. Since the kind of information used in the humanities retains its validity over long time-scales, the need for more easily manipulated and reliable long-term storage of digital materials is particularly pressing. The problems and issues are manifold, and cover virtually every type of computational issue that can be imagined.

The traffic is not all one-way. Humanities research offers scientists working in computing much new material. The need for the availability of "live;" image data for work in compression and storage issues has been widely acknowledged by computer scientists. The kind of images being collected by humanities scholars offers plenty of scope for precisely this kind of research. Humanities computing, in other words, offers the possibility for a realignment of the relationship between the scientists and humanities, one which will draw them back together in a close and fruitful partnership.

This panel will explore these issues by presenting three major projects in humanities computing which have all benefitted from collaboration between humanities scholars and scientists working in the computing and information fields. These projects are as follows:

The Electronic Beowulf. This major project is being jointly developed by the British Library and the University of Kentucky. The editor of the project is Professor Kevin Kiernan of the University of Kentucky. Beowulf is known only from a single eleventh century manuscript which was damaged by fire in 1731. Transcriptions made in the late 18th century show that many hundreds of words and letters then visible along the charred edges subsequently crumbled away. To halt this process, each leaf was mounted in a paper frame which prevented further damage but concealed many letters.The core of the digital archive being developed by the Electronic Beowulf project comprises high resolution colour images of the original manuscript which can be magnified and enhanced to explore damaged and unclear sections of the text. A digital camera has also made it easier to use special lighting techniques to explore the manuscript more effectively. In particular, fibre optic backlighting allows the letters hidden by the paper frames to be recorded. The archive also includes early transcripts and editions which provide key evidence for the history of the text. Hyperlinks allow these transcripts to be more readily compared to the original manuscript. Computer scientists at the University of Kentucky have already started to explore ways in which the computer can assist in identifying particular scribal formations in the manuscript.

The York Doomsday Project. This project is based at the University of Lancaster and the University College of St Martin, Lancaster. Its directors are Professor Meg Twycross and Dr Pamela M. King. It is assembling a multimedia archive of materials connected with the York mystery play of Doomsday. It problematises the centrality of the script and the homogenising of other `records', proposing that the play existed only in its variant performances: the play itself is essentially ephemeral, while there are a number of discrete texts, from the York register to the street plan of 15th-century York, which bear witness to it. The exercise of amassing the archive has led to research findings which revise assumptions about aspects of pageant production in medieval York. In its second phase, the project will generate multi-media books which will make the archive available in a number of configurations. Primarily it will provide high-resolution images of manuscript materials, but users will also be inducted into the processes of decision-making which are part of reconstructing a pageant waggon, and will be able to experience a virtually real processional performance in the streets of medieval York.

The Excalibur Project. Excalibur PixTex/EFS is a product which uses Adaptive Pattern Recognition Technology to facilitate more accurate and efficient searching of text generated through OCR. The use of fuzzy matching allows highly accurate searching even when the OCR-generated text contains errors. Searches conducted using the PixTex/EFS search engine are therefore different to those in traditional text retrieval systems. The British Library has been working with the Department of History and Department of Information Studies at the University of Sheffield to evaluate this package. The project also hopes to build up a corpus of historical materials accessible and more easily searchable through PixTex/EFS.

The panel will briefly introduce each of those projects, explain how they have benefitted from collaboration with computer and information scientists, and discuss the prospects for further collaboration.

The chair of the panel will be Dr Seamus Ross, Assistant Secretary with responsibility for Information Technology at the British Academy. He is a Visiting Fellow in Humanities Computing and Multi-Media at London Guildhall University and is the author of many books and articles on humanities computing. A full cv is available on-line at http://britac3.britac.ac.uk/seamus. Postal Address: The British Academy, 20-21 Cornwall Terrace, London NW1 4QP, England. Tel. (+44) 171.487.5966. Fax: (+44) 171 224 3807. E-mail seamusr@britac.ac.uk

Speakers are
Kevin S. Kiernan is a Professor of English at the University of Kentucky. He is editor of, and an academic director of, the Electronic Beowulf. He is the leading authority on the history of the Beowulf manuscript, and the author of Beowulf and the Beowulf Manuscript (1981) and The Thorkelin Transcripts of Beowulf (1986). Postal address: Department of English, 1215 Patterson Office Tower, University of Kentucky, Lexington, Kentucky 40506-0027, United States of America. Tel. (+1) 606 257 6989. Fax: (+1) 606 323 1072. E- mail: kiernan@beowulf.engl.uky.edu

Pamela M. King is Head of the English Department at the University College of St Martin, Lancaster. She is co-Director of the York Doomsday project. Her publications include The York Mystery Plays: A Selection in Modern Spelling (1986), edited with Richard Beadle. Postal address: English Department, University College of St Martin, Bowerham Road, Lancaster LA1 3JD. Tel. (+44) 1524 63446. Fax (+44) 1524 68943. E-mail: xtg018@lancaster.ac.uk

Dr Mark Greengrass is Lecturer in the History Department at the University of Sheffield. He is one of the Directors of the Hartlib Project, a pioneering electronic edition of the papers of Samuel Hartlib. He is also involved in a number of other major historical computing projects, including the Strafford Papers project and the Excalibur project. Postal address: Department of History, University of Sheffield, Sheffield S10 2TN. Tel. (+44) 114 2824807. Fax: (+44) 114 278304. E-mail: m.greengrass@sheffield.ac.uk

Dr Andrew Prescott is a Curator in the Manuscript Collections at the British Library. He is also a Visiting Research Fellow in Multi-Media and Humanities Computing at London Guildhall University, and co-Director, with Seamus Ross, of the Centre of Information Management and Technology for Scholarship at London Guildhall University. Postal address: Department of Manuscripts, British Library, London WC1B 3DG. Tel. (+44) 171. 412.7509. Fax: (+44) 171 412 7745. E-mail: andrew.prescott@bl.uk