Networked Moving Images
Royal Scottish Academy of Music and Drama
100 Renfrew Street
Glasgow, G2 3DB
This paper describes and demonstrates the work of the Performing Arts Data Service (PADS) in piloting the networked delivery of moving image resources for use in teaching film and TV studies and social history at the University of Glasgow. Streaming high quality moving images poses many technical questions, and we outline our solution which has proved robust within the limited testing arena of the project. The main focus of the paper, however, is on user issues, covering such areas as the choices we made for the design of the user interface, what we have learned from our user trials and evaluations, and what effects networked access to a library of moving image material might have on pedagogic and research practices in the humanities.
Overview of the project
A University of Glasgow consortium led by the PADS was chosen at the end of 1997 as one of two pilot sites for a new project to develop the delivery of moving images to UK academic institutions via networks. The short-term aim of the project, which was initiated by the British Universities Film & Video Council (BUFVC) in partnership with the British Film Institute and the Joint Information Systems Information Committee (JISC) is to test both the technical and pedagogical viability of networked delivery of moving pictures to universities. Over the longer term, it may be the opening activity in a radical new network service for UK higher education.
The collections of digitised moving images and related textual resources are mounted on servers for access by lecturers and students via a secure web gateway, with moving images streamed to individual workstations, computer labs and lecture rooms in the university. Of particular interest to film and TV studies academics are the collections of resources on Hitchcock's Blackmail (including substantial extracts from both the sound and silent versions of the film for comparison, a BBC documentary and a number of related textual resources) and a collection of archive British TV of the 1950s and 60s.
Once cleared for copyright via the BFI, the moving image material was transferred onto Beta SP for digitisation by the Manchester Visualisation Centre at the University of Manchester. From there it was forwarded to the client sites. At Glasgow we have concentrated on MPEG1 format, but we are also trailing MPEG2 format for some selected content after consultation with academics on the desirability of high quality pictures. We have focussed on campus-wide delivery, comparing two methods of storing and managing data: from the PADS SGI servers (using MediaBase and HyperWave) via fast ethernet and ATM and from the SUN servers of our partners in the Revelation project at the Department of Computing Science.
Using the collections
HyperWave, the PADS information gateway system enables users to employ a number of searching strategies (from simple searches on full text content or specific metadata attributes to more specialised filters) and to have results displayed in different ways, including graphical representations of the structure of the information. Users simply click on hyperlinks in the search results to play the movies via the MediaBase player. If browsing, again, there are a number of choices; we have found that users value the opportunity for browsing, an opportunity often denied in digital library environments. In browsing mode, users are given a subset of the Dublin Core metadata attributes to help them make their choice (including a brief description, date and link to related resources). They can also choose to access the full descriptive metadata for each item.
We have carried out evaluations of the use of this material in contrasting academic situations. For example, the Hitchcock materials have been used as part of a junior undergraduate lecture series, with materials made available to students to follow up in their own time in a student lab; we have also evaluated how senior undergraduate students majoring in film and TV studies use the material for personal research, and we have gathered data from academic colleagues in other universities on the utility of this method of delivery and its implications for teaching and scholarship. There is a striking attitudinal difference between faculty and students: students regard this innovative method of delivery with some nonchalance (this is no more then they have come to expect from their increasingly digitally-based study environment), meanwhile faculty, having experienced the difficulties of locating and using moving image materials in traditional formats for teaching or research, are much more forthcoming about the potential benefits. Priority areas for further development work are better indexing and easy-to-use tools to manipulate content; on the wider scale, rights management needs to be addressed urgently and a strategy formulated for digitisation and access priorities.
In comparison with traditional methods of access to moving image materials, networked delivery has clear advantages, including ease of access for larger numbers of users than is traditionally possible, and facilitation of access to a wide range of supporting materials alongside the moving images. Also some disciplines which have hitherto underused the rich variety of moving image resources available might be encouraged to take better advantage of them. However, although the technical setup necessary to achieve networked delivery of large amounts of moving image data is no longer at the "cutting edge" it is still unavailable to those without a fairly sophisticated technical architecture. This, coupled with other obstacles such as the thorny issue of rights management, make the future widespread availability of this kind of service still uncertain. At the end of the project we believe we will have demonstrated significant pedagogic benefit and a practical technical setup with which to achieve it, as a useful contribution to the debate about future directions.