Powered by
JSPWiki v2.8.2
g2gp 17-01-2009
View PDF

Creating and Using Virtual Reality: a Guide for the Arts and Humanities #

Virtual Reality Case Study Library#

Case Study 6: Quest for Canterbury's lost Roman Temple and metadata case study#

Nick Ryan, University of Kent

Public presentations, whether on the Web, in museums or in the broadcast media, can be much more than a repackaging of earlier forms in a fashionable and spectacular medium. They do not need to be limited to telling a single story or to presenting one of many possible interpretations as an established 'truth'. Figure 20 shows an introductory frame from a multimedia presentation, 'Quest for Canterbury's lost Roman Temple', developed for Canterbury Museums. One of the purposes of this system is to show museum visitors how archaeologists can make significant inferences about the layout of a Roman town and the form of its buildings from very incomplete excavation evidence and a knowledge of comparable structures elsewhere. The presentation has been designed as the first part of a system to provide information about each of the main public building complexes in the city, although the current implementation concentrates on a postulated classical temple and its precinct. The other complexes (the forum, public baths and theatre) are shown only as still images with minimal descriptive text.

Quest for Canterbury

Figure 20: Quest for Canterbury's lost Roman Temple


The temple is thought to have been situated in a precinct adjacent to the forum and theatre complexes. A portico surrounding the precinct has been located in several excavation trenches, but there have been few opportunities to explore the enclosed area. Where this has been possible, evidence of an extensive courtyard surface has been found, often cut into by post-Roman structures. Materials that could come from a significant classical building have been found re-used or re-deposited in these later contexts within the precinct area, but the exact location of any temple remains unknown. However, there is anecdotal evidence to suggest that its foundations were observed, and possibly largely destroyed, during building work in the 1960s, before the establishment of a full-time excavation unit in the city.

The presentation runs on a touch-sensitive flat-panel device standing next to a display case containing finds from the area. These finds include a number of masonry fragments, mostly from one or more Corinthian columns, and a variety of marble and other stone wall-cladding materials. The presentation, therefore, seeks to breathe some life into these otherwise rather dull finds that might normally be labelled only with dry statements such as 'Masonry fragments, possibly from a Roman temple'.

Interaction is minimal. At a few points, the user can choose which building complex to visit, or make minor changes to their route through an otherwise fixed sequence of frames. Buttons are also provided to skip backwards or forwards if the user finds the default timing too fast or too slow. The latter is more likely because the default timing is aimed at the needs of younger and potentially slower readers.

Most of the display sequence uses still images and video fly-through or walk-through sequences of complex solid models of the city in its surrounding landscape (Figure 20) and individual buildings, together with textual annotation in a choice of three languages. Animations are used to overlay models on excavation plans and to place photographic images of the finds displayed in the adjacent case onto rendered solid model representations of Corinthian columns and capitals.

The simplicity of the system and the use of 'guided' or 'directed' sequences, rather than allowing users to roam freely around a virtual Roman Canterbury, was a deliberate design choice. Canterbury Roman Museum is small, yet it attracts large numbers of visitors and is popular with parties of children from local schools. The design of the museum itself follows a similar pattern, with visitors encouraged or constrained to follow particular routes through its various rooms. The undeniable attraction of interactive displays, particularly to younger visitors, led to the requirement that the maximum duration of the display should be tightly constrained to prevent queues forming.

At an early stage of the design process, VRML was considered as a means of implementing the three-dimensional models, and simple prototypes of the temple and theatre were shown at various 'open days' to give an impression of what future displays at the museum might look like. In the end, the need for directed sequences and the relatively poor image quality of most VRML renderers led to the decision to prepare all images and video sequences using a solid modelling and ray-traced rendering solution. Vue d'Esprit by e-on software was used both as a modelling and rendering tool.

Source data used in the model included a DEM derived from out-of-copyright map data, and known building plans which were derived from both published and unpublished excavations undertaken by Canterbury Archaeological Trust and other groups. Many building fragments were initially drawn using various CAD programs and exported as DXF or DWG files, whilst others came from earlier projects using other modelling and rendering tools such as POV-Ray. The remainder of the model was developed within Vue d'Esprit.

Despite the availability of several more sophisticated multimedia authoring packages, this system was developed using Microsoft PowerPoint, a medium more often associated with conference presentations and lecture slides. Apart from the author's familiarity with this package, this choice was made because of PowerPoint's simplicity and adequacy for the relatively simple requirements of the display system. Indeed, the system could have been realised using almost any of the authoring packages currently available.

One of the benefits of the early VRML-based prototypes was that they were intended to be viewed using an HTML browser, with an appropriate plug-in, and so might also be published on the Web. Although the system installed in the museum employed a conventional multimedia approach, the possibility of Web-based delivery was not wholly abandoned. The same content has been used to produce several, albeit incomplete and unpublished, versions of the system as test-beds for more recent approaches. One of these is discussed later.

The image in Figure 21 is a frame from the beginning of an initial video sequence that flies in from an aerial view to land in the centre of the town. The 3-D model of Roman Canterbury and its surrounding landscape is based, where possible, on archaeological and environmental evidence. However, in a model of this size, there are large areas for which no such evidence is available and in these the model elements are no more than informed speculation. Even where evidence from archaeological excavations is available, it is invariably partial. Excavations rarely recover complete building plans and structures rarely survive above floor level and are often represented only by robber trenches.

Video frame from the quest for Canterbury

Figure 21: Video frame from the quest for Canterbury's lost Roman Temple


The model represents a composite of much of what is known of Canterbury during the first four centuries AD. It is intended to convey an impression of the town's appearance during much of this period, rather than to be a strictly accurate model of any particular date. The date of 300 in the title was chosen as a compromise to allow the late 3rd-century walls to be included, even though the building complexes that are the main focus of the presentation are mostly of 2nd-century origin. Indeed, some earlier buildings in the model may have gone out of use and have been demolished or replaced by later structures by this date. Most of the circuit of Canterbury's medieval walls remains standing today, although large sections are more recent rebuilds, and these are founded on the earlier Roman walls. These walls are, therefore, an important local landmark and their inclusion in the model helps viewers to orient themselves.

This model and the system by which it is presented to museum visitors are quite typical of current uses of Virtual Archaeology for public presentation. The presentation is a little unusual in that it seeks to convey an understanding of the incomplete nature of the archaeological evidence and the resulting uncertainty inherent in its interpretation. Other than this, however, it offers much the same benefits and suffers many of the same limitations as most similar systems in current use.

Table 9: AHDS Core metadata for Canterbury Museums VR presentation

Information typeScope note
TitleQuest for Canterbury's lost Roman temple
CreatorNick Ryan, University of Kent http://www.cs.ukc.ac.uk/people/staff/nsr/index.html
Subject.disciplineArchaeology
Subject.typeTemple
Subject.periodRoman
Subject.typeExcavation
Subject.typeReconstruction
DescriptionA multimedia presentation for Canterbury Museums
PublisherCanterbury Museums http://www.cs.ukc.ac.uk
DepositorNick Ryan, University of Kent
Date20 October 1998 - 21 February 2001
TypeInteractive resource
Formatapplication/vnd.ms-powerpoint
IdentifierNot given
SourcePlans and excavation data
LanguageEnglish
Relation.archivePlans and excavation data, Canterbury Museums 1980-2001
Relation.Has partVideo clip - Aerial fly in
Relation.referenced byhttp://www.cs.ukc.ac.uk/people/staff/nsr/va/des/fly_in.html
CoverageCanterbury
Coverage.administrative areaKent
Coverage.countryEngland
Rights.copyrightMuseum display: Canterbury Museums 2001
Rights.copyrightPlans and excavation data: Canterbury Museums 1980-2001
Rights.copyrightComputer model: Nick Ryan 1998-2001
AudienceMuseum visitors
MediatorCanterbury Museums
Education StandardThis resource does not conform to an established education or training standard
Interactivity TypePoint and click presentation
Interactivity LevelLow
Typical Learning Time30 minutes