Internet Archaaeol. 6. Edmonds and McElearney. Inhabitation and Access

Visualising the landscape

One of the consequences of this shift of focus has been a sharper concern with the form of archaeological texts, in particular with the possibilities opened up by experiments with writing. Such a concern is wholly justified, but it could be argued that the potentials offered by visual media are equally important. In a discipline that, on the page at least, remains resolutely `plan based', the scope offered by the Internet is substantial (cf. Pollard and Gillings 1998).

As a first step towards exploring these potentials, we have been experimenting with QuickTime VR. This technology situates the visitor at the centre of a digital photographic panorama, wherein they can pan their field of view (left to right, up and down), and zoom in or out of the image. Using these panoramas, we have begun to explore how movement within images might provide a more satisfactory basis for conveying a sense of the topography, setting and character of much of the archaeology. Links between panoramas can also be established, allowing the site visitor to move, albeit at this stage in a series of jumps, from one location to another.

Creating panoramas on the website has enabled us to provide a photo-realistic and (limited) interactive impression of what can be encountered on Gardom's Edge today. These have been useful. But the panoramas have focussed on significant archaeological features in the area, and although they can be linked together, the participant is not so much situated at, but anchored to, a visual experience from each panoramic `node'. The links from node to node convey some sense of movement, but these were arbitrarily set by us, and relate largely to our initial desire to link sites that we deemed important, or vistas that we felt were striking. Incoherent in chronological terms at least, these journeys give a taste of the place and of the extraordinary palimpsest of material. What they don't, as yet, allow, is the possibility of exploring the area and its development over time, in a variety of ways. They offer no effective sense of the changing character of this landscape, nor any real freedom of navigation that would touch on the experience of place.

As one way of addressing this problem, we are beginning to experiment with the use of GIS to model patterns of movement and activity at different spatial scales, our intention being that the project database will form the basis for these models. Given world enough and time, it should be possible to move in and around particular buildings, to map the changing configuration and pattern of the landscape, and to take a range of different paths. We're also exploring other avenues. Landscape generation packages such as VistaPro allow the creation of highly detailed landscape images from simple elevation data, using a series of complex fractal algorithms. Because they offer the ability to alter parameters such as vegetation, cloud cover and haze, it is possible to create many different representations of the same `vista'. We are therefore able to experience a series of `fuzzy viewsheds', which are arguably far more humanised than a series of irregular polygons overlaid on a digital map (Pollard and Gillings 1998). VistaPro also allows the generation of fly-throughs (albeit along pre-ordained paths) and because of the way that the software renders the images, these allow us to follow patterns of movement and activity at a number of different scales - from the house to the enclosure, from field-system to cairn, or across the gritstone moors and the limestone plateau beyond.

With the flexibility that multimedia technologies allow, we should also be able to explore how specific forms of movement, and the occupation of particular places, might have been caught up in social life. How did the form and setting of a chambered tomb or the banks of a henge encourage certain patterns or structures to proceedings? Where might we see rock carvings from, and from where could they be approached? Any attempt to broach these questions requires a considerable degree of flexibility in navigation through virtual monuments, and for this, we are starting to experiment with Virtual Reality Modelling Language (VRML). VRML enables the creation of three-dimensional spaces and objects that can be viewed and interacted with via the web, often referred to as `scenes'. As well as providing free movement through these scenes, they enable animation of components, and the spatial location and movement of sound. Of all these techniques, VRML appears to offer the most potential, though it also requires the greatest degree of preparation.

Our hope is that these experiments move us towards developing a visual medium for portraying and exploring some of the varied experiences of place that were possible in the area at different points in the past. We do not, however, imagine that our experiments will ever `bridge the gap'. After all, we are dealing with electronic media that necessarily create particular forms of distance and particular forms of gaze, however much they are allowed to wander (Berger 1972). In many cases, the metaphor of the map is still prominent.