AGOCG logo
Graphics Multimedia VR Visualisation Contents
Training Reports Workshops Briefings Index
This report is also available as an Acrobat file.
Back Next Contents
The Design of Virtual Environments with particular reference to VRML

The uses of spatiality

What can we use space to mean? What do we use space for in real life? Space can be physical; it can also be metaphorical. We should not omit the uses of 2D space, which may also help us in designing in 3D.
For example:

The use of fully dimensional space is one of the defining characteristics of VR. So far discussion of three-dimensional objects and worlds in VR has tended to look at the technology - what is possible - and the application - what is demanded. However there is an entire layer missing from the discussion, a layer between the technology and the application, which should deal with the qualities of spatiality itself. In other words: 'what can we use Space for?'

We can begin to answer this question by looking at how space has been used in other media and other contexts, rather than just within VR. As with many applications of computing, there is benefit in intelligently applying critical ideas from other media and other technologies, rather than looking only in our own backyard.

Do we know what space is?

Readers might be impatient at being asked what space is. After all, does it matter? Must we waste time on definitions? We suggest that some awareness of the debate about definition is needed, especially in the context of transcultural software development.
It may be that all human beings have the same perception of space at the biological level of perception. But certainly every society uses its space differently, both technologically and artistically.
J David Bolter 1986 Turing's Man p80

Bolter's may be is important. We should look briefly at the question of biological perception, before considering his second point.

The two opposing schools of thought can be briefly characterised as follows:

This is the nature/nurture controversy in one of its many guises: do we perceive space according to universal optical and perceptual principles on which social and environmental conditioning has no effect; or do we acquire some aspects of our perceptual system from external influences? It should be emphasised that we are leaving aside for now the question of its representation. If the very perception of space is culturally determined, then a new set of variables enters the equation.

The first widely acknowledged suggestion that the perception of space might be culturally determined arises in the work of Edward Sapir and Benjamin Whorf. Their work and that of their followers has acquired a remarkable popularity, becoming in some circles an orthodoxy in its own right. A favourable view of Whorf's work is found in a paper prepared for AGOCG (Hopgood 1993) which was prompted by a debate on standards for specifying time and space in the PREMO standard.

The key argument of the Sapir-Whorf hypothesis is that perception of space is determined by culture, and particularly by language. Whorf (1897-1941) was amongst other things an amateur linguist, who studied several languages including Hebrew, Aztec, Mayan, Hopi and Chinese. His view (in the words of Hopgood) is that:

Europeans have a notion of time and space that is generally assumed by them to be universal. This gratuitous assumption is naive, arrogant and wrong.
Hopgood 1993, p3

Hopgood is confident that the Hopi language contains no reference to time either explicit or implicit (ibid, p3). His argument in relating Whorf's theories to the development of standards for multimedia, is that, since no language system (or concept-system, in the view of this school of thought the same thing) can claim to apprehend space or time with greater truth than another, our reason for preferring one to another should be based on functional grounds. We should choose a model of space-time to serve a purpose, rather than trying to assert that it is somehow true. The criterion should be: does it work? For example, Newtonian physics is useful for prediction of everyday cause and effect, but unhelpful under special conditions such as high speeds or sub-atomic scales. Under these conditions we have to pick another model.

Transformative technologies

Most people, even where they disagree with the Sapir-Whorf hypothesis that language and culture influence perception itself, would endorse the idea that forms of communication are transformative. In other words, what can be said and thought is influenced by the means selected (or available) for expressing or representing it. The important consequence would be a recognition that Virtual Realities will themselves facilitate the communication of some kinds of information and experience, and militate against others.

To the Victorians, the idea that media are not transparent to the ideas they convey would have been unfamiliar. The whole period since the Renaissance has seen a largely unspoken belief that media are an unproblematic window on reality. This belief has been largely superseded during the twentieth century by that of transformative technologies. Almost notorious is Marshall McLuhan's dictum, 'the medium is the message'. Postman (1982, 1985) has made extensive use of a McLuhanite approach in his criticism of television and other predominantly visual media as against the merits of the word. Ong devotes all of Orality and Literacy (1982, p155) to the transformative effect of writing and printing on the world of ideas, and argues explicitly that 'These what we know in ways which make it quite inaccessible and indeed unthinkable in an oral culture,' (our emphasis).

Pinker (1994) aligns himself strongly with the Nature camp against the Nurture-based view, and adamantly refutes the Whorfian hypothesis. He does so both on the grounds that Whorf and his followers gathered faulty evidence, and that they argued falsely from these premises. It may be that there is no possibility of deciding between the opposed theoretical positions. However, in the context of harnessing virtual realities for particular purposes, it would seem that the functional view advocated by Hopgood is a productive one. The implication is, that we should represent in our virtual realities only those aspects of reality which serve our purpose at the time. This would suppose at the very least a selectivity in the attributes of reality which we might describe, and perhaps the deliberate construction of attributes which are 'unrealistic' but useful. It would also imply Virtual Object and Environment data which can be differently filtered and represented on an as-needed basis, rather than which have a fixed form.

Back Next Contents

Graphics     Multimedia      Virtual Environments      Visualisation      Contents