First is the need to distinguish between work that aims to represent artifacts, buildings, or sites and work that aims to document artifacts, buildings, or sites. The two aims are not incompatible, nor are they mutually exclusive. However, they are different enough that one must consciously aim for one or the other (rarely both).
|Images courtesy SimTeam and AGP.|
It's important for the original scholar to provide data in forms that permit others to query the model for dimensions or otherwise investigate it on their own-using only standard software. In fact, I would argue that the best work begins with documentation that will retain its accuracy and scholarly utility while providing a base for a more visually appealing representation. It seemed to me that this distinction should have been more strongly stressed in the article. (Of course, as an archaeologist, I would naturally think that.)
Second, I think more emphasis should have been put on the inadequacy of even the best model to represent the real item. When we build our dimensionally accurate AutoCAD model of the entrance building to the Athenian Acropolis (see www.propylaea.org), we may have every dimension right, but even so, we will have but a pale imitation of the original. I think we need to be careful to avoid suggesting that computer models can stand in for the real things. Not only does such a view suggest that there are no important aspects of an artifact or a building other than the visual, but it could lead to the frightening conclusion that objects need not be saved if computer models of them exist.
Finally, there are serious problems with these kinds of models themselves. They are never fully accurate. Even good models may show finished buildings when only partial re mains exist or painted items where only the unfinished ones survive. My point is that there is always the danger of replacing reality with our version of that reality without making sure that the audience knows the difference. (I make this argument in
an Internet Archaeology article "The Compelling Computer Image: A Double-edged Sword" (http://intarch.ac.uk/journal/issue8/eiteljorg_toc.html).
If we show people virtual reconstructions, we must first make clear what we really know and what we don't. We must resist the temptation to use technology rather than evidence to convince others that our reconstruction is the correct one. Otherwise, we will almost certainly misinform them.Harrison Eiteljorg, II
Director, Center for the Study of Architecture
We welcome any insights you have to offer that would further our readers' understanding of topics discussed in this issue, or that concern the computer graphics industry in general. We may edit your comments to conform to our style and space requirements. Please address letters to:
Letters Editor, Computer Graphics World
98 Spit Brook Road
Nashua, NH 03062-5737
Due to a production error, credits did not appear alongside the following images in the story "Preserving the Past" on pg. 24 of the September 2001 issue.
Image copyright2001 Learning Sites, Inc.
Image copyright2001 Powerhouse Museum.
Photo courtesy Alonzo Addison.
Images courtesy Bob Stone.
The images on pgs. 29 and 32 of the October 2001 story "Seeing with the Mind's Eye" should also be credited to Penny Rheingan, University of Maryland, Baltimore County, who worked with David Ebert to develop the perception-based software.