Graphics and Archaeology: Interpreting the Past
Snowbird, Utah
20-23 May 2000
Presentation:
- Open Problems for Archaeological Visualization
Presenters:
- Peter Shirley
Presentation Resources:
Summary:
Peter's primary interest in this Campfire was learning more from the archaeologists about future applications of visualization technologies. In a brief search of the web, he characterized some of the current approaches of archaeological visualization into three categories:
- scanning and survey data (e.g., remote imaging)
- hand-done reconstructions of 3D (e.g., object abd site modeling)
- resource-intensive reconstructions (e.g., the movie Titanic)
He gave an overview of various research technologies that required less intervention
- procedural rendering
- geospecific rendering (e.g., augmentation of USGS data, computer vision techniques)
- physics-based rendering
- simulated plants/animals
- simulated humans
- immersive environments (e.g., walk-thrus of reconstructed sites)
Other emerging technologies discussed included scanning (non-procedural), image-based rendering, visualization, animation, and physically-based sound. He showed a photo of a Greek amphitheater with human visitors and set that as a level of desired realism to converge upon.
Further in the presentation, he offered a hypothetical visualization system as a future target for graphics and archaeological visualization scientists to consider. It included the ability to dial in movement in time and space. It was populated by simulated humans, perhaps even with behaviors (mentioned MetaCreations' Poser). It had open data entry and could fill in procedural data wherever needed. A children's book, "A Street Through Time", gave an interesting analogy to these ideas, although there were still many issues of user interface (do you keep the time of day fixed in a year?)
He concluded that there were still big issues about navigation, the notion of science versus entertainment, and socio-technologic optimizations that needed to occur. Technologically, many things were possible now, but the data acquisition and organization was the most daunting. (Peter considered students as possible contributors to world archaeological databases, given the right tools to acquire and contribute data to an self-organizing database system.)
Conclusions / Issues:
- There is a huge need for the acquisition and organization of data to make such a system work. Without some entertainment value, we won't be able to attract those millions of possible data-entry people.
- Who owns and controls the data of any one reconstructed site?
- Example - Moab, UT - what if they do a virtual version of the petroglyphs? The municipality says we own the site (and the virtual site). If someone else does a better job, how could that individual make their work visible if the local owner controls the "domain name" of that project? Could you access the better data (and still not lose access to the original domain)?
- Can a funding model be created for such a public project?
- There is a need to develop a simple, no-brain-dead implementation of sharing and redistribution beyond HTML
- How would we blend different sources of data (perhaps from different database sites of the same region with your "time" knob?
- The system needs to be simple - needs more global and public buy-in
- What is this system like? A browser on the Internet? A CD-ROM with links to the Internet?
- We shouldn't lose access to multiple interpretations - the results
are richer.
- An Open Source implementation (a la Linux) might be desirable
- Consider simulations with people (a la Sim City), where populating sites with parameters for plausible behaviors - this might help exercise conjectures about a site's purpose.
- Digitizers of the future - high school students - there is a need to
harness the huge population of interested individuals - look at the popularity
of magazines like Archaeology.
- The potential here is much like the same potential unleashed by amateur astronomers - they contribute, too, to the science of the field.