When virtual reality was introduced into archaeology in the early 1990s, some of the concerns about its usefulness were precisely the same reservations that had been expressed regarding the use of photography in archaeology in the mid-19th century (see 19th c. vs. 21st c. archaeology methods at the left; hover over to enlarge). For example: that it is cumbersome and difficult to learn; few people in the field know how to use it effectively; it is expensive to create useful results; and the images cannot be trusted by the archaeological community. But as photography has become a standard, indispensable tool for the discipline, so VR is now becoming.

By the 1870s, archaeologists were beginning to see the benefits of photography for recording sites, excavations, and artifacts with a level of accuracy and realism not previously possible, and making these images available to a larger audience in an accessible, reproducible form. Since the early 1990s, VR has offered the same benefits for archaeology. And yet much more--VR can tailor visualizations to different end-users, reconstruct elements now lost, and link vast amounts of supporting documentation to the interactive visuals. Clearly, the benefits of VR for archaeology leap far beyond those of any technology previously available.

Over the last decade, much has been written about the increasing use of VR in archaeology and heritage projects, from museums to visitor centers to university computing centers to online re-creations and publications (see samples of VR environments at the left; hover over to enlarge). VR became such an integral part of cultural history projects that a new field called Virtual Heritage was born, and the field continues to attract new practitioners, new approaches, and new technologies. Virtual Heritage has become popular in part because the digital dissemination of information about the world's history in vivid visual form has fed an information-hungry, digitally literate world. But just as important, the excitement for the professional and student emerges from navigating 3D models that provide striking, sometimes quite surprising, reassessments of evidence.

I speak to you today as the president of two Virtual Heritage organizations: Learning Sites, founded in 1993, creates virtual ancient reconstructions with the intent of making them available to scholars, educators, and students. And, the Institute for the Visualization of History, founded in 2001, expands beyond ancient eras to other periods of history, and provides still more accessibility to the vast potential of VR for teaching, research and display.

This quick snapshot (see the timeline at the left; hover over to enlarge) serves to emphasize the rapid pace of expansion in the use of VR in archaeology over a decade--1993/94 saw the first virtual archaeology worlds, but there were just a handful in the US and England, including one of ours, the fortress at Buhen; back then virtual worlds ran only on high-end systems with graphics cards the size of small refrigerators that cost $250,000 or more. In 1995, VRML was introduced for running interactive 3D content on the Internet; and the first annual Virtual Heritage conference was held, in Bath, England (where we showed our Buhen project). In 1998, the Computer Applications in Archaeology conference was held in Barcelona and featured a special session on virtual archaeology, resulting in Learning Sites' sponsorship of the Virtual Worlds in Archaeology Initiative, and the book Virtual Reality in Archaeology, edited by Juan Barcelo, Maurizio Forte, and myself; also in 1998, in Japan, the first Virtual Systems and Multimedia workshop in virtual heritage was held, co-sponsored by UNESCO. In 2000, the Virtual Heritage Network debuted on the Web, with its intelligent portal to news and information about the discipline. Now, in 2005, graphics cards in laptops are hundreds of times faster and have more memory than the originals that were thousands of times the price; and this year alone, there are 13 international conferences on VH, and hundreds of VH projects worldwide. In the meantime, we at Learning Sites and the Institute have worked on 31 archaeological and historic sites, across 10 countries, covering more than 60 buildings and other structures.

We have come a long way from the initial enthusiasm of just being able to create virtual worlds, to the integration of virtual reconstructions into public school education, museum displays, archaeological site visitor centers, digital publications, and television. Yet, still today, the process of building a virtual re-creation of an historical place or event demands skill, patience, and hard work. Using as examples several projects developed under my guidance at Learning Sites and the Institute, I'll discuss what works and what doesn't work in the creation of virtual worlds for cultural heritage. I'll conclude with some words about the future of interactive visualizations (hover over the image at the left to see the topics).

Why do VH in the first place?

Creating images intended to reproduce visual experience has been part of human activity for millennia. From its earliest known images in cave paintings, humankind has exhibited an innate drive to represent the world around it (hover over the image to en. In addition to visual images, theater and literature have all offered views of other places, other experiences, other beliefs, and other times, to stimulate the imagination, to instill wonder about the fantastic, and to speculate about the spiritual. As technologies have changed, so have the means for stimulating the eye and the imagination, from mathematical perspective in the 15th century to motion pictures in the 20th century. People are fascinated by the illusion of being there.

Representation has also been used throughout human history to educate and entertain. Visualization is a recognized means of making data and concepts more easily comprehended and assimilated. Thus, instructional books are illustrated, and audiovisual materials are widely used in classrooms. And now along come computers, interactive 3D digital visualizations. Virtual reality is just another step in the age-old drive toward heightened visual experience based on reality.

A couple of years ago, historian David Staley [then Executive Director of the American Association for History and Computing] wrote that computer visualization when used "to represent simultaneity, multidimensionality, pattern and nonlinearity with…speed and efficiency" can do what "prose cannot capture." In his book Computers, Visualization, and History, Staley argues that the real impact of the computer has been as a graphics tool more than as a processor of words. The importance of 3D imagery lies in its ability to address longstanding concerns of historians who agree with the 19th-century writer, philosopher, and social reformer, Thomas Carlyle's observation that "Narrative is linear, Action is solid." Thus, the technical potential of computer graphics is that it can present a deeper and more richly rewarding history by giving a 3D solidity to past places and events.

With specific application to our discipline, VR can address a number of the problems that have plagued archaeology since its inception in the late 18th century (see the list at left; hover over to enlarge):

• problems of testing complex diachronic hypotheses,
• difficulties of accurate and precise data gathering,
• cost and space constraints of publishing images,
• the exasperating time lag between excavation and publication,
• traditional passive teaching methods can become participatory problem-solving-based distance education; and
• the overarching dilemma of how to envision and communicate to others a past that happened in more than just 2 dimensions and in vibrant color.

Other major benefits include:

• permitting remote access to sites that are difficult or impossible to visit;
• virtually re-placing artifacts back into their original contexts for fuller appreciation of their original functions; and
• preserving digital surrogates of cultural heritage monuments that are being lost or damaged through wars, natural disasters, natural deterioration, or heavy visitation.

Concerns over the deterioration of India's cultural heritage were first voiced in the 1870s, and, beginning in 1902 (first with excavations at major Buddhist sites and soon after with the discovery of Harappan sites), it was photography that was promoted as the documentary medium that could provide many of these same benefits--preservation, conservation, education, archiving, and remote access.

Over the years what has worked in VH?

The application of diverse new technologies has broadened the ways in which VH can integrate data from existing remains into virtual re-creations (see an example of a Learning Sites virtual world at the left; hover over to enlarge). VRML continues to be the output format of choice, despite the lack of interoperable VRML viewers (some don't have Java support, some don't work on Apple computers, some don't load well into the newest Web browsers, and there is no consistent adherence to the VRML 2.0 specifications). Yet VRML prevails, despite the apparent potential of X3D, Java3D, and numerous proprietary corporate interactive 3D visualization tools (such as those from Kaon, Adobe, Virtools, MindAvenue, Demicron, Macromedia, and those that come integrated with higher-end CAD packages).

What else has worked?

Well, the "wow" factor is still a big winner--when archaeologists experience virtual re-creations, they tend to be impressed, overwhelmed, and excited. The sheer visual impact can be stunning. In addition, when we work with researchers to construct virtual worlds from their data, they are forced to pay close attention to accuracy and precision, which leads to increased credibility in the results and more trust in the technology to deliver reliable re-creations (see for example the virtual reality model of the Athlit ram, a Roman ramming prow, at the left; use your left mouse button and mouse or touchpad to rotate and zoom). Researchers and students alike can explore new theories and test new situations impossible in any other media, such as about lighting, circulation, spatial relationships, and diachronic site development. The use of virtual reality for education has had proven advantages, because classwork becomes engaging, participatory, and exciting.

Unfortunately, the news isn't all good

There are some annoying drawbacks--annoying not so much because they cannot be overcome, but because they haven't yet been satisfactorily resolved, even after 10 years:

• there is a lack of standards, in browser support for VR s/ware; in VRML viewers; in a growing multitude of interactive 3D formats; in variations in display hardware that call for specialized output; and in archival methods for digital media (too many locally developed VH databases and worlds are not universally and globally accessible);
• the general public's expectations for all computer graphics has been dramatically raised by the modeling and rendering in newer animated movies and 3D TV shows, even though we understand that our real-time VR cannot yet match pre-rendered scenes;
• the precision of digital data capture devices far exceeds the level of detail or texture map sizes allowable by interactive 3D software; what do we do with all the extra pixels?; and
• the field is not taking advantage of the virtual world as a visual index for a fully digital site report, a completely new kind of research resource that recognizes the distinct value of interactive media over traditional static, codex-based publications and teaching tools.

Many of these are merely technical difficulties, but there are also some longer-term challenges to the discipline-wide acceptance of virtual heritage visualizations (the image at the left contains one of Learning Sites 3D model reconstructions of the excavations at Nimrud based on a single photograph; hover over to enlarge):

• how do we communicate the different types of data that informed our modeling: old vs. new excavation data; revised versions of our reconstructions; what is based on explicit evidence and what is hypothetical; and how do we integrate these varying datatypes into an easily accessible and easily navigable virtual environment?
• how are we to deal with differing output formats, different degrees of public access to broadband, differing levels of education among the viewers, and differing display hardware, when we are called upon to design for museums, schools, the Web, stand-alone kiosks, onsite visitor centers, digital publications, broadcast media, or movies?
• what do we do about the long-term survival of our work due to the ephemeral nature of digital data? Digital information is not like a cuneiform tablet; is our data too dependent on a technological infrastructure to be viable in the long run as an information medium? We have heard these concerns raised by users; is that holding people back from using VR?

These issues are being addressed by an the increasing number of international heritage collaborations; we look forward to continuing that trend as a result of this workshop. The upsurge has been especially noticeable among consortia forming under various European Union frameworks for virtual heritage, making progress on standards, archiving, open-source programming, data storage, augmented reality, remote museum robots, wireless heritage networks, and immerisve educational programs; such as, ARCHEOS, Archeoguide, EPOCH, Tourbot, LifePLUS, and DigiCULT.

By reviewing some of the projects Learning Sites and the Institute for the Visualization of History have been developing, we can examine some specific examples of VR's value for studying, teaching, and re-creating moments of history. The very process of creating virtual ancient worlds usually provokes novel and interesting questions. For example, the need to position light sources inside a 3D model immediately leads to questions about ancient sources of illumination, as for example:

[ed. note: at this point in the presentation several VRML-based virtual worlds were demonstrated; these can no longer be viewed in current browsers]:

Nimrud: new insight includes about the Northwest Palace's circulation, lighting, interior viewsheds, how the king saw the throne room, scale, and color; VR allows researchers to study the globally dispersed reliefs in context, the nature of the full ancient sculptural program, and accurately see the material in context.

Battle Monument at Actium: new insight includes about ancient shipbuilding techniques, the original purpose of the sockets, the size of Roman rams and thus their ships and thus the ancient fleet used in the battle; researchers can now accurately compare ancient textual sources to evidence of the actual fleet composition.

Kyrenia shipwreck: new insight includes about how VR can be used to re-sinked ancient vessels, to help determine the precise manner in which the ancient sailors stacked cargo in their ships, which is currently unknown, but since amphora wrecks are common and the pottery remains and the ships do not, VR is the only way to re-create the original vessel. Then, in reverse, can we come up with patterns in the cargo distribution that can help determine something about the ships?

The Acropolis: new insight allows researchers to envision a whole new Acropolis and test theories about the survival and design of its major monuments, examine new spatial relationships without the need for photorealism in the virtual worlds.

Tantura Harbor shipwrecks: new insight includes providing new tools to underwater archaeologists for the visualization, publication, and interpretation of their data. Underwater archaeologists have quite different excavation circumstances that present problems quite distinct from dirt archeology, such as inability to go back to the site at will later, difficulty obtaining good color imagery due to the water, particles in water, and absence of natural light, and difficulty appreciating the entire site from a distance.

Learning Sites javapanel: new innovative in-house tools for dealing with scale in a virtual world, to show distinctions between what has been excavated and what has been digitally reconstructed, and to access external databases for interactive research from within virtual worlds; we are also working on building tools for adding digital footnotes to virtual worlds.

Before looking ahead, let's examine a series of questions posed to the speakers at the first Virtual Heritage conference held in Bath, England, almost exactly 10 years ago to the day. Our responses show significant progress:

Virtual Heritage--is it authentic or fantastic?

Our approach, and that of most VH practitioners, has been to be as accurate and precise as the evidence permits, and only re-create the rest based on scholarly and studious investigation. When possible, we link virtual worlds to external databases holding the images and text that provided the evidence used to build the world. The goals of VH are education and preservation, not video games. One question remains, however: do we need standards for judging the integrity and accuracy of our creations?

Virtual Heritage--will access be local or remote?

The scale of current distribution formats and modes was not envisioned 10 years ago. The Web, wireless nodes, GPS-based locational devices, RFID tags have all emerged as viable remote access tools for virtual heritage data among a global audience.

Virtual Heritage--will viewing be passive or interactive?

Both can be entertaining and educational. The participatory VR experience and the passive animated flyover each has its benefits, depending on the audience, the goals, and the timeframe. The near-firsthand experience of interactive 3D environments, however, is unsurpassed for engagement and sensory immersion.

Virtual Heritage--will it be inside your head or outside your body?

Immersive VR using HMDs is mostly a thing of the past, though CAVEs, Immersadesks, and reality centers or vision domes are now in more widespread use and are dropping in price. VR projected or on a computer screen is now more widely available and affordable than 10 years ago.

Virtual Heritage--will it be seen only on supercomputers or also on PCs?

Again, we have progressed. Most of the VH projects created today are on PCs, while 10 years ago all were on supercomputers or needed very high-end, expensive systems. We now see--and prefer the benefits of--making our worlds available to a broad public.

So, where is VH headed now? In the hope of serving the needs of the Virtual Heritage community and also making Virtual Heritage projects available to the widest possible audience, we at Learning Sites have expanded our activities and formed the Institute for the Visualization of History (see the organizational chart at the left; hover over to enlarge).

The Institute's goals include the preservation, storage, and retrieval of Virtual Heritage data through a digital archive with legacy systems (see the chart at the left; hover over to enlarge).

Our public gallery will have high-end immersive systems for showing VH digital content to school groups; in conjunction with regional museum exhibits; and for the general public (see the chart at the left; hover over to enlarge).

We will encourage the digital publication of scholarly material, especially VR-based excavation reports, developed directly with field archaeologists (see the chart at the left; hover over to enlarge).

We also look forward to educating the next generation of content providers with the first comprehensive Masters Degree program in Virtual Heritage.

What is the future of VH beyond current projects (hover over the image at the left to enlarge)? Very soon virtual heritage worlds will be much more complex realtime, networked environments updated and tracked globally. These worlds will contextualize not only places and objects, but also people and events. The whole historical fabric will become available, including intelligent agents, people (with behaviors), vegetation, animals, and climate. N-dimensional semantic networks will link complex datasets to the worlds via routines based on simulations and game engines.

What is the future of the technology? Work is currently underway in a number of areas that look promising for the next major leaps in virtual heritage, such as ubiquitous computing (anywhere, anytime), projection holographs, floor-mounted 3D TVs, on-demand augmented reality; on-demand "world-boards"--handheld devices mixed with wearable systems with global access to information and location-aware 3D worlds; such devices will have universal statistically based translators for instant "any language to any language" translations for multilingual access to text or video.

Is this the virtual time machine? Can we and do we want to visit the past as if we were first-hand observers of ancient events? We can't change the past, but we can understand it more fully so that we can plan for a better future; we can learn from our shared heritage and strive toward the DigiCult goal of Ambient Intelligence.

In conclusion, we can agree that understanding the distant past is not easy--we weren't there. 500 years ago we could only muse about how surviving texts from the past might connect with a few known relics from the past (hover over the image to enlarge).

100 years ago we only had photography and hand-made sketches to record new finds and make sense of our interpretations (hover over the image to enlarge).

10 years ago we had 3D computer graphics, at a steep price (hover over the image to enlarge). Today we have 3D laser scanners, GPS, GIS, and numerous real-time visualization techniques. To paraphrase the character K in the movie Men in Black: Given this pace of change, just imagine what we'll have tomorrow.