Enabling Technologies and Methodologies

Jeremy Cooperstock (leader), Carl Gutwin (co-leader)

In the last few decades, our society has leapt from passive consumers of pre-packaged information and entertainment to more active participants in the production, selection, and consumption of such content. At the heart of this shift, the transformative media applications that have become commonplace in recent years (such as online games, electronic mail, voice over IP, Internet search, wikis, media-sharing sites, mobile devices, and social networking systems) were possible only because of numerous technological advances. These technologies -- including standardized network protocols, new rendering algorithms, publicly available application programming interfaces (APIS), methods for evaluating web usability, and the increasing availability of broadband connectivity -- enable designers, developers, and content producers to capitalize on the potential of interactive media-rich applications. The areas of computer graphics, human-computer interaction, networking and distributed systems, and information retrieval represent important infrastructural areas that support the interactive access to media content by a large population of users.

Over the coming years, we expect that the amount of geographically distributed digital media content, already massive, will continue to increase. Media data will be stored in machines with a wide variety of capabilities, from personal digital assistants to laptop computers to powerful servers. The capability of database management systems used will also be wide-ranging. Next, we anticipate growing interest in service models with increased support for choice, personalization, and differentiation. In addition, users will demand improved levels of privacy, security, and information filtering. Finally, important issues in new media research include high quality presentation for images, audio, and video; support for a range of end-user equipment, from cell phones and laptops to high definition displays; high quality animation and multi-user online games; scalable system architectures that can support additional users without degradation in performance; and quality of user experience.

Building on past successes and looking forward to these trends, the goal of GRAND's TechMeth theme is to identify the necessary advances in architectures, techniques, methods, and tools that will enable the next generation of graphics, animation, and new media applications.

Although the existing infrastructure has supported an impressive number of developments in these areas, applications are reaching hard limits in all of the projects proposed for this NCE. New technologies are clearly needed. For example, as video games become increasingly realistic, the limitations of current techniques for computer control of characters is evident; the game industry requires new solutions to the problems of creating and controlling believable, natural characters. similarly, interaction between distributed musicians or actors in a computer-mediated performance suffers from the limited quality of the media and the constraints imposed by the technology. And in our information-driven society, as data visualization becomes an increasingly important tool to facilitate understanding, the role of aesthetics in the visual presentation must be considered. Other examples can be found in graphics, animation, information retrieval, and content production. The invention of new forms of interaction and new types of applications requires that the limitations of current technologies be overcome.

This research theme is concerned with the identification and development of the building blocks that will be used to invent, design, produce, and evaluate the next generation of media applications. An important underlying objective of the theme is reuse -- that is, knowledge and understanding about new media applications that can be encapsulated and passed on to others in both research and industry. This ensures that the outcomes of our work will have powerful lasting value beyond the confines of the NCe. broadly speaking, the theme will concentrate on four categories of reusable knowledge that will be produced in the NCe: architectures, specific techniques, methods and methodologies, and tools.

Architectures and Frameworks

High-level conceptualizations of a design space provide application developers and content producers with a structure that organizes thinking about the domain, and that helps guide the search for solutions to novel problems. These structures can have revolutionary effects on productivity and access for an application area. For example, the concept of tiered architectures has had enormous impact on modern interactive systems, and the web services model has transformed and facilitated the development of software that involves the participation of multiple machines. There are similar needs, and similar opportunities, for highlevel structures (ENCAD) in the next generation of graphics and new media domains. Several projects will contribute new or refined architectures and frameworks -- conceptual frameworks of privacy and security (PRIVNM), new conceptualizations of design for assistive technology (INCLUDE), and theories of digital infrastructures (DINS).

Specific Techniques

The variety of problems encountered in GRAND's research related to the design and development of new graphics and media applications will result in many solutions that are specific to these challenges. Examples include new techniques for personalization to enhance the usability of media-rich applications, network techniques for minimizing latency and enriching the possibilities for coordinated interaction (HSCEG), graphics algorithms for improved models of human motion (MOTION), techniques for improved search in games, and new artificial intelligence strategies for more believable characters in computer games (BELIEVE). These techniques will be of enormous value for researchers and industrial partners; all of the technically-oriented projects in the NCe will contribute techniques to this theme.

Methods and Methodologies

Many forms of knowledge that will be created in the NCE's research can best be captured as processes. Examples of these include evaluation methodologies and development processes or design approaches that encapsulate a successful way of arriving at a solution to a problem. Several projects will emphasize the development of such methods and processes: from methodology for evaluating affective user experiences (AFEVAL) or the qualities of virtual worlds, to a process for modeling natural phenomena, and new methods for the development of serious games.

Tools

Tools are the most concrete form of reusable knowledge that will be created in the network. In this subtheme, our products will include reference implementations, toolkits, libraries, engines, and development environments. In all projects, researchers will work with industrial partners to develop reusable tools where appropriate to the products. Examples include a development environment for exercise games, a library for Monte-Carlo search techniques (MCSIG), techniques for parallelization of videogame platforms (PLATFORM), reference implementations for human motion models, and models of natural phenomena.

The industrial focus for much of GRAND's research means that infrastructure and reusable technical knowledge will be a vital part of our success. Our multiply interrelated network approach is a vital part of this theme, allowing for both a multiplicity of new techniques and methods from the thirty research projects, and the opportunity to share and test reusable technology across multiple sites.