HITC senior staff member Michael Harris is chairing a panel at the 1998 SIGGRAPH Conference titled "Interfaces for Humans: Natural Interaction, Tangible Data and Beyond." Huge advances in interface modalities are evident and imminent; Mr. Harris and five interface design luminaries will demonstrate and explore some of the most interesting, promising and clever of these, as well as their integration into powerful multimodal systems.

When users talk about computers, they usually describe the interfaces - because, for most users, the interface is the system. As Bill Buxton says, "The most powerful force in shaping people's mental model of the nature of the beast is that which they see, feel, and hear." It seemed to take forever for toggle switch panels to evolve into today's WIMPs, and both are still visual / motor based controls; in fact, switch panels were probably more haptically satisfying! "Keyboards only work for people who know the Roman alphabet. In twenty years, people will laugh at us for calling that technology," says Mark Lucente.

Now, thanks to exponential increases in commonly available computer power and versatility (and concomitant cost decreases), significant progress in interface modalities and their affordances can be perceived. This panel will explore the most interesting, promising and clever of these advances. We will emphasize demonstrable and practical stuff; we'll have hardware to monkey with, ideas to ponder and try.

This is a gadget-intensive topic, and we will present gadgets galore. Input devices that can tell systems where users are looking, the gestures they are making, the direction and content of their sounds and speech, and what and how they are touching. Display devices that image directly onto the retina, high resolution miniature LCDs, and spatial sound generators. Some of these innovative transducers operate not only non-invasively but invisibly. "No one should ever have to see a computer. The complexity should be soaked into the world around you," says Lucente.

While humans are adept at sensory integration and data fusion, computers are far less so. It is clear (and probably has been since Glowflow in 1968) that multimodal interaction is a seminal goal, and that achieving it is a formidable challenge. Now that computational power seems be catching up with algorithmic understanding, the panelists can report and discuss exciting progress in this area.

Our panelists have decades of experience in interface design. Their perspectives are theoretical and pragmatic, incremental and radical; their work is elegantly inspiring and often delightfully unconventional. All were considered visionaries, but now their visions are achievable and industry is paying attention. They are seasoned practitioners with their own viewpoints, all are articulate, and none is shy; the Q&A and discussion periods promise to be stimulating.