In this presentation we are going to show Grok Box from the perspective of its own interface as well as from the perspective of an actual context for its use. The latter will be a refugee camp where medical and other kinds of emergency relief workers are responding to an explosion containing a GI germ.

In this first scene we get a view of the entire Grok Box from behind the head of the user. We could do this several ways. We could start with a black scene which lights up as each perceptualization screen turns on, and the sounds of audio-boot, as it were, as we stand behind the chair or whatever it is that the user is "in". The other option would be to just have the full circus display as the scene opens. The point of this scene is to show a panorama of the fully active GrokBox to give the viewer a sense of its magnitude as perceptualization chamber. This scene could end with some more specific shots of particular screens, close-ups of the tactile feedback interfaces around the users legs and back, etc. Mark was right in pointing out that we probably want to stay clear of any real emphasis on the physicality of the Grok Box user space, be it a chair, a box, a hanging chair, what have you. We want to leave this aspect of the simulation quickly so that the viewers of the presentation are quickly behind the eyes of the user focusing on the renderings and the camp crisis scenes.

This scene will be comprized of video panoramas of the camp coupled with animated stills of the various goings on. There can be some more time spent on certain key shots, but the purpose of this scene is to establish context for use of the GrokBox.

Close-up of user's face in GrokBox. We see the reflection of all the displays covering his face. Suddenly The morass of reflections on his face is interupted by both a pronounced red display of some sort *and* a sound which interupts the many sounds that were previously heard. The user now turns to focus on the particular display/group of displays where the alarm icon(s) came from. Cut to that display(s). Viewer sees these displays closeup for a few seconds and then the scene quickly changes.

Viewer now sees [video of] a single man lying on the ground surrounded my medics and emergency workers. They are all hovering over him with equipment all over the place. Very important in this scene will be the focusing on the moniters he has been hooked up to (e.g., heart, blood, GSR, etc.). Then we want to see these all feeding into a laptop which is then sending this data back to Grok Box. Hopefully our video etc. will give us good focus on the moniter to laptop through LAN onto Grok Box part so we can really emphasize the connection between data gathering, data processing, and decision support which is the whole purpose of Grok Box. Scene ends with close-up on laptop/LAN.

Cut to GrokBox displays. In this scene we want to show how that data which is streaming off this dying man's body is going to be parsed across three sensory modalities: visual, auditory, tactile (in the third act we will show the multi- modal decision output by the user, but here we are focusing on the input from the field). This scene will require the most intensive computer graphics/animation, etc. For example, we will want to be showing how massive amounts of textual information about germs and antibiotics get transformed into dynamic 3D geometric shape "sentences" taken in by the eyes as well as tonal "sentences" taken in by the ears. Also, we will want to show how the tactile feedback stuff occurs over large body surfaces. As I think about this it is probably going to be important to keep perceptual contiguity between the data coming off the body and that being perceptualized, etc.

In the last act we viewed how data streams coming off a patient via physiologic sensors move through a portable computer then through a LAN type tool and on into the GrokBox data funnels, to be filtered and then spread across at least three sensory modalities. We saw in the last scene, in particular, how GrokBox, through intensive perceptualization and modulation of raw data, may render that data as highly refined information. The trained user is then in an exceptional position to make decisions about responses to the many urgent situations going on in the field.

While the last scene featured the 'rendering to the user of the input from the field,' this scene is concerned to show how the user now implements particular information towards decisions. What are some of the ways in which the GrokBox interface allows rapid output of commands?

• Joystick-like controllers (e.g., the flying hands)
• Voice recognition
• Foot activated pressure sensors/pads
• EMG sensors around leg and upper arm muscles

Again, the emphasis of GrokBox is to tap into as many bodily sites as possible for the perception and expression of information. Here we have given a few.

In terms of our guy on the ground we want to show the user making decisions which the team on the ground can immediately implement as they are sent as commands back out through the GrokBox system.

Capacity for the micro-response must not obscure the equally significant macro responses which the GrokBox makes possible. As the GrokBox user has just responded to an individual on the ground, he is also responsible for higher level integrations of data rich/critical multiple situations which comprise this whole crisis scene. Here he is tracking everything from water supplies, food and medicine inflow, evacuation tools (e.g., helicopters and stretchers), casualties, catagion rates, and even intelligence indicating if another attack is on the way. In all these and many more cases, the user is in a constant and dynamic interaction with diverse data sets and requirements. The GrokBox multi-modal perceptual and expressional interfaces put decision support and power in the hands of a single user with a velocity we believe to be unequaled.