CDR DesignSpace: Presence

DesignSpace

Spatial Data Presence

This page is under construction.
Please excuse the dust and debris.


Historical Preface

Early computer-aided-design tools (CAD) became useful with the invention of the pointing device. In this original implementation, even though the pointing device allowed simultaneous specification of two independent parameters, any task requiring parameter specification was issued one dimensionally from the command line with a memorized syntax. Menus on the graphical screen, in range of the pointer and with a specified hot area surrounding each entry forming a button, allowed the large selection of tasks to be visually tree searched, and thus freed the user from memorizing and typing entries in the ever growing command set. Still the interface was one dimensional, in that the button only scripted the command to the command line and prompted for the syntax, and the command remained a serial stream. Command combinations were impossible, frequently requiring systems to provide separate hybrid commands (with their own name and syntax) to facilitate complex tasks. A constrained copy/displace command would be a good example, resulting in array, offset, mirror, etc. Such was the state of the art in most all commercial CAD tools a decade ago, when AutoCAD, for example, became popular and found its way onto a quarter million desktops.

Then came the Graphical User Interface (GUI,"goohy") to give designers "direct manipulation" and a toolbox of icons that could be selected and applied to specific objects in constrained ways. The parameters to the tool, and thus the task, could be adjusted through a dialog box, alleviating syntax memorization. GUI's were sometimes frustrating to veterans of the command interface, since all the picking, clicking, and switching input devices (keyboard and mouse) sometimes slowed them down. To novices, GUI's were a great step forward.

At this point in its history, CAD would have been in good shape to continue as the prime provision of 2D drawing, design, database, and documentation tools for describing real-world artifacts with annotated orthographic viewsets. But 2D CAD was missing the entire point by being a model of paper which is itself a model of the volumetric objects which were under design. Why should the model in the computer not describe the object of the design instead of a piece of paper?

Today CAD is struggling somewhere in a transition between its roots (a paper, scale, and rule replacement) and its future (total design simulation with a full spatial model). While much of the research is being put into increasing the sophistication of the tools, such as via artificial intelligence techniques such as agents, the biggest battles are being fought over the paradigm shift in the design process product (formerly drawing and specification documents) and the user interface to access the spatial model within the constraints of a standard computer interface. DesignSpace proposes first that the standard computer interface need not be a constraint, and second that maybe the standard computer interface should change.

Presence

section in review

Evolving Human-Computer Interaction (HCI)

Spatial User Interface

Stereographics, spaceballs, flying mice.

Spatial User Presence

Immersion, virtual reality.
Putting the user into the data.

Spatial Data Presence

Data illusion.
Bringing the data out into the user's environment.


_________________________________________DesignSpace CDR Stanford PaloAlto ______
chapin@cdr.stanford.edu

last update - 15 Feb 94