[Excerpts from "A Case Study in Participatory Design" by Micheal
P. Snow, Jonathan K. Kies, Dennis C. Neale, & Robert C. Williges,
in ERGONOMICS IN DESIGN, Volume 4, Number 2, April 1996.
Copyright 1996 by Human Factors and Ergonomics Society, P.O. Box
1369, Santa Monica, CA 90406-1369 USA; 310/394-1811, fax 310/394-
A team from Virginia Tech recently completed the workspace design and prototyping of a new Usability Methods Research Laboratory (UMRL). The team was composed of professors and graduate students from the Industrial and Systems Engineering Department and the Computer Science Department with input from interested parties all over campus. During this process, we had the opportunity to gain firsthand experience with the practice of participatory design as well as a variety of design tools.
Participatory design has its roots in human-computer interface design and is a family of methods, techniques, and philosophies that leverages direct involvement of end users throughout the design process (Schuler & Namioka, 1993). We learned several lessons during our participatory design experience and found that various tools ranging from low-tech (transparencies and pens) to high-tech (virtual reality) proved helpful during different phases of design.
Phase I: Planning
The first step was determining system goals. The mission and goals of the UMRL were derived from a proposal submitted to, and supported by, the National Science Foundation. The mission of the UMRL is to provide a networked facility supporting interdepartmental research in innovative human-computer interaction (HCI) technology, formative and summative usability evaluation methodologies, and information access through digital libraries, networking, multimedia, and teleconferencing technologies.
An initial set of design principles and constraints included:
- Maintain maximum flexibility
- Provide disability access
- Accommodate research needs
- Ensure sound isolation
- Support meta-observers
- Connect to other laboratories
- Consider existing columns and load-bearing walls
- Provide emergency exits
Based on these considerations, computer-aided design (CAD) drawings were developed and workspace was allocated for different functions. The drawings were used in team meetings to discuss which functions were critical, how much space was necessary for the various functions, and how the space might be arranged. The drawings were helpful in identifying critical functions and behavioral specifications.
Phase II: Preliminary Design
After the overall dimensions of the UMRL had been determined, the amount of workspace allocated to each function and its arrangement needed to be specified. Space was required to support activities such as laboratory administration, central control of experiments, data analysis, electronic conferencing, equipment storage, library and reference, project development, subject welcoming and queuing, usability testing, usability methods research, video editing, workroom and shop, software development, and student observation and teaching.
Design meetings focused on CAD drawings. As the design process matured, issues concerning equipment requirements, electronic networking, and media formats began to surface. To help address these issues and better visualize the different laboratory designs, we constructed a desktop virtual environment walkthrough with Virtus Walkthrough Pro software. Although this was useful, it was almost too early in the design process to take full advantage of it because many basic issues remained unresolved. To help build consensus among the faculty who would use the UMRL, an extensive questionnaire was developed and distributed to ascertain the requirements for software, hardware, and facilities.
Phase III: Dynamic Space Manipulation
The five principal investigators were asked to manipulate the room structure and layout of the workspace. CAD drawings depicting the outline of the allocated room space were presented on the wall using an overhead projector. Opaque cut-outs of possible rooms with likely dimensions were then superimposed on the overhead so that the rest of the group could visualize how the floor space could be laid out.
Phase IV: Physical Walkthrough
The principal investigators, graduate students, and university architects met at the site where the labs were to be constructed. The group marked off where new walls were to be constructed, and was able to see how one particularly debated wall would have to be removed.
The outcome of the physical walkthrough was to confirm many of the ideas set forth in the functional specification document and floor plan designs. However, the tour provided participants a sense of the physical dimensions for which they were designing. Equally important was the architects' ability to estimate major costs, such as the removal of the raised floor, and to share this information with the investigators. The walkthrough also enabled participants to assess the sound-conductive characteristics of the current walls and methods for ensuring proper sound attenuation.
Following the meeting, the architects created the construction blueprints of the new lab based on the agreed-on specifications. The delay between this meeting and the scheduled construction date allowed the team to engage in a virtual mockup and walkthrough.
Phase V: Virtual Mockup and Walkthrough The software used to construct the walkthrough was Superscape VRT (Virtual Reality Toolkit). Superscape VRT's Control Language was used to allow people touring the virtual UMRL to perform some of the normal interactions that they would with a real environment, such as opening and closing doors and desk drawers, turning lights on and off, and raising and lowering projection screens. Interacting with the environment has been shown to increase a user's sense of presence in a virtual environment (Heeter, 1992; Slater and Usoh, 1993) and seemed to do so in our case as well. The walkthrough constituted an integrated 3D blueprint that could be used as a baseline for discussion.
The Value of the Virtual Mockup
It is difficult to place a value on the virtual mockup in terms of money or time saved. Usability problems identified in the virtual mockup not corrected prior to construction would likely have become usability problems that users would have had to live with. The value of the virtual mockup lay in its ability to allow designers and potential users to explore the future workspace and make changes before construction began and problems became "set in concrete."