| Design Rationale: Problems and Progress Program Workshop Notes |
|
||||||||||
Workshop aim: Typical design documentation merely describes the results of the process: a snapshot of the final product, with little insight into how these results were achieved. An alternative approach, which grew out of Rittel’s 1970s work on an Issue-Based Information System (IBIS), is to attempt the additional capture of the rationale behind the design. This includes the decisions required; all the alternatives evaluated; the reasoning behind the choice of one alternative over another; and the dependencies between earlier and subsequent decisions. This information can be useful in many ways, such as for redesign, reuse, maintenance, learning, documentation, collaboration and project management. Despite there being widespread acceptance of the potential usefulness of Design Rationale (DR) repositories, they are still not in general use. One obstacle has been the widely held perception that capturing the rationale is expensive, and that designers will be resistant to using a capture tool. Three years ago, the IBIS-derived Compendium dialog mapping tool was released for free download, and subsequently it was made open source. The tool is steadily growing in polish and sophistication, which has helped to foster a thriving research and user community. This community has largely focused on the use of the tool for collective sense-making, generally by specialist facilitators capturing real or virtual meetings. While the words “Design Rationale” now rarely appear on the Compendium discussion forum, the software’s sharp improvements in usability, functionality and desktop integration, raise the thought that it might after all suggest a way for designers to capture their own routine deliberations. Furthermore, it has recently been reported that a major international power and aerospace company has successfully adopted a simple IBIS-based DR capture tool, as a fundamental part of its design process. Use of the tool is reported to have spread from a trial of an early research prototype, largely by personal recommendation among designers who found it useful, without requiring any form of coercion. In addition to these recent advances in argumentation capture, significant progress has also been made in the capture of DR by the computational interpretation and augmentation of design sketches, and by improved computer support for simple but meaningful diagrammatic representations, in the early stages of design. All of this suggests that the time may be ripe to re-examine some of the pessimistic conventional wisdom concerning the viability of DR capture. Participation is invited from researchers in Design Rationale and related fields. Anyone interested in attending the workshop will be welcome to do so (space permitting). |
|
||||||||||
Workshop
format:
The half-day workshop will take place before the beginning of the DCC conference. It will be designed to facilitate discussion and will consist of presenting papers chosen to encourage this discussion.
Topics include but are not limited to:
Special Issue of AI in Engineering Design, Analysis and Manufacturing:
Dr. Janet Burge and Dr. Rob Bracewell will be guest editors of a special issue on Design Rationale to be published in 2008. All workshop participants will be invited to expand their papers to journal length (up to 40 pages) and quality for submission to the special issue.
See: http://web.cs.wpi.edu/~aiedam/SpecialIssues/Burge-Bracewell.html for details of submission to the special issue.
Submission
information:
Contributions
to the workshop should be submitted in the form of brief position papers or
extended abstracts, approximately 1000 words long. Figures and references can
be included. These papers will be assessed by the workshop committee for inclusion
in the workshop. These position papers should outline a piece of original research
or contain a succinct review of the state of the art.
Please submit your workshop position paper to the Workshop Chairs formatted according to the DCC06 paper guidelines.