(continued)

One reason for Sonata's failure in the U.S. market, Aish says, is that its concept of "enterprise computing" required a restructuring of teams within the industry. For example, it required more detailed input by architects for the benefit of engineers and contractors "downstream." But there was no motivation or compensation for the increased workload upstream. As a result, now-familiar drafting systems that could run on low-cost personal computers won over the U.S. market, where drawing efficiency could be marginally improved without requiring major structural readjustments.

"But the fundamental weakness here," Aish explains, "is that drawings distribute design intent across multiple independently editable sources. To any information technologist who understands the concept of data integrity, this use of computing by the design professions is completely anachronistic."

Many of the concepts within RUCAPS—multi-user access to a single building model, with drawings treated as 'graphical reports'—are the same as those now being presented by products such as Bentley's MicroStation TriForma. One important difference. Aish believes, is in the attitude of the user community. Having been through the experience of automating 2D drafting, they are now poised to accept more comprehensive design tools that do not simply replicate traditional techniques.

Now software developers are joining academic researchers in calling for the professions to make the necessary professional and organizational adjustments. It will require revamping the relationships between architect, engineer, builder, and owner and rethinking how various players are compensated for their contributions.

Until that happens, Aish says, Bentley is working on methods for easing the transition. Their ProjectBank technology creates a central data repository that is accessible to the entire team yet looks much like a traditional CAD environment.

"Design tools such as TriForma and collaborative engineering systems such as ProjectBank have an important complementary role," Aish says. "Although distributing design intent over multiple drawings was unsatisfactory for coordination, it had the advantage of allowing multiple architects and engineers to work at the same time on different drawings.

ProjectBank is the collaborative tool that gives multiple users simultaneous access to the single building model." This technology allows design managers to selectively "undo edits" or construct new design revisions from elements selected from different historical versions, while maintaining the integrity of the single building model.

Behind the development of such design tools and collaborative systems are important philosophical issues. Should software developers be creating tools that support existing approaches to design, or should they create new tools that suggest different workflows and possibly different organizational structures?

"Both approaches," says Aish, "may bring equivalent entrepreneurial rewards to the software developers but have very different effects on the design community." Software companies like Bentley are not just constructing technologies, they are constructing a 'culture of use' to match the technologies. However, Aish cautions, "We rely on feedback from users to ensure that these tools will be compatible with important values retained from the design traditions."

Bentley is working with researchers in the field of collaborative engineering to understand the key social and technical barriers that have impeded a restructuring of design-through-construction processes.

"This time," Aish says, "we want to get both the technology and the business conditions correct—and correctly matched." Re-engineering and integrating the entire industry, Aish believes, has the potential to foster the development of innovative research tools that will both improve design processes and support behavioral changes needed to promote collaboration.

Many of these issues that occur at the intersection of research, software development, and practice will be discussed at a day-long research seminar on September 17, the day before the Bentley International User Conference, in Philadelphia.

Chaired by Aish, this seminar will focus on how geometric and computational tools can be combined to support a new approach to "visionary" architecture. "One of the signatures of this architecture," says Aish, "is that it combines an exuberant geometric subjectivity with an amazing degree of constructional rationalization. This suggests—or demands—computational tools that merge traditional design methods with algorithmic thought, indeed tools which can span both intuition and logic."

This adds an optimistic note to the story of architectural research. With sufficiently powerful hardware and support software, all the clever research ideas—automating space planning, hand-crafting forms, calling up reference materials through sketching, and performing complex simulations early in design—will eventually become part of a practitioner's every-day toolkit. The obstacles are many but the rewards are compelling.

B.J. Novitski is managing editor for ArchitectureWeek and author of Rendering Real and Imagined Buildings.

An abridged version of this article first appeared in Architectural Record, December, 1999.

The Esplanade: Theatres on the Bay, a new performing arts center for Singapore, was designed and modeled in TriForma by the architectural firm of DPA, Singapore. Image: Bentley Systems Complex geometric structural elements in DPA's performing arts center would have been nearly impossible to calculate with 2D models. Image: Bentley Systems TriForma's single-building model technology made manufacturing the building's glass shells relatively simple and cost effective. Image: Bentley Systems   Click on thumbnail images to view full-size pictures.