The structural engineer was traditionally bound by the limits imposed by the available methods of analysis. Those available in the first part of the 20th Century were graphical: methods that were not necessarily simple, but methods that were distinctly limited. Since a Graphic Statics Analysis was completed most rapidly with determinate structures (i.e., simple beams, three- hinged arches), structural designers utilizing these methods usually designed structural systems which were also determinate. (There were, of course, notable exceptions to this rule.) It was only as graphical methods were gradually subsumed by convenient "rational" numerical methods that indeterminate structures (i.e., continuous beams, two-hinged arches, rigid frames), with their increased efficiency in utilizing the material of the entire cross-section, became commonplace. The addition this new method of analysis implied the availability of a broader structural vocabulary not only for the designing engineer but also for the architect. As the age of mass production impressed its urgencies upon the European western societies, the numerical methods became "mass-mechanized." Speed and accuracy of each and every calculation vied for primacy over craft and understanding. Indeed, with the computer, speed and accuracy became the only motivation.

The original mainframe computers were housed neither in plastic boxes nor single air-conditioned rooms. These megaliths, which could only add, subtract and multiply, occupied multiple floors of buildings. They spawned an evolution which continues in a direction that was the material for many science fiction novels and comic strips of the 1950s. Long complicated equations were no longer stumbling blocks in the search for efficient structures. The "number crunching" capacity of the black box also allowed the application of matrix analysis to structural elements. It provided the medium for finite element methods, which had been applied to the field of mechanical and electrical engineering, to cross over into structural design. Finite element analysis allowed again an even more rapid and exact determination of the behavior of a certain structural element or system. The structural designer has the capability of determining the stresses within a structural member at a speed, and to a degree which was unheard of just ten years ago. The iterative analysis work which then consumed hundreds of hours of "structural design" time is now completed in a matter of minutes; over 10,000 times faster. This speed implies an almost limitless capacity to precisely solve very complex functions. It further implies that due to this limitless capacity, the structural analysis method should no longer be the boundary for structural design.

As the analytical degree of exactitude approaches certainty, the question must be raised of what has been replaced. Has this exactitude increased the design freedom of the structural, or architectural, designer? The increase in the design vocabulary that the computer allowed was the shell. The main-frame computer and its finite element analysis allowed the iteration necessary to solve the multitude of possible forms for shells. This occurred in the 1950s and into the 1960s. The effects in architectural design need not be renumerated here. What effect has the personal computer had? or will it have?

Are there structural systems or building materials being developed which will take advantage of the lightning analytical speed? Will the dream spaces and fantasy structures of what one knows as the visionary designers become reality -- as the well known "wristwatch" of Dick Tracy will become reality in a matter of months? In 1951, Matthew Nowicki stated in the Student Publication of the School of Design, North Carolina State College(vol.1, no.1, p.14), "No new form of architecture could have been created without a new structure, and the psychological receptiveness had to wait for its fulfillment until the structural possibilities ripened." Tensegrity was one such structural concept. At the time of its inception it, too, was considered to be a structural fantasy. It was a system developed with a vision far beyond the capability of the contemporary methods of analysis. It stretched the imagination and understanding, it did not mimic comfortable images of previous structural designers.

The inverse of this is found in the designs of the office of the popular contemporary designer, Santiago Calatrava. The structural systems of his office, which seem to be clearly self- evident and a result of a logical system, are of a type that Nowicki described in 1951 as designs in which the "form follows form and not function"( Ibid. p.11). Indeed, the structural forms have resulted from a functional analysis but the analytical tools have been slaves to the desire to achieve a singular distinct image: the form supreme. They are not the result of the evolution of the form. Hundreds of hours of calculations are made until justification for the form is found. (There are many who find no fault in such a method.)

The profession on the "other side of the fence" has not escaped the influence of the computer. An American architectural office without at least one CAD workstation is rapidly becoming a rarity. The office of WHL Architects and Planners in San Jose, California, which traces its success to trading in their drafting boards for workstations almost ten years ago, works solely with related professionals who also are similarly equipped. The endless hours of drawing and re-drawing plans has been reduced to a matter of cutting, pasting and revising elements from a well developed design library. The two-dimensionality of the design process, aided as it was with tediously constructed perspective renderings, has evolved into a fully three-dimensional process in which the designer, or client, can virtually "walk-through" a completed design before a brick has been laid. The hours spent on checking and re-checking drawings, the source of most of the construction site confusion, is now accomplished with real-time on-line communication. Even the express mail disk traffic has been eliminated.

This evolution would imply that the decrease in time required to complete a singular design would allow an increase in the time available to explore alternative design solutions. However, this is not usually the case. The "best" design has a plethora of meanings. It is the definition of the word "best" that has been a determinant in the rift between most architects and civil engineers, as well as a contributor to that between the contractor and the designer.

The rift between designing civil engineers and architects is not widening or closing; it is being redefined. The process of redefinition has been hastened by the personal computer. The architect now has the possibility of adding modules to CAD programs which provide everything from construction cost- estimating and fast-track planning to energy analysis with on- line real-time design recommendations. It is commonplace for the mechanical engineer to design a new part at a workstation, send it to the workshop via cable, and have the part manufactured in the time it took to walk to the shop. Programs have been designed to allow specialty contractors to take dimensions directly from an architects drawings so as to streamline the bidding process. Developing technology allows for the "instantaneous" transmittal of changes or problems from the field to the office so that all associated documents are automatically updated.

Where is the limit to this modularization? This will unfortunately be determined not by practicality or logic, but by the issue of liability. This will not be discussed further in this paper. But knowing that the majority of the calculations for either a structural or an energy analysis, for example, are straightforward, iterative and/or linear, these can, and are, routinely accomplished with computers. It is only a matter of putting the right numbers in the right place. Therein lies the difficulty. Any expanded CAD program would have to be able to "pull off" the correct values from the architectural design documents to use for an analysis. This, too, is no longer an issue with contemporary programs. As a designer (taken to the extreme, it becomes a moot point whether an architect or civil engineer) can sit down and begin to draw, the modules of the CAD program can make recommendations on demand for any number of issues. One person could design an entire complex: or could he? Would the modularized CAD program be able to replace, or replicate, the creativity of the design team? The synthesizing process of the human brain is at this time irreplaceable. The amount of selection that takes place that is imperative to the process cannot be accomplished by the computer.

A computer could work through many times more calculations than a design team could in the same amount of time, but it cannot synthesize the volumes of information in an interactive and creative way. In rethinking the relationship between the analytical methods applied to a structure and that structure's design, one finds that the limiting factor was not the degree of exactness of the analysis but the knowledge of how to interpret the results. This limiting factor will remain the same, no matter the simplicity or complexity of the modularization. The user, the designer, must have a level of knowledge which will enable them to interpret the recommendations of the modules. With the current state of our educational system this is a questionable, and frightening, situation.

The personal computer allows each of the neighbors to peer into the "realm" of the other. Pre-packaged design programs can be purchased for as little as $39.95 -- structural analysis programs for approximately the same price. The fence is rotting in a figurative sense. When this occurs, the fence can either be rebuilt, or allowed to decay further. Perhaps we can replace the preoccupation with the quantitative with a desire for the qualitative in all aspects of design.

Material appearing in Architronic may be distributed freely by electronic or any other means, providing that any such distribution is without charge (unless for purposes of cost recovery by interlibrary loan services) and that Architronic is acknowledged as the source. However, no article may be reprinted in any publication without the explicit written permission of the author(s). This statement must accompany all distributions of Architronic, whether complete or partial.