Two-D, Three-D, Four-Dimensions, more. Just as in the age-old nursery rhyme, vendors now dance to a sing-song chant of escalating "D-ness." Before joining Mr. MXYZPTLK in the 5th dimension (or beyond), it may be worthwhile to D-construct the issues, D-bunk a few myths, and D-tail some D-velopments.
In case you're not a fan of nursery rhymes or Superman, how about a Dilbert metaphor? As a consultant, the Dilbert character with whom I identify most closely is Ratbert, the consultant. A favorite strip has Ratbert drawing the consultants' classic 2X2 matrix and labeling the resulting four quadrants "something, something else, another thing, and whatever" (the punchline, as if one were needed, is that "phase two will have arrows and circles").
So, imagine, if you will, a 2X2 matrix with the quadrants labeled "Two-D Paper Drawings (A), Two-D Digital Drawings (B), Three-D Digital Models (C), and Three-D Physical Models (D)" (reading clockwise from the upper left quadrant).
The actual constructed asset, be it an office, factory, or highway, sits at the logical center (0,0 origin point) of these various representations. Three-D physical models (Quadrant D in the matrix) pre-date drawings in any medium (Quadrants A and B) and all digital representation (Quadrants B and C), yet rarely are used today as part of the construction documentation process (however, the architectural firm of Frank O. Gehry & Associates often uses 3-D digitizers to capture Frank's clay or cardboard models into computer-based 3-D models that, in turn, do serve this firm as construction documentation).
Getting from A to B
The relationship of 2-D paper drawings to the constructed asset (A=>0,0) is today a well-understood process. There is ample custom, practice, and legal precedent to address every issue from priority of specs versus notes versus drawings, to questions involving wet signatures and embossed seals, and on to matters such as notice via certified mail, return receipt. We have not yet achieved the same level of clarity in our understanding of the relationship between 2-D paper and 2-D digital representations (A=>B)—for example, if there is a conflict between a plotted drawing and its digital "original" as viewed on a monitor, which takes precedence?
The relationship of 2-D digital drawings to the constructed asset (B=>0,0) is almost entirely undefined. Although Autodesk and Bentley Systems have each announced and/or shipped digital signature extensions or upgrades to their respective flagship design software products (representing 70% to 80% of all constructed asset documentation), the user community is only now beginning to explore how to use such tools in practice.
In large measure, I think the 95% failure rate among project collaboration network (PCN) startup ventures at the turn of the millennium can be attributed to their premature efforts to Web-enable and move on-line these B=>0,0 processes that have not yet been adequately defined or resolved off-line—a kind of premature e-collaboration problem. Furthermore, the potential payback of pure 2-D document sharing systems may not always justify the effort in project terms; knowledge management spin-off benefits are, however, another story (and one for another time).
The Shortest Distance Between Two Points
The relationships between 2-D digital drawings and 3-D digital models (B=>C) also is not well-settled. In the AEC space, a side-by-side comparison among Autodesk Architectural Desktop, Autodesk Revit, Bentley Architecture for Triforma, Graphisoft ArchiCAD, and Nemetschek NA VectorsWorks Architect—to name just the five leading solutions in North America—reveals widely varying approaches to assembling models and/or deriving drawings from them. Despite one product's outrageous claim of "perfect coordination 100% guaranteed," the issue of potential discrepancies between model and drawing has not been settled in practice or at law (again, only one among many unresolved issues).
Thus, if we're not clear about the real-world relationship of 2-D paper drawings to their digital equivalents, nor the relationship of digital drawings to the construction they represent, nor the relationships between 2-D and 3-D digital representations of construction, we shouldn't be surprised that the relationship between 3-D digital models and the physical building (C=>0,0) is an almost total unknown. To be sure, there is interesting work being done in rapid prototyping (creating physical components directly from 3-D digital models), especially in some of the more progressive (and well-funded) design schools, such as Columbia, and MIT (although some of that effort is directed at the relationship of 3-D digital models to 3-D physical models (C=>D)). Of course, there is always the counter-example of Frank Gehry, whose construction project manuals often bravely specify that "all dimensions shall be scaled from the (digital) model" (wow! Think about that one in early adopter terms).
A Long and Winding Road
Does all this imply that 3-D digital models are a waste of time and effort? To the contrary, the fact that so many folks are willing to struggle so mightily against the procedural obstacles to 3-D digital design and documentation is testament to the compelling benefits that can be achieved, if only for individual player's internal benefit in today's world. We are entering a transitional phase where engineers and architects will find ever-increasing value in 3-D digital models for their own purposes, even though they still must extract 2-D digital drawings and plot those as 2-D paper drawings in order to issue bid sets, contract sets, and permit sets. Similarly, enlightened contractors may choose to create their own 3-D digital models for costing and constructability purposes, even when the project designers have furnished only dumb drawings. As constructed assets grow more complex, owner-operators may want 3-D digital models of their facilities, even if the 3-D digital data has to be obtained, not from designers or constructors, but by laser-scanning the physical object (a technology of which Cyra and Quantapoint are the two leading proponents).
Cross-discipline, multi-enterprise collaboration is still today's leading edge of the wave of the future. Adding the fourth dimension of time and the 5th dimension of cost only makes the ultimate payoff that much more rewarding. Early adopter, one-of-a-kind outfits like Frank Gehry's design firm, or Texas designer-builders the Beck Group will continue to "shoot the curl" of the technology wave. Forward-looking endeavors like the National Institute of Building Sciences' IAI/IFC efforts, the BLIS Project, and the Virtual Building Model research among Stanford, Lawrence Berkeley Labs and the Finnish national technology agency, Tekes—among many others—are to be applauded and supported throughout the mid-to-long term.
In the sort term, however, individual project entities, from designer to constructor to owner-operator, still need to spend a few years exploring and learning the many benefits and occasional "gotchas" of true multi-dimensional design. Only then will the AEC, plant/process, and infrastructure industries be able to cross over to the mainstream, mass adoption phase of multi-D technology.