Originally published: May 2008

protoBuilding
To BIM is to Build.
Abstract: For generations, architects have viewed drawing as the best way to understand and 
communicate design intent. Building Information Modeling is generating a new awareness and 
power  to  represent  buildings.  More  than  just  a  new  representation  tool,  genuine  BIM  is  a 
building activity in its own right.   

Rene Magritte – La Trahison des Images. www.lacma.org

© René Magritte Estate/Artists Rights Society (ARS), New York/ADAGP, Paris

Introduction
In a famous 1929 painting, The Treachery of Images, Renee Magritte painted a pipe under
which he declared “Ceci n’est pas un pipe’”(This is not a pipe). Initially puzzling, he later
explained – “Could you stuff my pipe? No, it’s just a representation.” Magritte’s work is
one of the best-known commentaries on the limits of representation in capturing important
aspects of real-world objects. The painting was a good facsimile of a pipe, but it lacks
crucial aspects of ‘pipeness’

There is an obvious similarity to the world of building design. Buildings are intensely
complicated, and so immense, that the most reliable way to convey design intent has been
to rely on pictorial views which try to describe their configuration. As a result, the
architect’s work has traditionally become an exercise on paper trying to represent
something that has a different, more physical reality.

As a profession, we have devised standards and workflows in which drafting and

specifications, produced with standardized graphic conventions, serve to reasonably

© 2008 John Tobin, AIA, All Rights Reserved 1

 Originally published: May 2008

document the 3D building. We understand however, that there are limitations to this
drawing-based method, with its multitude of related 2D viewpoints on numerous
disconnected sheets.

To minimize the potential for confusion or conflict, many of us have ‘General Notes’
paragraphs which in effect say “This is not a building” (we actually say, “Do not scale
drawings”; “In case of discrepancies, call architect”, etc..) This is because we understand
the limits of representation and know that drawings are just our best conventional tool to
represent a building.

Building Information Modeling, it can be argued, offers an entirely different proposition,

one that allows us to move beyond representation and into a simulated building process.
Though many designers continue to use BIM to produce drawings, - and a recent industry
survey favored that issue over several others1 - this is probably a transitional phase for
BIM. We would be better served to look beyond using BIM merely as a more powerful
representation tool, and instead to treat the models we create as ‘protobuildings’.

The Greek Proto - as in prototype – suggests something that is a first version of an entity
which will later be replicated. BIM models clearly fall under that rubric, being the first
iteration of a building. The process of creating these proto-buildings we might call
prototectonics, but more recently it has been called virtual construction.

After two decades of continual advances in 3D object-oriented technology, architects now

have the opportunity to build information models with the same rigor as one would build
a building. If we are sufficiently motivated, we will not model to merely show others the
intended outcome of a different, separate process. Rather we will proto-build - that is, make
a first-building. In doing so, we will gain an unprecedented advantage over any previous
architectural era - the opportunity to build twice.

BIM representation.

1
Top Criteria for BIM Solutions: AECbytes Survey Results, Oct. 2007

© 2008 John Tobin, AIA, All Rights Reserved 2

 Originally published: May 2008

Generations of BIM activity

To understand how this opportunity will materialize, it would help to look at both the
current state and also the likely future trajectory of BIM. For the purposes of our argument,
it helps to think of BIM as having three different generations, which we will call BIM 1.0,
2.0 and 3.0.

Let’s examine the generations in sequence.

BIM 1.0 - CAD on Steroids: In the BIM 1.0 era, model-based software emerged against a
background of a 2D CAD production workflow, where the major benefit was better
coordinated and faster production of documents. The goal was to model with objects and
to minimize tedious drafting by having one 3D object handle multiple 2D representations
when placed in a project.

Over many years of development, increasingly sophisticated object-based CAD programs

emerged that improved drawing coordination and also added data fields to the objects.
These tools now permitted real-time generation of schedules and 3D representations. In
comparison to previous software, these applications were like CAD on steroids; smarter
than reams of 2D CAD, reducing drafting chores, and also eliminating tedious counting.

Because the tools were now incorporating data, the term ‘Building Information Modeling’
was soon born. The objective was to use the data features for schedule generation, which
would ultimately end up on paper. Though sometimes troublesome to work with, the
software grew in acceptance largely for these coordination benefits. It’s worth noting that
the BIM 1.0 workflow was still tackling traditional representation tasks.

The most significant characteristic of the BIM 1.0 era however, is that it was still in the realm
of the A/E alone, so designers could elect how deeply they wanted to implement the new
techniques.

Detailed BIM models.

BIM 2.0 - The Big Bang in Reverse:

The second phase of BIM - the 2.0 era - dawned when the realm of people involved in
building models expanded beyond the A/E professions.

© 2008 John Tobin, AIA, All Rights Reserved 3

 Originally published: May 2008

We could call this stage The Big Bang in Reverse because the diverging galaxies of
designers and builders - which had been hurtling away from each other over several
centuries - suddenly reversed direction. Two groups that had grown apart now found
themselves drawn quickly back together by the potential of this new technology.

In contrast to BIM 1.0, the BIM 2.0 situation becomes far more complicated for architects,
because various groups are seeing completely different possibilities - and requiring
different performance - from the same artifact. What architects continue to need as a
production tool for documentation now also becomes a bill of materials or a logistics tool
for contractors, as well as a facilities management tool for owners. In the BIM 2.0 era,
contractors are helping to popularize terms such as 4D (time) and 5D (money) models,
which allow contractors to use them for phasing and quantity takeoff purposes. Not to be
outdone, design sub-consultants start to see BIM models as an engine for energy and
environmental analysis. The expectations of the BIM process thus increase significantly.

BIM 2.0 is producing a divergence of opinion in how BIM models should be built, who will
properly build them, and what their purpose is. Architects are struggling to absorb how
to build these models correctly. Though it does not seem unreasonable, it takes us outside
of our comfort zone, and certainly our typical education. BIM 2.0 – which is currently
ongoing - is a highly tumultuous era with architects trying to learn the new techniques that
BIM models enable.

The Association of General Contractors (AGC) added significant momentum to this

reverse Big Bang when they published their Contractors Guide to BIM in late 2006.
Somewhat in response, the AIA accelerated the shift in gravity by developing an initiative
known as Integrated Project Delivery, which encourages close cooperation between
designers, owners and contractors. Owners groups such as CURT have also acknowledged
shared common needs between owners, contractors and designers with the 3xPT initiative.

Before too long a familiar term, interoperability, - meaning how various information gets
exchanged from one partner to another - emerged with new urgency as an industry-wide
issue, and designers are now faced with creating models unlike anything they had
previously seen. Clearly the BIM 2.0 model has moved beyond a tool for representation
and is being thought of as an intricate work in its own right.

Detailed BIM models.

© 2008 John Tobin, AIA, All Rights Reserved 4

 Originally published: May 2008

BIM 3.0 - Post-Interoperability: Though interoperability is currently a thorny and

seemingly insurmountable problem in BIM 2.0, we need to take a look beyond to a third,
future phase for some direction. There are several efforts afoot – such as the IFC and
BuildingSmart initiatives - to promote inter-disciplinary exchange of information and
expertise across disciplines. The BIM 3.0 era – the Post-Interoperability era - imagines what
BIM will be like when working together is a seamless proposition. In BIM 3.0 and beyond,
the various parties will construct a model, not as a representation for their intermediate
purposes, but as a full dress rehearsal for construction, working with the model as they
would a real building.

The BIM 3.0 sandbox will probably be a net-centric database, one where BIM models, now
proto-buildings, are constructed and populated collaboratively in a web-hosted 3D
environment accessed from anywhere. All participants including contractors, trades and
owner will understand how to collectively contribute through their own discipline-specific
modules. The architect’s initial proto-building will ideally have continued value
throughout design, construction and fabrication efforts. It may alter form, and overall
content, but it will be a continuum of activity.

BIM 3.0 may appear at first glance to be a rather idealistic scenario. But the intent is not to
ignore real present-day problems, but to look beyond them for a moment so we can make
some steps now towards a better BIM process. The interoperability challenge will be
overcome eventually, and architects need to be prepared for our role when it does happen.

Innovations in computing technology will provide a piece of the puzzle necessary for BIM
3.0 to occur – faster WAN speeds, and more robust software will enable the networked
model to become the vessel for all manner of design input. But attitudes will also be an
important ingredient, one that technology alone cannot overcome. One of the hallmarks of
success in BIM 3.0 will be a cross-disciplinary attitude that looks at buildings as a shared
enterprise, especially between designer and contractor. Also important will be a belief that
architects can initiate a process that continues throughout construction.

This prompts the question - if the trajectory of BIM materializes as imagined above, what
will be the impact on, and the role of, designers?

.
Building BIM Doors

© 2008 John Tobin, AIA, All Rights Reserved 5

 Originally published: May 2008

Build it as it will be
As outlined above, the development from BIM 1.0 to BIM 3.0 is marked by a shift from
representation to proto-building. In the BIM 3.0 era, drawings will not be a critical
component of the design-build workflow. What will be important is how well the model is
built to embody and enable the building process. The act of building BIM models therefore
will in fact become a credible building process in its own right. It may be digital, but it will
still feel like a process of construction.

Some early BIM adopters are already getting a taste of this. Throughout many years of
using BIM tools, there have been numerous points where, confronted with a
documentation issue, the question arises – “how should I model this?” The inquiry
inevitably turns to the question; “first, what is this element in real-life?” Usually the path
to success is found from a deeper understanding of the element being modeled, and then
modeling it as such.

This prompts two observations. First, architects are now often forced to ask basic,
fundamental questions about something we thought we already knew quite well – such as
creating doors. It is difficult to believe something as mundane as a door could prompt a
deep philosophical discussion, but that is what appears to happen as we try to grasp the
construction implications around a simple BIM component.

In traditional 2D work, a door is often signified by a panel swing and a frame. Other
information gets added in a schedule, and yet further elaboration is created in a door
legend. In a BIM environment, deeper thought is required. This is because how that single
door object will get transferred throughout the downstream processes influences how it
gets created in the first place; (is it a frame with a panel inside, or a panel with a frame
surround?). Take the thinking behind the door and apply it to all aspects of buildings;
windows, ceiling, finishes, and you begin to see a trend.

This leads us to the second observation: if we are deliberately creating individual

components in the manner that they would be built in the real world, then why don’t we
do it comprehensively, and model the entire building as it would be built? When done in
this manner, a design model approaches a proto-building, emulating many real life aspects
of buildings in construction. After working with BIM for several years, many architects
find themselves modeling in ways that don’t necessarily make sense if 2D representation
is the end-goal. But the goal is no longer 2D representation – instead, knowing with some
certainty how a building works in 3D is invaluable even if it appears many aspects will not
make it onto a sheet of paper,

Though it may take some effort for designers to re-examine how construction objects
should best exist in a BIM environment, this new awareness can only prove to be beneficial
in the long run, because it prompts architects to learn even more about components they
are using all the time. In our firm, and in several others I know of, we already strive to
build our models according to accepted industry standards. For example, projects are
broken into ‘worksets’ according to the UniFormat system of Substructure, Core and Shell,
Interiors, etc. Furthermore, when we create BIM models, we try to mimic construction -
place foundations, columns, beams, and slabs on beam, then wrap columns, and then have
finish walls separate from enclosure walls. We do this so that we might better understand
the process of construction we are initiating.

© 2008 John Tobin, AIA, All Rights Reserved 6

 Originally published: May 2008

There is considerably more for architects to understand before we attain full protobuilding,
but as far as possible we are moving to shadow construction activity as closely as the
current software will allow. What this means then is that a segment of the architecture
profession is moving beyond representation, and embracing a proto-construction
mentality, carefully but inevitably.

Detailed BIM models.

One Model? Several models?

The impetus for this article was triggered by a simple exchange of viewpoints. Not too
long ago, I attended a joint Architect-Contractor AGC workshop where contractors were
becoming resigned to the view that the architect’s BIM model was a throwaway, that it was
often useless for the purposes of construction and that they would simply build a new
model for the construction phase of a project. This is a dire development in my opinion.

To the extent that it is true, architects are missing a tremendous opportunity to re-align the
efforts of the architect and the contractor. The architect’s model should not become a
throwaway. It may be a basic first effort that can be developed later, a first prototype, but
it should not have to be thrown away.

Separate design and construction models would be disastrous for many reasons
1) It is a waste of time to duplicate modeling work,
2) We may lose important design intent in the shift from one model to another,
3) It would reinforce the notion that the architect’s design work essentially lacks
value, and
4) It supports the idea that design can proceed separate from construction
implications.

One of the reasons the Architect’s BIM model is being viewed as a throwaway may be
related to current software shortcomings where BIM applications cannot yet deconstruct
an assembly into the trade-by-trade subcontracting tasks that contractors need. This is not
just another software glitch, it precipitates a difference of vision between architect and
contractor and for that reason we need to work with others to overcome this kind of
obstacle. An excellent source of discussion on the limitations of architect models to

© 2008 John Tobin, AIA, All Rights Reserved 7

 Originally published: May 2008

construction teams can be found in numerous posts at http://bimx.blogspot.com,

especially some excellent early posts on wall composition.

If BIM 3.0 is our ultimate goal, we will want to keep our focus on what will be possible
when these bugs are surmounted. As long as architects continue to remain the primary
contact in the design process, we still have the opportunity to make that first BIM model.
Whether it will become the final BIM model depends on how seriously we take it as a
proto-building.

Conclusion:
Contractors have understood the value proposition of ‘virtual construction’ for several
years, and architects can assist them in maximizing it. Architect’s models - based on
representation - are not yet conveying important aspects that contractors’ BIM models are
expecting to find. So why not? Part of the answer is due to limitations of the tools we use.
But it is also that we are not yet embracing the full potential and implications of BIM, and
its ability to emulate building. This may be because we’re not expected to do it - but it may
also reveal a hesitation whether we should shift beyond our role as ‘representers’.

Yet, another factor is that we need some input on what to model, and how to model it as
we are designing, so contractors can use it. One of the great hopes for the recent, ongoing
(…and ongoing…) National BIM Standard (NBIMS) process is that it will clearly articulate
a course on how BIM models can become ‘virtual buildings’ with value throughout the
entire construction process. Just as the previous National CAD Standard was a set of
graphic representations that codified the construction document phase for 2D CAD work,
the NBIMS will hopefully propose a set of standard conventions suitable for proto-
building. In the interim, a growing number of firms, and some large agencies like GSA,
have started to lead the way and are pursuing a variety of ad hoc BIM model standards.
Many of these standards give us a peek at how we can begin to build protobuildings which
can emulate certain construction processes.

This is a great and exciting time to be an architect. Digital proto-building presents us with
an opportunity to fashion our designs into significant, valuable building prototypes that
offer better control and feedback over the design outcome, and to heighten our
contribution to the construction and operations process. Our chance for success may be as
much an issue of attitude as it is technology. What can help to propel this endeavor is a
vision of BIM as more than just an extension of an architect’s documentation drawings, but
to understand it as a legitimate form of proto-building in its own right. To BIM is to build.

End

© 2008 All Rights Reserved.

John Tobin is an architect and Principal at EYP Architecture & Engineering. He resides
in Albany NY. He spearheads BIM implementation for EYP’s five A/E offices advising on
projects using Revit Architecture, Structure and MEP packages.

© 2008 John Tobin, AIA, All Rights Reserved 8