Home  /  BIM and ARCHICAD   /  Four BIM Flavors   /  Self-awareness vs Environmental-awareness: A working definition of BIM

Self-awareness vs Environmental-awareness: A working definition of BIM

The dividing line between CAD and BIM

What is CAD (Computer Aided Design)? What is BIM (Building Information Modeling)? What is just 3D modeling? Where is the dividing line between CAD and BIM? This division matters because BIM is about effectiveness and adding value: things that architects desperately need. Understanding what differentiates BIM from older methods, and understanding what BIM promises will help us be better architects—we’ll be able to harness more creative power. On equally grand terms, if my view of BIM is understood and accepted by the software developers, then amazing things will happen. Because how can we push the boundaries of what’s possible if we disagree on where the boundaries are and what’s at stake?

Ready? Let’s start with what differentiates CAD and BIM and then end up somewhere much more interesting.

CAD is a computer based drawing made of unintelligent lines and fills. An amalgamation of basic vectorial elements are placed together on a field to imitate hand drawings. What is a created is a representation of a window, door, wall, etc. It’s a window because we point to it and say window; because everyone who looks at the image agrees on the graphic symbols. But if one were to go into the program and select the window or door, what would be highlighted would be a bunch of individual lines, polylines, fills, etc. They could be grouped and that group could be labeled ‘window’, which is a step towards BIM, but this is pretty basic and not BIM.

Many of us made this first tentative step as we got tired of the traditional AutoCAD way before we left that program for good. Back in 2004 and 2005, my last 2D CAD drawings (done in Vectorworks and then AutoCAD) were filled with groups and instances of those groups.

BIM on the other hand is different. Fundamentally different. BIM is a collection of intelligent, often parametric elements that have basic understanding of what they represent and how they relate to their environment. A window element in BIM knows it’s a window and knows that it lives in a wall. Furthermore that wall knows it contains a window and can tell you (if queried) various pieces of information about what it means that it contains a window (surface area minus openings, number of openings, etc.). This collection of intelligent elements could be a 3D model, a 2D model, a 1D model, or 4D, 5D, etc. And that intelligence covers a range from very dumb to hyper detailed and aware.

By this working definition of BIM, the difference between lonely little bim and CAD is a big one. And from the narrow view of 3D models, a 3D BIM created in ArchiCAD, Revit, etc. is also clearly different from a surface model built in a non-BIM program (instead of pointing to lines and calling it a window in a 3D surface model we are just pointing to a 3D mass and calling it a window). These lines of course are blurring as traditionally non-BIM programs like SketchUp are getting smarter and officially crossing the threshold into BIM (more on that soon).

Blurred Lines

I think we can describe the difference better though. Or more specifically I think we can look at CAD vs BIM from an idealized point of view, a perspective that sees CAD and BIM as two points on a larger continuum that also includes hand drafting and processes yet discovered. If we think about the intelligence of the elements created while drafting or modeling, then all our tools fit on a nice graph. Physical methods of design and production are as dumb as you can get. In the graph below they are all the way to the left. A pencil line is dumb. A piece of basswood is an inanimate hunk of dead tree. Regardless of the dimensions (one, two, three, or four—assuming some incorporation of time or movement) hand sketching and physical model building, no matter how wonderful, are as dumb as you can get (though [spoiler alert] in the future we might build physical models with smarter materials). Now as we start adding intelligence we enter the realm of CAD and then BIM: elements can be grouped, copied, pasted, queried for information. In broad strokes I think this helps us understand what is CAD and what is BIM based on how smart the elements in the program are. Parametric objects (things able to react to various numerical inputs, etc.) are smart. Fills and hatches grouped together and labeled “chair” a bit dumber. A wall that can recognize a window being added to it: smart. A wall that has to be manually cut to receive a window: dumber. This works, but only to an extent. Where is the dividing line? It seems like there is a tricky gray area between the two in this definition. Plus as the dimensions increase, where the CAD/BIM divide gets even more confusing. It could be argued that the more dimensions there are, the “smarter” an object is; that at some point an element with so many dimensions starts to act very BIMish. If that’s a term. That seems logical, right? If an element has so much data in it (ie, lots of dimensions), then it must be BIM, right? Right? Maybe. But maybe not.

Intelligence of Elements vs DimensionsActually, I believe the number of dimensions in an object or file really doesn’t have much baring on whether something is CAD or BIM. A 3D surface model is just CAD (albeit 3D CAD). A really powerful spreadsheet can be BIM, even if it is just numbers. In fact the graph above is essentially a one dimensional line from dumb to smart: from sketching to CAD to BIM, though sketching isn’t labeled. So we could argue that all that matters is the intelligence of elements, mostly. That it’s not the amount of data that defines if something is BIM, but something else. In fact I think even the obsession with parametric elements is a red herring in the quest to understanding BIM. Having lots of toggles and check boxes is great, but is that inherently better than an element which is less parametric? We’ll return to that question in another post. For now I’ll defer to Rob Jackson.

@jbacus @SketchUp @BuildEdge It’s not really about parametrics. It’s about relationships of the data to each other.

— Bond Bryan BIM (@bondbryanBIM) March 1, 2014

It’s not the data that matters but the utilization of the data. You can have every bit of information in an object, but so what? If that information is disconnected to other bits of information it is useless. Thus instead of intelligence, I think the proper term should be awareness. Which actually is in line with my arguments from this post about Building Intelligence Models. The dividing line between CAD and BIM is awareness. Self-awareness and environmental-awareness. Here’s a graph:

Environmental-Awareness vs Self-Awareness

My working definition of BIM: a system of self-aware and environmentally aware elements.
In the graphic above there is once again a tricky gray area between CAD and little lonely bim as self and environmental awareness increases, but that dividing line really doesn’t matter. If you look at something and aren’t absolutely sure if it is CAD or little lonely bim, then it might as well be CAD because the elements don’t have enough awareness to start harnessing all the benefits inherent in BIM.

Self-awareness vs Environmental-awareness

All things, physical or digital, can be described on the graph of self-awareness and environmental-awareness. This means we can use these same two axes to examine not just our architectural tools but also our built works and the products within them. The Internet of Things promises to raise the self-awareness and environmental-awareness of our world. A smoke detector that knows the difference between a false alarm and danger, or what to do when its battery is low. A door that can tell you when it is open or closed via your smartphone. Or even a beer can that can tell you if it’s cold or not (okay that’s not the IoT, but same point). BIM provides that for our architectural processes. A Door in a file that knows that it’s a door, that knows its price, that knows what to do when it is in a CMU wall or a stud-framed wall.

Once we see this connection between how BIM something is and the future of our built environment, the magic really starts to happen. In BIM circles we already talk about BIM for Facilities Management and some people (me) get excited about augmented reality. Seeing both our production and design tools and our end products moving towards increased awareness (self and environmental) there is an obvious theoretical convergence. If a door knows it’s a door in a BIM file, at some point does that door know when it gets built? And does that imaginary digital door become one with the physical one that is built? And if so, does that mean the physical door can thus track things like how and when it’s opened, where it is touched, its condition, its need for maintenance…

We talk about BIM for FM but then stumble when we think about the facilities crew that will never bother to learn a BIM program or who currently have good-enough processes for replacing light bulbs and knowing what paint color the walls in each room are. We acknowledge this disconnect between theory and reality—that BIM is great but perhaps adds to much complexity to a good-enough solution. Now think again of increased Object awareness, both self-awareness and environmental-awareness. And think about digital Objects and built physical objects. As all that merges into one, things start to automate. The door says it needs repair, the light bulb messages that it has just died, the load bearing column asks to be inspected. Perhaps you think this is all decades away. You are wrong. It is closer than that. And it could be much closer if we consciously design this future.

If you’re on board with my vision outlined above, here’s a handy flow chart to help you answer the burning question of “is this BIM?”

HOW TO TELL IF IT IS BIMSo this was supposed to be the fourth part in a series of posts, but it’s coming a year late. And I have expanded on these issues so much since January 2013. The first three posts were Why you failed at BIM (you were impatient), There are Four BIM Flavors, and Primary Benefits of BIM. Each post looked at one aspect of understanding a major shift in our industry that is quickly gaining steam. The first post looked at struggles with BIM adoption. The second clarified what the different types of BIM are. The third helped focus discussions on the value of BIM. They all pretty much set the tone for about 50% of the blog last year. If you haven’t read them yet, seriously? How is that possible.

Subscribe to my blog to read more about the tricky world of being an Architect in the 21st century: Shoegnome on FacebookTwitter, and the RSS feed.


  • March 5, 2014

    Nice article although I have a couple comments as I think you really skirted around the issue of 3D models vs BIM. Programs like Solidworks, Catia, SolidEdge & Pro-E for years have been creating 3D “smart” objects, that react to their environment, have ton of data attached that you can extract, know where they are in relation to other objects, can tell if there are clashes, can be used for motion simulation and is some respect are smarter than Archicad, Revit or Vectorworks models. See this video of Creo (Pro-E) from 3 years ago where collision and constraint are automatically detected without the user have to define them.



    But I wouldn’t exactly call them BIM. I think it’s more about which data is attached or used.

  • March 5, 2014

    Thanks. Just wanted to say that the article was thoughtful and it is problem one of the best attempts to date to define what BIM is or define Building Information Models. Maybe skirted wasn’t the right word, as yes they do fall somewhere on the your graph. Although my point is that if you are going to start discussing where or not you a 3D surface or solid can become BIM by adding data to it, I think you should include other more intelligent modelers as is some ways they are ahead of standard BIM platforms. Even Maya with it’s inverse kinematics and physical environment simulator has a lot intelligence built in, but most people wouldn’t classify them as BIM models. While I partially disagree I can see there argument as it doesn’t have the right data.

    • March 5, 2014

      Yeah and crazy I didn’t realize you were from Seattle also.

  • March 11, 2014

    I get your point, but to me it’s all ‘CAD’, meaning ‘Computer-Aided Design’….and ‘Drafting’ is also part of the deal. I’m 48, & now use them all. I see it all as the evolution of CAD. Just another set of tools to add to the ever expanding toolbox. In 10-20 years, someone will be scoffing at the term ‘BIM’……I’ve always seen it as a marketing buzzword, myself.

      • March 11, 2014

        Not to me. Refer to original comment….heehee
        So when will the name change to ‘ArchiBIM’?

  • November 27, 2014


    I would categorize it in rough terms as follows:

    *CAD =Geometric objects+ Operations + Transformation + Properties (System and/or User Defined) + Display + Management (layers, groups, negative space, X-refs, etc).

    *Parametric Modeling = CAD + Association (relationships) + Constraints + [some come with automation].

    *BIM = Parametric Modeling + AEC domain specific [Properties, Associations, and Constraints].

Post a Comment