Saturday, February 20, 2021



A request I received this week:

I am an avid reader of your grevity Blog. I really loved your lollipop trees and used them in my projects till my third year of architecture college and I seem to have lost them and your link is also not working. Is it possible to get a working link so that I can use them in my thesis too?


So I have reworked the lollipop tree family and will provide a link to the new version lower down.


The original post is here        


And below is a screenshot from a file I assembled for a presentation I gave in 2017 at BiLT EU in Aarhus, Denmark. 



I’ve been using Revit for more than 15 years now and one thing that fascinates me is how often I can come back to an idea like this and improve on what I did before.  So inevitably when I started on this post I found myself rethinking how the parameters work.  But even before that, I had to remind myself how in all works. 


The result is a family with ten types, grouped according to 5 basic shapes. 



In practical use the end user will probably want to simplify this down to 4 or 5 types, and maybe tweak the parameters to emphasize the differences between the shapes.  Height is an instance parameter so you can scale the shapes up and down at will without having to create new type names. 


The whole idea of the lollipop tree is to keep things simple and abstract.  It’s useful for campus style projects, large housing developments, urban design studies.  So having too many types is just counter-productive.


How you expose parameters to the end user is always important, but difficult to get right under time pressure.  I like to use short names with intuitive meanings.  Not saying I always succeed.



Under the hood, there are formulas, driven by these parameters to adjust the geometry and resize a nested family which displays as a symbol in plan views and matches the circular footprint of the canopy.


No doubt I will revisit this again at some point in the future.  Maybe you will find some useful ideas in the way that I have put this family together.  I’m a great believer in simple generic families, not always of course, but  often enough.



New link!AobWJ3ol0-6RrYByzD01Hd2O62kr9w?e=c8ukZA





Monday, February 15, 2021


 My “studying history with BIM” work goes back many years, but the pivotal moment in terms of public engagement was Project Soane.  That was also the thing that got me deeper into the Classical Orders and how to represent them in Revit, having made a start via discussions with Paul Aubin over the preceding year or so.  Paul and I tend to have “different but complimentary” approaches, which is great.

After that came Notre Dame, and some truly collaborative work, diving into the Gothic tradition which sits between the classicism of Late Antiquity and the classical revival from the Renaissance onwards.  That’s a huge sweep of European history, stretching well into the era of global trade and the industrial revolution which was already underway during John Soane’s lifetime.


Then came a period of parallel styles, mostly “revivals” of some kind, including very strident arguments pitting Classical against Gothic.  Polarisation of opinion is not altogether a new phenomenon although it does seem to ebb and flow.  In recent weeks I have stumbled into the work of Victor Horta, towards the end of the 19th Century.  People were trying to find a way of moving into the future, to find a “style of today” or even a way to progress beyond the notion of “style” as a set of formal conventions.  One of these days I will get back to looking at the Arts & Crafts movement which was part of that search for a new direction.  But right now it’s the turn of Art Nouveau, a much more lavish and hedonistic approach to generating form.

All the same, there is often a deep attachment to ideas we would associate with Arts & Crafts, or even modernism.  Looking at the window grille by Hector Guimard which I photographed in the Victoria & Albert Museum in London in 2017, I was fascinated by the way he used standard mild steel sections, executing simple cuts and bends, just as blacksmiths had been doing for generations, but generating wildly organic shapes, as if by magic.  So this is not just willful nonsense, as you might think at first sight, but an attempt to work within craft traditions while heading in new directions.

Back to Baron Victor Horta and the Maison Hallet in Brussels.

I decided to tackle the cloakroom window next to the street entrance.  I had used a placeholder family here.  Just a cut opening, drawn in a wall-hosted template and assigned to the Windows category but with no timber framing of glass, and not picking up on the subtle inflections of the stone facing.  I am assuming this is limestone, carved by hand.  It could have been faience (cast ceramic) but I don’t think so in this case.




I have to convert an opening cut into a void extrusion.  This is because you can’t mix the two in the same wall-hosted family and I need some stepped rebates to take this family to the next level.  Luckily we can copy-paste the same sketch lines used for the cut opening to define the void extrusion.

  • Select the cut opening
  • Edit
  • Select all the sketch lines
  • Copy to clipboard
  • Close the sketch
  • Delete the cut opening
  • Create void extrusion
  • Paste aligned to selected levels
  • Choose ref level
  • Close the sketch.


The void will probably not be cutting the wall.  So you have to use “Cut Geometry”  Also it may well be using the centre-line of the wall as its Work Plane, so Edit Work Plane and specify “Back” not sure why that’s the name of the work plane that is locked to the face of the wall, maybe because it’s the back of an object that sits on the face of the wall. Don’t expect templates the date back to the earliest version of Revit to always work the way you might have wanted them to.

Now it’s cutting right through the wall, but that’s only because the wall is 150 and the extrusion defaulted to 250 (6” & 8”) When I bring it into the project the wall will be 500mm thick so I will have a recess.  Sometimes that’s what you want, and you can always add a parameter to control the depth.  Either way we need a reference plane parallel with the wall.  Lock the void extrusion to that, and lock the reference plane to the inner face of the wall.

That takes us back to where we started from, but with the crucial difference that Revit will allow us to add further voids.  First I modified the shape of the main extrusion to pick up more of Horta’s subtle curves.  Then I added a shallow void extrusion to cut some delicate rebates in the top edge of the lower openings.  These have the effect of tricking the eye into seeing vestigial capitals atop the columns that divide this lower portion into three parts.  It’s a clever little detail that I hadn’t noticed before, one of those little bonus observations that come your way when you put in the effort of trying to recreate someone else’s work.  That applies whether you are drawing it by hand, modelling in Max or Sketchup, or going “full BIM”.  But the BIM pencil is the most demanding of your attention and offers deeper insights in my experience.



One small aside.  Shapes that taper away to nothing can be challenging in Revit.  That’s why the “triangles” in my sketch end with a radius rather than a sharp point.  Just stay alert for this possibility when you get an error message saying Revit can’t make that geometry.  Revit doesn’t like tiny little short lines, and it doesn’t like tiny little thin slivers.  There are always limits, you just need a plan for adapting to that reality.

My third extrusion was a big rectangle on the inside half of the wall.  When I added this I pulled the outer void back so these two meet in the middle.  That’s because I wanted the inner sill to be slightly higher than the outer sills.  I’m thinking about rain water.

None of this is parametric.  I could add a certain amount of parametric behaviour later if I thought it offered some value, but right now this is a one off and flexing the size is going to be a nightmare with all these subtle curves.  Maybe that’s a bit strong.  It’s going to consume quite a lot of time and effort which doesn’t seem justified for current purposes.



So the two solid extrusions to represent wooden frame and glass are also free of constraints and parameters.  Just practicing eye-balling in these tricky shapes.  It’s all good practice.  Drawing with the mouse using Revit’s little palette of tools.  I placed a little groove to represent the rectangular outlines of the inner frames, which could be hinged or fixed.  I really don’t know and I’m not too concerned right now.  Much more interested in taking a crack at the external grille.

This has something in common with the one by Guimard that I showed at the beginning, but Horta is just working with flat bar.  It’s more delicate and the fixing rivets are becoming more significant in adding to the organic feel, as if they are little buds on some climbing vine perhaps.  I could spend a lot longer capturing this, and adding the ones on the entrance door, but there are lots more windows, plus balcony railings.  I had to decide how far to take this exercise, the purpose of which is to increase my understanding of Horta’s work. 

So… I wrapped up my day with an image that combines shaded and rendered views in photoshop with a bit of masking and “coloured pencil” filtering, shared that to LinkedIn and signed off.


Friday, February 12, 2021

BIM is the new pencil

 an article for Middle East Architect, published in 2013


Fragments of ochre found at Blombos Cave suggest that drawing is almost as old as our species, Homo Sapiens.  A thin sprinkling of both two and three dimensional works spans the archaeological record since then. Among the most famous are the lion man carving and the remarkable cave paintings of Altamira and Chauvet. Visual thinking is deeply embedded in the collective subconscious.

 Architecture is also very ancient. We could trace it to mammoth bone huts from the depths of the last ice age, but it is more conventional to start with the emergence of urban life in the Middle East some 7000 years ago. We don’t know what drawing techniques the architects of the White Temple of Uruk used, but we can trace the trace the growth of geometry from ancient Iraq and Egypt to Greece and Rome.   Vitruvius gives us our first detailed account of the skills and knowledge an architect should have and his work was closely studied by architects of the Italian Renaissance who also benefited from ancient knowledge transmitted via the Islamic world.

 Brunelleschi and Alberti were two such architects who played a key role in the development of perspective drawing around 1400.  This can be seen as a major breakthrough in drawing technology based on innovation in both hardware & software.  The concepts of a horizon line and vanishing points lie clearly in the software realm, but drawings by Durer and others show a variety of ingenious physical devices that were used during the unravelling of these mysteries.

 Many of these historical giants were also sculptors, and physical models have long been part of the design process.  In fact the distinction between drawings & models is a false one.  Energy models, financial models and freehand sketches are all simplifications that help us to focus on key issues.  We deal in abstractions.  We capture ideas in diagrams that isolate some aspect of reality and allow us to see design problems more clearly.  If you study the concept sketches of famous architects you will detect a remarkable facility for condensing a complex array of information into a single image.  This ability to organise information and to 'see the wood from the trees' is perhaps the key skill of the architect.



During the eighteenth century, a group of French Mathematicians unlocked the secrets of descriptive geometry enabling the techniques we now know as orthographic projection.  The current mania for '3D' has blinded us to the importance of this breakthrough.  Reflect for a moment.  Our cities can be built, our projects tendered & priced, because of the use of orthographic projection: a fully dimensioned set of plans elevations & sections.  The real breakthrough of Building Information Modelling (BIM)  is that it combines the virtues of a 3d model with the explanatory power of orthographic.  Once again it is a question of condensing vast complexity into a single 'diagram'.


The last century or so has seen many advances in drawing technology: blueprints, parallel motion machines, tracing paper, drafting pens, photography, letraset, Xerox.  Architects have always been split in their reaction to these innovations.  Some have seen the dangers of becoming distracted from their essential purpose by slavish devotion to gadgetry.  Others have grasped the new potential enthusiastically.  It might be helpful to examine the role of photography in Fine Art.  Initially there was controversy, heated debate, experiment, uncertainty.  Ultimately the fuss has died down and photography has become another tool like a pencil or a palette knife.  You can use it to create fine art.  You can use it in a purely pragmatic way.  You can use it to create the most dismal rubbish imaginable.  It all depends on the power of the ideas you have and your ability to use tools as a natural extension of your hands, eyes and brain. 

And so we come to BIM.  I have skipped over CAD which I regard as a false start, a fleeting detour along our journey.  BIM is a much more architectural way of thinking and working.  It is holistic, volumetric, integrated, dynamic.  There is a danger that it will become bogged down in the blinkered thinking of 'business speak'.  Far too many magazine articles about BIM fall into this trap.  You hear lots about “Return on Investment”, “Change Order Management” and similar technocratic terms.  But the potential is there to transform our industry, if only we can find our proper roles within the emerging digital world.





Here at GAJ, we began our BIM journey in earnest around 8 years ago.  From the start, our motivation was simple.  It was a new and better way of drawing.  We had found ourselves living in an increasingly fragmented workspace, using multiple computer programmes to deliver our designs.  The multi-tasking environment of Windows 95 had made this possible, but it did not solve the coordination dilemma.  Someone spends 3 days constructing a section through the site, passes it to a second person to beautify in Photoshop , then as this work nears completion a third person moves everything around on the site plan.  Disaster. 


 Don't make late changes, you might say (especially if you are an engineer) but change is the essence of the iterative design process.  We make these drawings, abstractions, models of reality, in order to see the design problem more clearly.  We pause and reflect; internalise the working process of the past week; see through the mist to a much simpler solution.  We suffer frustration with clients who suddenly change their minds when half way through detailed design.  But isn't this part of the same phenomenon?  You can't firm up on a budget without building a financial model.  You can't make clear decisions about the building you want until you see it taking shape.  Everything is inter-related.  Enter the “single building model.”


There is a well-known diagram for the adoption of new processes.  It describes a roller-coaster ride through excitement and disillusion before emerging on the “plateau of productivity”.  You cannot reach that plateau unless you board the BIM bus.  No pain, no gain.  For us it was a rather lonely bus ride for the first few years.  We were among the early adopters in the Middle East and it was difficult to find sub-consultants willing to commit.  Perhaps you are familiar with the arguments. “Yes we have the software, but we are short of operators, maybe next time” or “the programme is too tight on this project.”  And so on. 





Right now we are entering the collaboration phase in a big way.  Engineers have got the message and Quantity Surveyors are starting to bite.  Clients have become aware, but may lack the experience to make realistic judgements about what BIM can deliver at this stage of the game.  Contractors are mostly out of the loop, but some have grasped the potential of BIM, especially for large, complex projects with the potential for thousands of clashes.  Suppliers and manufacturers are often totally unaware of what BIM means but we need them on the bus too.  The next few years are going to be very interesting.  We could learn a good deal from the UK, where the right kind of government intervention is persuading all sectors of the industry to engage in the process of developing collaborative digital processes.


 At Godwin Austen Johnson we are fully committed to BIM.  It is, quite simply, a better way of drawing.  We have used BIM processes in dozens of projects ranging from master planning to single villas, traditional resorts to modern apartment blocks.  But BIM means collaboration and we are acutely aware that the full benefits to clients and end users will only come when the industry as a whole has embraced BIM wholeheartedly and invested the time and energy in making it work.  That's an exciting future, and we aim to be part of it.


The Middle East is still very much in love with the dream of progress: bigger, faster, more iconic.   BIM is seen as a management tool for streamlining processes, giving a competitive edge.  Nothing wrong with that, but perhaps we need to look a little further.  Architects are generalists.  Our role is to manage teams of specialists, to coordinate their work while keeping an eye on the bigger picture and managing the expectations of clients.  We are trained to think out of the box, to redefine problems, to distinguish the wood from the trees.  We are also trained to consider the best interests of society at large.  This is at the heart of the professional ethos which architects still embrace despite the pressures of commercialisation.

Recently I have been reading a book by William Rosen on the history of steam power.  He talks about the role of visual thinking in science & technology.  “Any inventor's moments of insight ... are primarily visual.  Pyramids, cathedrals and rockets exist not because of geometry, theory of structures or thermodynamics, but because they were first a picture ... technology has a significant intellectual component that is both nonscientific and nonliterary.”  He goes on to suggest that the inventor’s insights are not just visual, but also tactile.  Leonardo's hands were as important as his eyes.  "the beginning of the eighteenth century (brought) an unprecedented enthusiasm for scale models" 

Drawing is a most remarkable activity.  Somehow, hand, eye and brain come together in a seamless process of creation and communication.  In a social context it has immense power.  Think of a group of people sitting around a table with sheets of tracing paper, cycling through design possibilities.  What is the BIM equivalent of this kind of rapid brainstorming?  Drawing helped to initiate the technological chain reaction that we call the industrial revolution, to create a society based on growth and innovation.  This has brought many benefits and improvements to human society, but in recent years we have also begun to see that continuous growth and change cannot be sustained indefinitely.  Are we caught in what is sometimes called a “progress trap?”  If so, how can we escape?  Can we design our way out?  Can the new drawing technologies that are currently emerging help us to see past the current progress trap, to visualise a more sustainable world?



One of the key ideas in the debate about responsible business practices is engagement.  How do we help clients become more engaged in the design process?  Would it help if, instead of standing back and asking us to achieve the impossible, they participated in hands-on design workshops where ideas can be transformed into walk-through simulations in a matter of minutes? 

There is a danger of getting carried away by science fiction fantasy, of worshiping clever gadgetry for its own sake and losing sight of the bigger picture.  Architects have always used diagrams to abstract and simplify, to isolate the essence of a problem, to visualise the underlying shape of a solution.  This is exactly how we need to use new drawing tools: the single building model, reality capture, social networks, cloud services. 

Building Information Modelling means many things to many people and we do not doubt its potential as a powerful management tool, but it is also the new pencil.  We see the software and hardware tools that are coalescing around BIM as continuations of the age old tradition of visual thinking.  The world needs fresh ideas and architects are uniquely placed to provide them.  We bring well-honed visual and tactile skills to the table, embracing new tools that help us to engage our clients in the design process.  Our goal, as always, to serve society at large and build a better world.


Andy Milburn is an associate at Godwin Austen Johnson with special responsibility for BIM strategy and Revit implementation.  He has had a varied career ranging from bricklaying in the UK, via curriculum development in Zimbabwe to architecture in the Middle East.  He is a well-known figure in the global Revit community, speaking recently at the BIM Show Live in London and the Revit Technology Conference in Auckland.  You can follow his blog at