Monday, January 31, 2022


 Take the ruthless purism of a Greek temple and dial it back a notch. Now you have the pragmatism of the Romans: a balance between formal beauty and practical convenience.

The Greek temple form, with a porch on all four sides is rare in the modern era. There are examples, but not so many.


The Roman type, with a porch at the front and shallow pilasters to maintain the rhythm on the other three sides is much more common. It works for banks, and for government buildings like the Virginia State Capitol. You can argue functional efficiency while maintaining simplicity, unity of form and truth to historic precedent.


I haven’t attempted to maintain a single family that morphs into multiple buildings just by typing in parameters. Perhaps I will walk back in that direction a little once I have a a dozen or so examples from which to generalise. For now, Roman pragmatism has jumped the blood brain barrier and infected my Revit strategy.  The Virginia State Capitol, but Thomas Jefferson with Charles-Louis Clerisseau (who seems to have been a kind of “gun for hire”: architect, draughtsman, antiquary and painter of picturesque ruins)


So we are ‘saving as’ and adapting each family in an ad-hoc way. It’s an interesting journey.  See how we have sidestepped the religious origins of the Roman Temple archetype. As we move into the industrial age, secular institutions begin to usurp the role of the church. Boundaries blur. The religious impulse finds other ways to express itself. 


The Bank of Pennsylvania is dialed back yet another notch from purity of form. In a way it is more symmetrical, with porches at both ends, but not really. At the sides continuity breaks down. Arches suddenly make an appearance. (not yet modeled) No Greek precedent for that. Latrobe was English but of French Huguenot descent. He grew up in Yorkshire (like me) and was apprenticed to John Smeaton of Lighthouse fame. 

For the bank he nested a mini-Pantheon inside his Roman temple, expressed externally as I a cube with a shallow dome and lantern pushing through in search of light.  


I started with a rational solution to a parametric pediment, extended this into a Roman Temple and varied that theme to become Jefferson’s reworking of the temple form as a house for the abstract gods of democracy.

Latrobe pushed contingency much further in his “house of mammon”, hinting at a hybrid form. Am I engaged in a comparable dance of improvisation?

Mix and match stands in contrast to purism. Picturesque displaces the tyranny of Reason. Romanticism contends with Rationalism.

Faith v Works. Reason v Emotion. Art v Science. How do we integrate these dichotomies? Inevitably the way we build reflects these struggles within society. Architecture becomes a mirror for the human soul.

Saturday, January 29, 2022


 I’m going to come back to the Pantheon family, but for that I will need an “all-purpose pediment” which I can nest into multiple classical temples and churches.

Ideally it should have a small number of driving parameters which can be hooked up to the host family.  We can take that all-purpose starter and tweak it to better suit the host… depending on stuff like “is it Doric or Ionic?” or “what level of simplification do we want in this situation?”

Sometimes good ideas pop into my head just as I’m settling down for the night.  Best to jot things down before they evaporate from short-term memory, and visual ideas usually get captured on my Samsung note using Sketchbook Pro.  I saved a couple of versions of the “thinking aloud” sketching for my “Generic Pediment” family.

The main issue here is how to keep things fairly simple while producing a convincing pediment and allowing for swapping out the profiles to better represent a variety of treatments for different buildings and contexts.  I want this to work for the main roof, a window or door head, a portico ...  whatever.




The core of this family is an extrusion, constrained be various parameters and equality constraints.  The upper sweep has a “pick edge” path so that it turns the corner while responding to different roof pitches.  The lower sweep is horizontal, so it can be hosted on a reference plane. 

I started off simply, labeling the various parameters A,B,C,D,E so I could focus in getting the formulas to work.  We can clean up for later when it’s working properly and I can clarify my thoughts about which parameters to expose and choose suitable user-friendly names.


My profiles rely heavily on equalized reference planes.  It’s one way of reducing the number of parameters and formulas.  The results are not necessarily academically pure, but bear in mind that these are massing models, intended to facilitate rapid assembly of multiple buildings for comparative studies.

In the example below, the entire profile is driven by the “Depth” parameter. This can then be linked back to the "Width" parameter (A) in the host family.



That work was all done on Friday afternoon.  So, I was able to come back with fresh eyes the next morning and clean things up.  The result is a family driven by 3 instance parameters which set the Width, Depth and Pitch of the Pediment, plus 2 more to tweak the Depth of the Entablature and the relative size of the upper profile which sweeps around the eaves and the verge.



The final step is always to flex the family in the project environment to make sure it behaves as expected.  This actually happened before I finished setting up the two “Factors.”  So The flexing process revealed the need to have those adjustments.  A window head will have much chunkier mouldings in proportion to its width than the roof of a large building, for example. 

In the first place, these pediments will not be placed directly in project space.  I’m setting them up for nesting into generic families of Temples and Churches.  Of course, they could also be nested into door and window families for use in full-blown projects.  But first let’s use them to study some Roman Temples.  

Stay tuned.


Friday, January 28, 2022



 Third post of 2022 and continuing with my massing studies of churches and temples.

My collection of Revit powered religious architecture stretches back to Greece and Rome. The origins of the classical style which has been so influential almost everywhere.

My Pantheon model of 2012 is a “linked project” not a “family”, although it is a much-simplified interpretation of the original. So there are subtle gradations of approach here. We can go from the crudest of massing models, all the way to a detailed exploration like Project Notre Dame. 

As an aside ... always impressed how quickly you can generate interesting visuals from an old clunky model just be bringing it into Enscape3d.  (sigh)


What would the simplest version of the Pantheon look like? And why would we do that? What is the value of a thumbnail sketch? And is there a BIM equivalent?

It turns out that the Pantheon can be generated from a Cube that is nested within the inner space, where the corners of the cube, in plan, touch the internal edge of the circular space.  You can then copy this cube to define the footprint of the portico.  The diameter of the circle is going to be the diagonal of the square, and the dome will be the top half of an imaginary sphere which just touches the ground floor level.

Following this logic it was quite easy to create a series of reference planes that constrain the geometry of a parametric family.  This family will scale, based on a single input: the width of that original cube.


I bought a book about the Pantheon on Kindle. So many great insights. Gotta love the interplay between “action research” and “literature review” Spend a few hours, messing with Revit. This stimulates an internal narrative. Endless connections and rabbit holes. Chase up some of these on Wikipedia or Duck Duck Go. Read a book. Collect images. Then back into Revit, or maybe some hand sketches, a morning with brush and paint on canvas. 

Learning all the while.  Searching for deeper meaning.

Anyway, there is a diagram in this book that illustrates this generating geometry of cube and sphere.

There is something about the oscillation between activity and reflection. It’s basic. Theory and Practice. Waking and sleep. Cyclical processes of design or scientific enquiry.

That’s what I am trying to do with my Revit massing families.  It’s quite different from the way I use Family Editor in my day job.  The techniques are similar, but the aims … not so much.  Once again I am trying to use my BIM pencil as a thinking tool, a crow-bar to lever open the packing crates of history.

I intend to develop the family further.  Two more rings to add to the drum, and sweeps for the cornices that define it’s three layers.  Columns within the portico and probably an outer shell for the dome (although this may be harder to control parametrically)

What is the value of making this family scalable?  I haven’t set this as a universal aim for my church families.  To be honest I embarked on this because it seemed like an interesting challenge given the simplicity of the Pantheon’s basic massing.  But having started, I realise that it forces me to understand the relationships of the parts at a deeper level. 

By attempting to scale the whole family proportionally, I make every failure of understanding immediately obvious as the various instances placed in my collections either update correctly or break.  I grew up with the mantra “if you want to understand something, try to draw it.”  Now this has expanded into “try to model it parametrically”

I’m not going to apply this to all my families, only those which appear amenable to this idea, where there seems to be added value on offer.






Thursday, January 27, 2022


 Exploring history is analogous to design work.  You need to work at different levels.  Focus on detailed tasks, but also stay aware of the bigger picture, the social and cultural context, the technologies of the day.

Project Soane and Notre-Dame soaked up hours, days, months, years. Fantastic experiences, but we also need the view from 40 thousand feet. That’s where the massing families come in, and the collection files.  Stand back and look at 40 or 50 different religious buildings.  Group them in different ways.  Develop taxonomies like the relationships of fruit and vegetable from my 2012 studies. 


Temples and Churches are houses for Gods. Those gods are pinnacles of meaning. They sit on top of Hierarchies of value. They signal to us when we are straying from the path. I don’t believe in the god of my childhood. That is part of being modern. But when one God dies, another God is born in some dark corner of our brain chemistry. Because patterns of meaning require structure. We need rules of thumb, preferences, visceral reactions to potential danger. Embodied cognition. Russian dolls of metaphor.

So the Winter Wonderland competition came and went.  My church collection hasn’t progressed far enough to count as an entry, but I started to tie things together, to look at the bigger picture, to think about what it all means.  It’s great to belong to a community: “the church of BIM” perhaps.  Perhaps my role within that church is to emphasise the intuitive, artistic, right-hemisphere approach. 

My good friend Alfredo Medina sent me a copy of Sir Banister Fletcher’s wonderful “bible”.  That was a very moving gesture for me, unexpected but well timed.  Last week, my son shared a picture of his Fiancée in a Hampshire church.  There are so many of these small village churches within driving distance of their home, connections across time to deep traditions of daily life full of human emotion and meaning.

Let’s not lose that spirit in our fervour for all things technocratic.



The sequence of architectural styles first caught me attention as a schoolboy.  At first, I was absorbed by learning to recognize various styles and how they evolved over time.  Later on, I started to draw connections between those styles and the societies in which they emerged.  What was it about Italy towards the end of the medieval period that led to the transformation from Gothic pointed arches to the rebirth of classical style, the 5 orders, reverence for Greece and Rome?

How was it that the changes of fashion began to accelerate during the 19th Century, leading to many different historical revivals existing in parallel, and attempts to completely novel styles for the modern age.  This was not unique to architecture of course.  Music, Painting, Literature also experienced similar fragmentations and searches for novelty.  Most people would agree that this is closely related to the ever-increasing pace of technological change and the social instabilities that this also engenders. 

Some years ago, I started to model Le Corbusier’s chapel at the top of a hill in Ronchamp.  This was clearly an attempt to reimagine the idea of a church without reference to styles of the past.  Is this true?  There are still towers, a Nave, confession boxes, coloured glass.  To what extent is this still a functioning religious focus to a community and to what extent has it become a tourist attraction, as so many older churches also have?

I want to ask these deeper questions in parallel with the more pragmatic, hands-on business of building models.