Don’t put the cart before the horse

Every now and then, at a first production meeting, I hear something that reminds me of this:

I’m sure you’ve heard it too: “Let’s get (or build) a revolve.” “We have a cart we used last year.” “We have that staircase from [name a show].” “We can use periaktoi.” And similar comments.

Granted these ideas are all intended to be helpful, but it’s so easy to home in on one of them and let a piece of scenery drive the set design. I’ve mentioned this one before: a couple of years ago I was at a first meeting for Les Miz at a large community theater, and of course the barricade came up right away. Someone immediately suggested borrowing a 12′ revolve from another company, and several heads nodded in agreement. At which point I brought up two things: one, a 12′ barricade on a 40′ wide stage would not have looked very impressive, and, two, we didn’t need a revolve just because the barricade turns around. So they looked at me and asked what I would do instead, to which I replied “I don’t know yet; this is our first meeting and I haven’t started on the design.” Not what they wanted to hear at that moment, but it had to be my answer.

In the end, they had a much more impressive barricade that turned just fine, and they didn’t have to deal with a revolve. You can read about here, at The barricade in Les Miz.

It’s fine to keep stock pieces in mind, and they can certainly help with the budget or the schedule. However, it’s so much better — in terms of the audience experience — to create a solid, compelling design first, thinking about the story and the characters, and then (and only then) look for stock pieces that fit into the design. A staircase, for instance, built for one show may not work for another show due to size, style, or some other consideration. Ditto for doors, windows, and lots of other items, and the same holds true for furniture and accessories.

By the same token, popular and frequently built scenery pieces — like periaktoi — are not always the best solution to a design problem. What I’ve seen happen often (too often) is that these items become a construction project that drives everything else. You can end up deciding to build three or four periaktoi right up front, before considering the overall set, and then find yourself painted into a corner: “Okay we have them, now what do we do with them?”

If you want to consider periaktoi as problem-solvers (which they are), and not as short-cuts to designing a set (which they are not), you may want to check out my post on them, at Periaktoi: an ancient solution that still works.

Take your time. Study the script, the characters, the period, and all those other things that make up the story, and come up with a physical environment that supports it in the most creative, theatrical manner possible. Then feel free to see what pieces you have that fit in perfectly or that can be modified.

I love that photo above, BTW. It’s like the horse is breaking the fourth wall to ask us, “What’s wrong with this picture?”


Revolves don’t have to be round

My first year in junior college we built a revolve for a show. It was a huge affair with three different scenes, one of which had folding panels to change the setting quickly. In the center of this revolve was what we called the “delta unit,” which was a triangular space that led to each of the three scenes. I kid you not: it was easy (and known) for people to get lost in the thing.

That was my first revolve, and of course it was round. And, of course, going by the textbooks, we built up the floor around it so the whole surface would be at one level and the turntable wouldn’t be visible. It worked beautifully, but, being the first revolve most of us worked on, it also gave us the impression that revolves have to be round.

Not so.

Your typical revolve, out of a textbook, is a round turntable in the middle of a built-up area. The backs and sides of the set, and the structure on it — and of course the dimensions — can be anything, but the basic idea is the same:

Revolve 1

Now and then we see a revolve without the built-up area around it, which looks like a round platform:

Revolve 2

Nothing wrong with this, as long as it fits visually into the overall design.

But it doesn’t have to be round. Years ago I designed a small, non-round revolve for Equus, which was about six feet on a side:


But it wasn’t square: the edges were ragged, so it looked like a rough wooden platform:

Revolve 4

We can also take this idea a bit further and stack several levels on it, creating a sort of revolving hill or rock formation:

Revolve 5

And here’s another way to look at a revolve, from a set I did for Shrek:

Revolve 6

For this one, we used half of an existing 12′-diameter round revolve and built a square piece on the other side to create a tilted stand for a huge book:

Shrek book

So the revolve had the fairy-tale book on one side and Fiona’s bedroom on the other. You can see more on this on my web site, at Shrek.

Then, of course, there are those pieces that we think need a revolve, but really don’t. The barricade I designed for Les Miz, for instance, was a free-standing structure that turned all the way around but didn’t sit on a “typical” revolve. This photo shows half of it, from the “rebel” side. The complete unit was twenty feet across:

Barricade 2

You can read more about this unit at Problem solving: the barricade in Les Miz.

Revolves can be wonderful tools to help tell a story, but the real trick is to think in terms of what the story needs to say, instead of what the physical piece of scenery “should” be.

Get to the root of the problem

When I came up with the idea for this post, I thought maybe I should send it to The Huffington Post instead of placing it here.  🙂

As I’ve said in previous posts, I often find the solution to a problem, or the answer to a question, in a totally unexpected place. Case in point: the United Airlines passenger who was dragged off the plane in Chicago some weeks ago. Did United ever go back and take a good look at the real cause for the incident?

Was it a real “doo-doo happens” emergency? Was it a scheduling error? Did somebody drop the ball? Did the software crash? Did somebody misunderstand a supervisor’s instructions? Was it a last-minute knee-jerk reaction by a manager somewhere? Did four crew members get sick at the same time in Louisville?

Obviously, I don’t know if they ever found that first domino. But, putting aside my own personal experience with this airline, I would like to think that they, or anyone else, would want to dig back, if only to learn something from it and try to figure out how to prevent a similar incident in the future.

Now, putting this in the context of set design and tech. I ran into a similar issue years ago when I designed the sets for The Odd Couple and the show immediately following, Play It Again Sam. I wrote about how we used the same set for both shows in Two shows on one set (re-tasking a set) so I won’t repeat it here. But we had a very simple problem right up front — that would have been very simple to solve — which created other problems down the line. Unfortunately, even though we did discuss it afterwards, the real reason for the problem was not addressed: it wouldn’t have been “politically correct” to do so. So similar problems continued to occur.

Briefly, the structure called for a series of platforms to go right up to the back wall of the stage. From there, flats and platforms would angle downstage to form a box set, which would terminate just behind the proscenium wall on both sides of the stage. I designed it that way to simplify the look, as the entire set would be behind the proscenium arch. But, because we were using the same set for the second show (which required several secret panels), the position of the walls relative to the platforms, steps, escapes, and other elements was critical. Plus, of course, there were parts of the theatre building which could not move.

The problem began when the TD left two inexperienced carpenters to begin assembling the platforms while he went to lunch. By the time he came back, a number of platforms were up and connected, but there was about a four-inch gap between the back edge of the platforms and the back wall. This was due to a decision the carpenters made, and by not following the instructions. I pointed out the gap to the TD, but he didn’t want to correct it, as he felt it would take too long and not really create any problems later.

To make a long story short, pushing the set downstage by those four inches created several problems. Because of the angles and steps, the secret panels had to be modified, which meant revising some of the architectural details. But the panels could only be shifted so much, which now meant it was awkward to go through a couple of them. Then, because the set could not now end behind the proscenium arch, I had to re-design the DS termination, which involved building and painting new flats. And of course, that four-inch gap at the back wall had to be closed to avoid anybody stepping in it.

The final score: several new pieces had to be built, painted, and installed, and the gap had to be closed, all of which took longer and cost more than it would have taken to move those platforms when the TD came back from lunch.

We discussed it afterwards — the managing director, the TD, and me — but the message didn’t get through. Nothing was learned from the experience. And yes, similar things happened several times afterwards.

Back in college and grad school, and certainly at a professional scene shop where I worked for six years, the TD or his equivalent would have made those carpenters take the platforms apart and place them per the drawings. That is, after reading them the riot act.

But so often nowadays we don’t seem to be willing to correct errors or even to admit them. Which is too bad, because understanding why mistakes happen is a great way to learn from them and reduce the chances of their happening again.




The punch list

I know… I know… the first time I heard this term (very soon after I started working in the architectural field)  I had a couple of interesting mental pictures too. And of course they were based on that too-often-asked question, “Whose fault was it?”

But a punch list is nothing more than a list of things that need to be completed before the project is handed over to the client. Generally, when a building contractor tells a client that the job is “substantially complete,” he (or the architect) also provides a list of items that still need to be addressed. Maybe a piece of carpeting keeps pulling up, or a door needs to be plumbed, or an A/C diffuser needs to be moved… or someone hit a wall with a cart and the drywall needs to be repaired. Making these lists is a standard practice in the industry, and what it does is make accounting for these items a formal, expected part of the process instead of a last-minute finger-pointing exercise.

I’ve never seen this term used in theatrical companies (although I’ve tried to introduce it several times, with often limited success), but, in reality, I’ve found that the TD or shop foreman, or designer, often does make up a list before opening. The main difference I’ve noted is that this list, and the response to it, is often seen as a negative (again, “Whose fault was it?”) instead of just being part of the process of building and installing a set.

I don’t know where or how the term “punch list” originated, but an article on it in Wikipedia says that the term probably comes

“from the historical process of punching a hole in the margin of the document, next to one of the items on the list. This indicated that the work was completed for that particular construction task. Two copies of the list were punched at the same time to provide an identical record for the architect and contractor.”

(I mention this in case you want to introduce the concept in your own company but people object to the term “punch”  — which I’ve seen happen.)     🙂

I’ve written here several times that, if a set construction project is managed well and given enough time, the set can be pretty much ready to go by First Tech, which is usually the weekend before opening night. If this happens, then Tech Week, for the shop, becomes just a time to go through the punch list and take care of all those little details that’ll make the set look finished and become a valuable part of the story.


Flats are over-rated

In September 2016 I presented a Set Design Mini Boot Camp at the Educational Theatre Association’s conference in Las Vegas. The three-hour workshop covered mostly how professional set designers go about creating a set, what the process is, how long it takes, and so forth, with lots of photos of real sets. One of the things I covered was urban legends often believed in theatre (one of my first posts in this blog), and one of those was that “all sets are made of flats.”

In a way, it (sort of) makes sense that this is a common belief. Many tech theatre classes either start out by having the students build a flat, or they do so early in the course. Because of this emphasis, it stands to reason that the students may think this is a very important subject, and therefore may believe that flats are an essential part of any set.

But they’re not.

A flat is a solution to a problem, a device invented many years ago to represent a flat surface, such as a wall, on stage. It’s just not practical to build sets, especially touring sets, like real buildings, and more especially if the scenes change during the show. But flats are light, easy to move and store, can be re-used, and travel well. They can become blank walls, or have openings for doors or windows, or even be cut to represent trees and other objects. They’re versatile, but they’re not always needed.

For instance, here’s my set for Marc Camoletti’s Don’t Dress for Dinner, which takes place (according to the script), in a renovated French farmhouse, a structure dating back a few hundred years. It’s the interior of a building, so flats are a good solution.


Here’s another way to use flats, in a production of Macbeth I did for a high school. The director wanted a non-literal set (no castles) with a little bit of a steampunk feel. I don’t have any good photos, so this is a SketchUp model.


In this case, the large back wall was a perfect candidate for lots of stock flats. The tapestry was rented, those blocks behind it were painted like rusted metal panels, and the sconces were purchased from a place that sells steampunk lighting fixtures. There was a matching wall opposite this one, but the whole center part (where the blocks are) was a rear-projection screen.

Now here’s another show, Spewak and Porter’s Kiss Me Kate. This one has a lot of locations and usually a lot of scene changes that tend to slow down the action, something the director and I agreed to avoid right up front. The set uses no flats at all, except for a few placed against the back wall to look like… well… flats stored in a theatre. This was the backstage set (under work lights):


And this was the onstage set, again under work lights:


That drop was lowered in full view of the audience by a cast member dressed as a stagehand. In fact, all the changes, and the few pieces that were carried or rolled in, were handled by cast members playing the crew. Lighting played an important part in this one to separate onstage from backstage, but the show flowed from one scene to another, and it didn’t use a single flat. The back wall of the stage, however, was painted to look like old, grungy, dusty  brick.

Here’s another example, My Way, A Musical Tribute to Frank Sinatra, by David Grapes and Todd Olson. It would have been really easy to stage it in a night club, complete with flats on all three sides. Instead, I designed a back wall that consisted mostly of open space backed by curtains.


Lighting was important in this one too. The curtains changed color several times, the sconces on either side of the center opening had three different lighting configurations, and a full moon was rear-projected onto the stage-right curtain during one of the numbers.

Flats are a good solution to specific problems, but they are not “the building blocks of a set,” and many very interesting and useful spaces can be created without them.


Problem solving: slam the door, don’t shake the wall

Every now and then a director will want a character to slam a door as he enters or leaves a room. It can be a very dramatic moment, but too often the entire wall shakes when the door gets slammed. Not good.

Fortunately, there’s an old trick to help prevent this, and it consists simply of having the door flat and the framed door be separate units and not attached to each other.

Here’s the back of the door flat. I’m showing a “Broadway” style flat, although this works with “studio flats” too:



And here’s the self-contained door unit, complete with front molding and a brace:


Here are the two pieces in place:


And here’s a cross-section thru the door and flat, looking down:


All you really need to allow is about a quarter-inch gap, all around, between the framed door and the door flat, including the molding. Depending on the set design and the sight lines, you can have one or two braces on the door unit, secured to the floor with a sandbag or bracket (although of course you don’t want to create any tripping hazards). The wall flats are then braced as they would normally be in your case.

The same idea can be used for a window or any other framed opening. It’s a simple trick and will help keep the audience focused on the story and not get distracted by a shaking wall.


How long does it really take?

One of the most common misconceptions I’ve run into over the past thirty-odd years, mostly in non-professional theatre companies, is that “everything always happens at the last minute.” Although I’ve run into some companies where this idea is so ingrained that you can’t break through it, I’ve also seen a few companies that did change the way they did things and were very successful at it. And all it really took was sitting down and looking at the last few productions and realizing why things happened at the last minute.

For the most part, the problem was just a matter of not allowing enough time for what had to get done in that particular company’s realities.

As I’ve noted in other posts, creating a set is just a process like any other process. There’s nothing magical about it: you do this first, and then you do that, and then something else. Each step takes a certain amount of time, but it also has built-in variables, one of which is Murphy’s Law. Say you schedule five weeks for building the set, but the primary carpenter comes down with the flu up front or has to work elsewhere. Or the painter has to finish another show that’s running late. Or the director requests a major change. That’s where things slow down, but, since opening night doesn’t change, it all comes crashing together near the end.

I worked with a theatre company some years ago that had this ongoing problem, and here’s what we did. This company mounted four to five large productions each year, had a full shop and staff working Monday thru Friday, and hired mostly professional designers.

First, we realized and accepted that the set always did seem to get finished at the last minute, even though the schedule always called for it to be complete by first tech.This is where the ugly specter of “whose fault was it?” comes up so often, and we decided to not go there. We just looked at the reality.

Then we looked at typical production schedules over the past few years and noticed that the production process for each show took three months, from the fist meeting to opening night. So, by the time the director and designers were on the same page, and the basic design was complete, we were down to two months from opening night.

At that point the set designer and TD always got into the argument about when shop drawings would be ready. The designer needed time, since he or she was working on other shows at the same time, and the TD wanted them as soon as possible. So the usual compromise was to agree on when construction would start, and to request the shop drawings on that same date.

But here’s the problem. Construction couldn’t start on the day the drawings arrived, because the TD needed time to go through them, order materials, get the crew and the painters lined up, and so forth. So, when Murphy’s Law expressed itself, which it always did, there was no room to maneuver. Suddenly construction would start a week (or more) late and run well into tech week.

And we noticed that this happened over and over again.

So we looked at each other and asked, “why does the production schedule run three months from first meeting to opening night?” And the only answer we could come up with was, “because that’s the way it’s always been.”

And we decided to change it. We started at opening night and worked backwards: one week for tech, five to six weeks to build and paint, one week for the TD to go through the drawings, three or four weeks for shop drawings, and so forth. I covered this in my post on the project schedule, so I won’t repeat it here. By the time we agreed on all this, the production schedule for each show had expanded to six months instead of three. Because we did several shows a year, the schedules would overlap, but this was not a problem since each show would be at a different stage of production.

It made a huge difference even with the first show on the expanded schedule. Now there was time to do the work properly and to account for Mr. Murphy.

Then came the second part of the solution. Because the designers’ contracts had never specified the dates on which each deliverable was due, there was nothing to fall back on. So we added due dates to the designers’ contracts for each phase of design, based on the particular production schedule. A little negotiation usually had to happen here, but it was resolved by the time the designer signed on, and the dates could be put on the production schedule for all to see. By the way, designers were hired prior to the six-month production period; in fact, they were usually all hired the previous season, during the planning process for the new season.

It took a bit of work to make all this happen, but the results were amazing. Now the production schedule could include dates for everything else, including lights, costumes, props, and sound, and we all went over it at the first production meeting, with everyone around the table at the same time.

The primary — the most important — thing that made this work was realizing and accepting  how long it really took to actually build one of our typical sets (what our audiences had come to expect from us) given our shop and our crews. After some discussion and reminiscing, we agreed that six weeks was adequate for building most of our sets, but smaller ones could possibly be done in five. But we kept it to six weeks throughout, figuring that the worst that could happen is that we would finish early.