Editing Workflows

Conveyor belts, computers, and the new order of things
Download the PDF

What are the benefits and hindrances of different editing workflows?

An exegesis submitted in partial fulfilment of the requirements for the degree of Master Arts Screen: Editing

Australian Film Television and Radio School

2024

Key Words

Animation

Animation historian Dr. Maureen Furniss defines animation as “a broad range of practices in which the illusion of motion is created through the incremental movement of forms, displayed sequentially” (2016, p. 12). I will narrow this down and focus on two types of animation:

  1. Cel animation, so called for the celluloid sheets on which elements were combined and photographed (Okun, Zwerman, 2021, p. 671); and
  2. CGI animation, an abbreviation of computer-generated imagery, also confined for the purposes of this paper to illusions created before virtual photography.

Editor

Editing practitioner and educator Dr Karen Pearlman says “the editor is the creative collaborator with an expert knowledge of movement of story, emotion, image, and sound” (2016, p. 234). Edward Dmytryk, a practitioner of a previous generation, identifies an editor’s “keen sense of timing, a compulsion to seek out the scene’s hidden values… and a mastery of the technical skills needed to bring all these talents to bear on the film” (1984, p. 2).

Non-Linear Editor

Non-linear editors, or NLEs, are computer softwares for film editing, so called to distinguish them from the linear nature of editing on previous technologies such as film or tape. For example Avid Media Composer, often simply called Avid. This is the designated NLE for masters students at AFTRS.

Visual Effects

Visual effects, or VFX, is a catch-all phrase that the experts can’t agree on. The industry standard guidebook for visual effects supervisors notes that “even the Visual Effects Society (which ought to know) hasn’t come up with a satisfactory definition” (Finance, Zwerman, 2010, p. 3).

In-keeping with my working definition of animation, I will be using VFX to mean illusions created after photography. This was the accepted definition before the introduction of CGI, whereby “if a shot required some sort of treatment in postproduction and rephotographing on an optical printer, it was called a ‘special photographic effect’” (Finance, Zwerman, 2010, p. 4). Computers have blurred these lines, as will be discussed.

This is distinct from special effects, which are created during photography.

Workflow

The Visual Effects Society defines a workflow as that which “defines a procedure, or series of operations, through which a task is performed, and a deliverable produced” (Okun, Zwerman, 2021, p. 703). This can be broken down further into “a series of departmental workflows, beneath which are artists workflows, and beneath those are task workflows” (p. 703).

This paper is largely focused on departmental workflows (standard practices within editing) and artist workflows (my personal ways of working).

Chapter 1: Introduction

“Live-action films shoot first and edit later—in animation, you begin to edit first, then ‘shoot’ later” (Leland, 2021).

At first glance, the editing workflows of live action and animation couldn’t be more different. One is linear, the other iterative. The two have markedly different deliverables, and historically required their editors to do entirely different jobs. Yet, in researching their differences, I found that thanks to computers, these editing workflows are growing more and more alike.

By questioning the benefits and hindrances of each, I have come to the conclusion that in the digital age there is nothing to stop an editor from taking the best elements of each and combining them into a hybrid workflow. That is, nothing except for history, habit, and the hard work that inevitably must come with trying something new.

This narrowed my focus down to three smaller questions. What are the legacies of film editing which are a hindrance in the digital world? How did the advent of computers benefit editing workflows? What happens when you try to edit a live-action film like an animation?

As I tackle these questions, Edward Dmytryk’s observation of an editor’s “compulsion to seek out the scene’s hidden values” (1984, p. 2) will be significant, particularly for live-action editing. Likewise, Dr Karen Pearlman’s idea of “editing thinking” (2016, p. 232) has proved a helpful framework for understanding how creative collaboration might function in animation. This is the notion that “the intuitive process of sensing, hypothesizing, and realizing structure and rhythm” (2016, p. 232) is not confined to only the editor, but rather what the editor is expert at.

Even though my research and analysis is confined to departmental and artists workflows, I will of course need to reference the broader stages of filmmaking. For the sake of simplicity, I’ll be defining the three stages as follows:

pre-production = planning

production = photography

post-production = assembly

The remainder of this introduction will establish my methodology, the relationship between technology and workflows, and the evolution of my own practice as it relates to digital editing. I will then address the benefits and hindrances of different editing workflows in three chapters:

1. Live-Action Workflows

2. Animation Workflows

3. Hybrid Workflows

Methodology

I will address my research question using creative practice-based research, the process by which “practitioners become practitioner-researchers and use their knowledge of and/or experience in practice” (Batty, Zalipour, 2024, p. 2). I will draw on my studies as an editor over the past two years, and incorporate reflections on my creative practice in that time.

This creative practice includes work on short films Interlude and Solo un poco Aquí, an internship at Cheeky Little Media, and general coursework that I undertook at AFTRS. Most of my written reflections will be drawn from personal records kept at the time of practice, as indicated.

Digital writer, researcher, and creative practitioner Lyle Skains writes that “reflection… can be an unfortunately fallible method” (Skains, 2018, p. 87), which I do acknowledge. Thus, my reflections will be supported by reviews of scholarship on editing workflows and contextualised by scholarship on technology and collaboration. I will further position my research by including references to a number of case studies that further illuminate my research question.

Skains further notes that practice-based research is still impossible without personal reflection (2018, p. 88). Batty and Zalipour agree, writing that “personal and professional creative experience intertwine with academic enquiry in creative practice research… While arguably all researcher backgrounds have a bearing on their disciplines, in creative practice research it invariably shapes the questions being asked” (Batty, Zalipour, 2024, p. 5). To that end, I will be including short autoethnographic narrative moments in each chapter, where I believe my background is relevant to the questions I am asking.

In broad strokes, my background is this: as a teenager, I taught myself to edit on iMovie, the NLE that came pre-installed on our new family computer. My earliest filmmaking obsession and first foray into animation was Lego stop-motion, but once I figured out how to download YouTube videos, there was no turning back. Many hours were spent recutting scenes from my favourite live-action films, and it was here I discovered how much a story could be shaped and reshaped in the edit.

I’m also an extrovert. This will become relevant later.

Technology & Workflows

Changes in workflows can often be traced back to changes in technology. I found this to be true of my own experiences with Avid.

Avid and I got off to a rocky start. When I started at AFTRS, I had experience with a range of NLEs, but I’d never even heard of Avid. The learning curve was steep. My workflow at the time was very drag and drop, with minimal use of keyboard shortcuts, and by the end of last year I had pain all up my right arm from overuse of my mouse.

In the new year, I replaced my mouse with a tablet and pen. Using my cursor now feels more like drawing, plus I reconfigured the controls so I can use the pen to scroll in any direction (e.g. up and down through lists of video files, or side-to-side in an editing timeline).

With one hand busy holding a pen, however, my other hand was left to navigate the whole keyboard on its own. The few keyboard shortcuts I had been using were very spread out, so the natural next step was to reconfigure my keyboard settings, so that my left hand can reach all the essentials without moving across the whole keyboard. Many of the Avid shortcuts are designed to encourage a much more pre-meditated workflow of only placing into a timeline what you need, in order, frame accurately. I adopted these, letting them change my workflow.

The simple act of changing from a mouse to a tablet had a chain reaction on my personal workflow. But whilst I finally feel able to be creative in Avid, the technical aspect is still not intuitive for me. This is a problem, because editing as a practice is both creative and technical.

In examining this problem, I have found that learning the history of Avid can be very helpful. My journey with my new tablet is an example of how technology can impact my own way of working. Next, I’d like to look at how changing technology can impact standard practices within editing. To do so, allow me to set the stage of the history of Avid.

Reel to Unreal

“The advent of computer-generated imagery (CGI) represented a change to narrative cinema as seismic as the introduction of synchronised sound” (Kinder, O’Steen, 2022).

My struggles with Avid, as it turns out, have a lot to do with tensions between legacies of film editing and the digital world. These tensions have arisen not because workflow changes can be traced back to technological change, but because after the advent of computers, existing workflows didn’t change.

For me this realisation started when a teacher made a throwaway comment in class about why the software uses “bins” to store media rather than discard it. My peers and I were thrilled by this new piece of trivia. Our heartfelt collective oooh seemed to surprise the teacher, for whom this fact was either obvious or inconsequential. No doubt for him, a student of the film era, it was.

Embed from Getty Images

Bin is bottom right, below stored film. Source: An editor works on a 35mm commercial with a Movieola machine, Soho, 1966. [Photograph], by Walter Nurnberg, 1966, Getty Images.

A bin in Avid for Solo un poco Aquí fine cuts.

Avid was designed to replicate the tools in a real cutting room as closely as possible, so that editors could easily make the transition to digital without needing to change their workflows. Yet I’m of the generation who take for granted the ability to undo mistakes instantly. Cutting film, on the other hand, “was destructive. Undoing splices to change the cuts damaged the film, so the more preplanning the better” (Kinder, O’Steen, 2022, p. 17). This is why Avid’s keyboard shortcuts are designed to encourage a much more pre-meditated workflow. There is a keyboard shortcut for undoing mistakes, but then why the steep learning curve to think like a film editor? On one hand the pre-meditated approach has made me a more disciplined editor, particularly when cutting scripted dramas. Yet on the other hand I have often mourned the loss of easy dragging and dropping when cutting documentaries with no predetermined structure. I find that linear workflows are effective when a story has a beginning, but less so when not.

For a young George Lucas, digital filmmaking was but a dream (Baruh, 2020). In the 1980s he invested heavily in the development of the EditDroid (Bouzereau, Duncan, 1999), an early NLE that mimicked the analogue editing workstations of the time. The EditDroid was an integrated desk and computer: footage was stored on laserdisc or VHS, and inserted as needed (Klingeren, 2014). In the 1990s it was sold to Avid Technology, who had found a way to store footage locally on a computer, and the modern Avid Media Composer was born (Bouzereau, Duncan, 1999).

Reception to NLEs was mixed, to say the least. Ed Catmull, in charge of Lucas’ computer division at the time, said of the EditDroid that “while George wanted this new video-editing system in place, the film editors at Lucasfilm did not” (2014, p. 30). In 1999, Lucas finally got to cut a feature film on Avid for the first time, and he was very positive about the results:

It allowed me to actually create shots and scenes in the editing room, rather than just cutting them. I could move things around, cut people out of one shot and put them in another, change sets, or take a scene from one location and put it in another location. I could completely reconstruct and rewrite the story in the editing process” (Bouzereau, Duncan, p. 137).

For me, starting at the beginning of any project doesn’t come naturally. I prefer to follow ideas, and if my first idea pertains to the middle of a project, I start there. This is true for my editing, my songwriting, even my grocery shopping. So why do we continue to use workflows which encourage linear editing, when computers are decidedly non-linear? What is helpful and what is habit? What are the benefits and what are the hindrances? This tension between legacies of film editing and the digital world is at the core of the debate.

Chapter 2: Live-Action Workflows

In live-action filmmaking, it is standard for an editor to join a project in post-production, having had no involvement up to that point (Pearlman, 2016). Photography happens first, then assembly, always in that order.

This workflow is born out of the realities of working with reels of film, a medium that dictates a methodical, meticulous process. After photography, film would need to be chemically developed before the results could be seen, let alone edited to tell a story. I myself have sent off many a roll of 35mm film to be developed, only to discover an important photo didn’t work at all.

Photo shot on 35mm at AFTRS: a light leak meant everyone was cropped out of the photo except for Sophia, one of our producers on Solo un poco Aquí.

Technology has changed so much, and yet live-action editing workflows less so. That is what I will be investigating in this chapter, in much the way that Dr Pearlman phrases her own investigations in a recent addition to her textbook Cutting Rhythms: “Do linear production models reflect the constraints of the old tools rather than the opportunities of new tools for capture, cutting, and exhibition?” (2016, p. 231) More precisely, I am asking What are the legacies of film editing which are a hindrance in the digital world?

Assembly Lines

When Henry Ford created the first assembly lines, he had an impact on more than just the automobile industry (Sutton, p. 41). Ford himself predicted this in 1926: “Within a comparatively short time I believe the practice will be so general in industry that it can be made universal” (‘The 5-Day Week’, p. 12). He was right, and I can spot two elements in particular that are universal in live-action filmmaking.

The first element is conveyor systems, which journalist John Younger described in 1928 as “one of the most spectacular of the Ford achievements” (p. 488). Conveyor belts would carry work-in-progress cars through the factory, and workers would make their modifications without ever taking the car off or stopping the conveyor belt. These systems set a controlled and predictable pace of work (Younger, 1928), allowing for tight schedules and budgets.

The second notable element is what is now called deskilling; instead of employing one person who could do lots of things (i.e. build a whole car), Ford employed lots of people who could do one thing each (Kirby, 2010). Again, Younger noted of the factory:

“At first sight it is a bewildering maze of machine-tools seemingly unrelated to one another. Closer observation reveals that the plant consists of a multitude of small shops, each concentrating upon one product. Pistons, for example, are made in their own department and nowhere else. Never does the drilling of a piston-pin hole stray into the drilling department for connecting-rods. Each article is kept distinct, even at the expense of some duplication of operations, and much resulting economy is affected” (1928, p. 488).

Doesn’t that sound like every movie behind-the-scenes documentary you’ve ever watched? Some confused presenter leading a camera operator through a “bewildering maze,” one department at a time?

Live-action filmmaking is not unlike an assembly line. A film sits on a theoretical conveyor belt as it passes through pre-production, production and post-production. The more distinct the departments, the more predictable the assembly line’s outcome and its “resulting economy” as Younger puts it. This is of great benefit to those monitoring a film’s schedules and budgets.

Fix It in Pre

INT. FILM STUDIO – DAY

DIRECTOR

(pointing to a ladder)

Would you be able to take that out of the shot?

VISUAL EFFECTS SUPERVISOR

Oh yes, absolutely.

A beat. No-one moves.

DIRECTOR

No, we mean now.

When Christopher Nolan told this joke at an awards night for The Visual Effects Society, it went down a treat. “If that joke didn’t play here, it would never play anywhere,” he sighed with evident relief (The Movie Rabbit Hole, 2024). Post-production are accustomed to hearing a flippant “eh, we’ll fix it in post” when something isn’t working, typically on set. But when we don’t solve problems as early as possible (like a ladder in shot that shouldn’t be there) we risk making our own lives harder later on.

This year, I discovered for myself how difficult it is to send the conveyor belt backwards. After a late screening, we realised that our class project Paralysis needed a shot changed from a wide to a mid. It was just one shot, how hard could it be?

The process quickly got out of hand. Changing one shot started a domino effect, which affected the rest of the picture. Because a first pass of the colour grade had been completed, I couldn’t simply re-deliver a finished edit. I had to deliver individual elements to be slotted into the colour grade. The whole process took a couple of weeks, instead of the few hours I expected.

It should have been a simple matter of asking Avid to spot the difference between the old and new edits of the film, and generate a list of the changes. These changes could then be applied to the colour-graded version of the film. Yet instead of information, Avid kept spitting out error messages like “(MissingTC) Opt. #1.”

Long story short, it was discovered that the error messages meant some shots were being treated as “optical shots” and left blank. This harks back to the way that VFX shots pre-CGI involved “rephotographing on an optical printer” (Finance, Zwerman, 2010, p. 4). VFX shots would be left blank in an edit, leaving space for the completed VFX shot to be added later. But I didn’t want them left blank! I needed data! This legacy of the film era had caused quite the stumbling block in my workflow.

As I wrote in my personal reflections after updating Paralysis, “iterative workflows are very difficult, especially if you don’t intend to be nonlinear from the very beginning” (4 April 2024). As Ford himself said of his assembly line, “unless parts are all made accurately, the benefits of quantity production will be lost – for the parts will not fit together and the economy will be lost in the assembling” (‘The 5-Day Week’, 1926, p. 11).

The solution, then, must be to plan ahead. I have often heard the screenwriters at AFTRS say “fix it in pre.” After all, the writing stage in pre-production is the cheapest of any stage (Okun, Zwerman, 2021). The earlier you can anticipate and address problems, the more working backwards you can avoid. But what about the problems you simply can’t anticipate?

Find It in Post

Problems are inevitable in the world of filmmaking. Ed Catmull argues that if you are able to proactively tackle every issue before it happens, you’re probably making derivative work (2014). Problems must, at some point, be dealt with reactively.

The lot of reacting often falls to the editor, as the first person to see all the footage after photography. Just as an editor’s job is both technical and creative, it can also be helpful to think in terms of both technical and creative problems. But where “fix it in post” applied to technical problems annoys me, I’ve noticed that “fix it in post” applied to creative problems excites me. That, after all, is editing.

Instead of “fix it in post,” I prefer to think of reacting to creative problems as “find it in post,” the act by which we begin to put the puzzle pieces together in our “compulsion to seek out the scene’s hidden values” (Dmytryk, 1984, p. 2). Or as Pearlman puts it, when “the editor’s developed sensitivity to movement kicks in and she writes the last draft of the script in movement” (Pearlman, 2016, p. 234).

It was this desire to more readily find it in post that drove George Lucas’ dreams of editing digitally in the first place:

“My whole focus on filmmaking is as an editor. The script is just a rough sketch of what I’m going to do, and the filming is just gathering the materials – but the editing is how I create the final draft” (Bouzereau, Duncan, 1999, p. 135-136).

Whilst this wasn’t impossible with film, computers, with their undo and drag-and-drop functions, have made this style of reacting to problems exponentially easier.

With our short film Solo un poco Aquí, we always planned to edit this way. It is a post-apocalyptic sci-fi about a young man named Theo. The director Diego and I were both very influenced by the film Arrival (Villeneuve, 2016a), which plays with time to explore similar themes of grief as “an intangible thing,” according to editor Joe Walker (American Cinema Editors, 2017). Walker worked closely with the director in an explorative style of editing where “the clay was always wet, and we could keep tweaking these ideas until they’re right” (DP/30, 2016). They had planned for this in pre-production (Villeneuve, 2016b), so we did too.

The script for Solo un poco Aquí had scenes in the past, and others in the present. We knew we wanted to bridge the two with a match-cut, where Theo falls asleep in one, and wakes up in the other. But we also knew we wanted the opportunity to play with story structure in the edit, so our cinematographer built flexibility into his shot-list. He filmed Theo falling asleep at the end of Scene 3, then filmed match-cuts at the start of both Scene 4 and Scene 5. This gave us room to put these scenes in any order we liked.

Still from Scene 3 of Solo un poco Aquí, not final grade.

Still from Scene 4 of Solo un poco Aquí, not final grade.

Still from Scene 5 of Solo un poco Aquí, not final grade.

Sometimes putting together the puzzle was fun. On one occasion, I drew pictures of each scene on some paper, which Diego coloured in, and we then moved around to our hearts’ content.

“With each paper edit, I would open the Avid timeline, re-order the scenes to match, then export without watching anything. While we worked we naturally found ourselves chatting about The Beatles, marathon training, laughing at funny stories from onset, and coming up with a couple more film resonances as we had ideas with our little paper scenes… Diego really enjoyed the morning (as did I!), said it was “a good exercise” and now has 7 versions of our film (assembly + 6 alternate structures) to take home with him over the weekend” (Personal reflection, 16 August 2024).

On other occasions, it was hard work. Reacting to creative problems is stressful. There were long periods of time where we weren’t sure if we would even like what we found in post.

Contrary to Avid’s legacy function of leaving “optical” shots blank, Diego and I put work-in-progress VFX shots directly into our edit as they came through. On one occasion, we even extended the length of a shot because the VFX was so beautiful. There is space for iterative reactive work within live action (increasingly so as technology evolves) but going backwards entirely is still out of the question. To quote editor Lee Unkrich: “Live action editing is a very reactive game. You have what you have, you have what was filmed on the set, you have to make it work. Which is great, and it’s nice to have that challenge” (DP/30, 2012).

Unkrich started out in live-action editing, before helping to pioneer the role of animation editor by cutting Pixar’s first feature film Toy Story (Lasseter, 1995). As I will cover in the next chapter, this is an entirely different approach, one where if a shot is missing, “we can just create it. So it’s very liberating, and thus the editor gets their hands much messier in the creative process” (DP/30, 2012).

Chapter 3: Animation Workflows

In animation, an editor joins a project during pre-production, and stays through to the end of post-production. Whereas assembly happens after photography in live action, animation is done in reverse. Rather than a find it in post approach where a large selection of footage is sorted through and whittled down, you edit first, then animate only what you know you will need.

Instead of working linearly, in the spirit of Ford’s assembly lines, Pixar Animation Studios prefers to think of filmmaking as an iterative wheel, the “hub” of which is the editing department (Kinder, O’Steen, 2022). In this paper I will focus on Pixar as the model case study, because it was they who created the role of animation editor in the first place.

Pixar’s “wheel” for The Incredibles. Source: Kinder, O’Steen, 2022, p. 27

The role of an editor in this model is equal parts creative and technical – both storytelling and keeping track of the latest version of the film – from writers’ room “all the way to the bitter end of finishing the sound mix and the colour timing,” as Lee Unkrich, editor of Toy Story phrases it (DP/30, 2012). But that wasn’t always the case.

Before the advent of CGI, animation editors didn’t exist. To understand why, it is necessary to look further back through the history of animation editing. This chapter will cover brief histories of editing in cel and CGI animation, plus reflections on my animation editing internship at Cheeky Little Media

Throughout, I will be asking: How did the advent of computers benefit editing workflows?

Synchronised Footsteps

Synchronised sound launched the career of Walt Disney. The Jazz Singer was released in late 1927 (Crosland), and just one year later, Steamboat Willie (Disney, Iwerks, 1928) followed in The Jazz Singers’ synchronised footsteps. Phil Vischer (creator of VeggieTales) puts it best:

“Within twelve months, Mickey Mouse was a national craze. Walt didn’t invent the animated film or synchronized sound, but he envisioned a combination of the two technologies – married to a likeable character and a serviceable story – that audiences found intoxicating” (2006, p. 97).

The job of animation editor did not yet exist, and wouldn’t exist until the advent of CGI. Instead, cel animation was assembled by cutters, purely technical workers who had one job: keep the picture and sound in sync (Kinder, O’Steen, 2022). There was no creative collaboration in the role. It was the job of the director to do most of the “editing thinking” (Pearlman, 2016, p. 232).

The Disney studio grew, ultimately settling into a company workflow which can be broken down into four stages: story pitch, story reel, layout, and animation (Kinder, O’Steen, 2022). Each stage was iterative, to a point. Until all the elements were placed together on the celluloid for photography, a certain amount of flexibility was required to make sure all the pieces of the puzzle would fit together. For example, background painters at Disney worked with watercolour or tempera, because these paints allowed for easy correction (Thomas, Johnston, 1984). Even this could have disastrous effects. According to Epic Games’ Animation Field Guide, “even a minor change in color would send the process back to be repainted by hand, and more detailed changes could mean starting a sequence from scratch with new drawings. Days, weeks, sometimes months of work could be lost” (Bousquet, 2022, p. 12).

Thus a culture was established of finished sequences never hitting the cutting room floor; they had simply taken too much time and effort to be discarded (Nerdstalgic, 2024). This remained true until the mid 1980s when new head of animation Jeffrey Katzenberg personally made some cuts to The Black Cauldron after a disastrous screening (Kinder, O’Steen, 2022). But it was too little too late for both The Black Cauldron, which was a box office flop (Nerdstalgic, 2024), and for Disney.

Cel animation’s linear workflow had grown to suffocate the pioneering spirit that had led Walt and Ub to make Steamboat Willie (1928) in the first place. Paul Wells, animation scholar and practitioner, argues that this spirit of invention is so intrinsic to animation as to define it: “Animation has always sought to embrace new technologies to extend its vocabulary of expression” (2012, p. 15). It seems that in the wake of Walt’s passing in 1966 (Nerdstalgic, 2024), Disney had forgotten this.

Enter computers, Pixar, and the animation editor, whose chief creative responsibility would be iteration at every stage of a project.

Reinventing the Wheel Hub

After helping Lucas’ digital dreams come true, Ed Catmull left Lucasfilm to pursue his own dream of making the first feature length CGI animated film (Catmull, 2014). With the release of Toy Story in 1995 (Lasseter), this was achieved. In the process, Catmull’s new company Pixar Animation Studios also completely reinvented the role of the animation cutter, thereby creating the animation editor (O’Steen, Kinder, 2022).

Before touching on this history, I want to briefly compare Pixar to the two elements of Henry Ford’s factories that most impressed John Younger.

Conveyor Belts

Rather than react to creative problems in post-production, Pixar’s wheel allows problems to be identified early. An iterative way of working inevitably allows for lots of testing of ideas, which is very different from the days of hoping that developed film works out as intended. Catmull himself mentions conveyor belts:

“If all our careful planning cannot prevent problems, then our best method of response is to enable employees at every level to own the problems and have the confidence to fix them… A culture that allows everyone, no matter their position, to stop the assembly line, both figuratively and literally, maximizes the creative engagement of people who want to help” (2022, p. 163).

Deskilling

A certain amount of deskilling is inevitable when growing a company and delegating tasks to departments. What is notable about Pixar, however, is the way in which they upskilled the editing department. Very early Pixar editors had t-shirts calling them the “Richmond Point Cutters” (Kinder, O’Steen, 2022, p. 9). But it wasn’t long before the role of cutter was expanded from keeping everything in sync, to being the wheel hub.

Lee Unkrich arrived at Pixar in the spring of 1994 (Kinder, O’Steen, 2022). Because of his background in live action, Unkrich’s skills extended beyond the technical task of syncing; he was already versed in the kind of creative “editing thinking” (Pearlman, 2016, p. 323) that thus far had been missing in cel animation. He found himself becoming increasingly useful inter-departmentally, and whilst he is credited on Toy Story as editor, he also collaborated on staging and camera work (DP/30, 2012).

In the course of one film, animation editors had gone from non-existent, to adopting the standard practices of a live-action editor, to establishing “wheel hub” status (Kinder, O’Steen, 2022) and being “very very involved, much more than an editor is actually involved in live action,” as Unkrich himself puts it (DP/30, 2012).

Pixar was using Avid, still a relatively new NLE at the time. Version 6.5.1 required thirteen installer floppy discs in 1995 (Kinder, O’Steen, 2022), but even by embracing such cumbersome technology, the pioneering early-adopter attitudes at Pixar had changed careers of animation editors forever.

Creative Collaboration

Visual media educator Volker Pantenburg writes that the editing room “isn’t a place of social interaction but of interaction between images. Much of the time at the editing table is ‘dead’ time, spent winding film material back and forth” (2015, p. 154). But I’m not sure that’s true any more.

It made sense in the era of the cutter to wish to be undisturbed. Work was physically demanding and tedious. Certainly even now, in the time of computers, there have been times when it is unreasonable to keep a director in my edit suite while I focus on a technical task. When cutting Solo un poco Aquí, I would at times excuse my director for any length of time ranging from thirty minutes to half a day. Sometimes it was because I was mixing sound and needed quiet, or at other times I was consolidating lots of new ideas and didn’t want any extra ones. 

Research at the University of California found that after an interruption, the average person will complete two new activities before returning to what they were initially working on (Thompson, 2013). This doesn’t sound very conducive to completing the methodical, technical tasks of an editor. But given that the role of editing is both creative and technical, it stands to reason that the editor should also be where the creative collaboration occurs.

In his 2020 memoir, Paul Hirsch, editor of the original Star Wars movie (well before the EditDroid), observes that “the culture of filmmaking forbids anyone to see the editor’s work until after the director has seen it and made changes… Today, however, advances in technology have somewhat eroded this tradition of secrecy” (p. xi). He cites the iterative relationship with VFX as an example.

This was paralleled in Solo un poco Aquí, but I must admit that when Diego invited the whole team to an early rough cut screening, I was nervous. Villeneuve called Arrival “a movie that was fragile until the end” (DP/30, 2016), thanks to the constant uncertainties of finding it in post. Our film felt the same, and I was nervous to show it to the rest of the team too early in the editing process.

Maybe on some level, I was aware that those who work in a wheel are used to collaborating with editors, and therefore trained in “editing thinking” (Pearlman, 2016, p. 232), whereas a team in an assembly line, like ours, would struggle to know how to contribute constructively at a screening. Or maybe eroding the “tradition of secrecy” (Hirsch, 2020, p. xi) was scary. Ultimately, the risk paid off. But the point is that it was, indeed, a risk.

I’m not advocating for having everyone in one big room just for the sake of it, but personally I do need a balance of time out in the thick of it and time alone in my edit suite. In her writings on creative collaboration for the Harvard Business Review, Leigh Thompson says there is “decent evidence that extroverts make good team members” (2013, p. 74). This innate desire of mine to get out of my edit suite more is what led me to research animation editing in the first place. But there was only so many times I could read that on the Lego Batman Movie “editing was carried out in Animal Logic’s open-plan offices throughout the film’s development process” (Miller-Zarneke, 2017, p. 15). I needed to put the books down, and go try it for myself.

Cutting in Movement

In April of this year, I undertook an internship with animation studio Cheeky Little Media. I was on the editing team for season three of their ABC preschool show, Kangaroo Beach.

To my great chagrin, I discovered that even in the collaborative world of animation, editors can still be locked away in their own rooms. In fact, the editors are the only people in the whole of Cheeky Little not to be sat in the large communal office. By happenstance, my edit suite wasn’t ready when I arrived, and I was put at a temporary desk in the big room on my first day. I was so evidently happy there (and able to stay on top of my to-do list) that I was allowed to stay there for the remainder of the fortnight.

My desk was next to one of the directors, and she was unendingly generous with her time. Whenever she caught me curiously poking my nose around her monitor, she would pause to show me and explain. The storyboard artist who sat opposite me once raised a pumpkin above her monitor to show me her latest vegetative acquisition (I believe one of the 3D artists had grown it and left it on her desk). That wouldn’t happen in an edit suite!

When browsing books on the office shelf, I read that this spirit of openness as standard practice in animation can also be traced back to Walt:

“In most of the established studios, both on the East Coast and in Hollywood, the top animators guarded the secrets of their trade, but at the studio that Disney founded in 1923 there was an insistence on an open atmosphere, with the men learning from each other’s discoveries” (Thomas, Johnston, 1984, pg. 7).

My first job at Cheeky Little was to learn what I now realise is a wheel, although I don’t know if that’s how the studio thinks about it particularly. It felt more like a tennis match. The editor organises dialogue recordings, the director picks performances, the editor turns them into a radio play, a storyboard artist draws storyboards, the editor combines the storyboards and radio play to make an animatic, the director reviews it to give notes, and on it goes.

By the time I had tried these things for myself, I had made my first animatic. An animatic is a full episode of the show, with dialogue, temp music and sound effects, but only using motionless storyboards. From here, the animators begin to introduce movement, slowly replacing the storyboards with animation. While making my first animatic, I hit a snag.

Dmytryk proposes seven rules of cutting. Rule three is “whenever possible, cut ‘in movement’” (1984, p. 27). Dmytryk doesn’t mean cut whenever there is movement, but rather don’t cut without it. This proves very tricky when there is no movement.

In live action, I have gotten used to showing the first frame of movement in one camera angle, then finishing the movement in another. How do you ensure your cut is in movement, when you can’t see the movement? Do the animators know to start movement one frame before a cut? I asked my supervisor, and the response was that because this is an iterative process, adjusting the timing of movement is the kind of feedback an editor would be welcome to give. I was starting to understand in greater detail exactly why an iterative wheel is so necessary in animation, and was really enjoying being even a small part of the wheel hub:

“I can easily see myself enjoying edit assist work in this environment. You’re engaging with interesting WIPs in a way that live action assistants aren’t, that always seemed so menial without being inspiring. In that situation, there’s no-one to pop your head out of your office to annoy or bounce ideas off of. Here, they’re all still working, surrounded by drawings and sketches and people making silly noises and acting things out” (Personal reflections, 15 April 2024).

Chapter 4: Hybrid Workflows

Having investigated live-action and animation workflows, I have found benefits and hindrances to both. In this chapter, I outline the benefits of a hybrid editing workflow. Might not the ideal editing process be one that draws on the most valuable contributions of both live-action and animation editing workflows? Is it possible to develop a bespoke hybrid workflow according to the requirements of a particular project?

In their handbook of standard practices and procedures, the Visual Effects Society supports this notion of a bespoke workflow, stating that “focusing on the artistic goal is essential… Technical and usability requirements should be designed to make achieving artistic goals as simple as possible. Reducing complexity wherever possible is key” (Finance, Zwerman, 2010, p. 707). Ford made “only one model of car, in one factory” (Younger, 1928, p. 488) and nearly went out of business when building a second car required building a whole new assembly line (Kirby, 2010), because the first one hadn’t been built with adaptability in mind.

Are live-action and animation workflows too technically different to be blended? I don’t think so, thanks to the affordances of digital software and NLEs, such as Avid – which has formed a significant part of my research.

What about creative differences? I have found there are none. Dmytryk’s seven rules of cutting apply to both mediums. As Carol Littleon, editor of E.T. the Extra-Terrestrial (Spielberg, 1982) notes: “As a live action editor, imagine my surprise when I discovered that animation editing shares the same dramatic values of live action: story structure, characterization, performance, pace, and emotional impact” (Catmull, 2022, p. i). Live-action editing and animation editing, in our modern digital world, are more technically and creatively alike than I had first thought.

I was able to put this to the test on the live-action short film Interlude, which I approached like an animation editor. The team had already agreed to approach the film in a non-hierarchical, collaborative way, so were very supportive of my experiment. In this chapter I ask: What happens when you try to edit a live-action film like an animation? I explore this question, and discuss my findings, in relation to the three stages of filmmaking: pre-production, production, and post-production.

A Musical Case Study

Interlude is a short live-action film directed by Elder. It is a musical about a grocery store worker named Sam, who never speaks but does sing. Early on, I trepidatiously pitched Elder my idea to cut our film like an animation. To my delight, this aligned with his own preferred way of working. He had recently directed the music video for Ed Sheeran’s Amazing (Elder, 2023), for which he had made a kind of live-action animatic in pre-production, and had worked with an onset editor throughout the shoot. I couldn’t believe it! We both agreed that this workflow would bolster our pre-planning, something necessary to help us stay on top of our tight budget and quick turnaround for Interlude (only a three day shoot). Lesson one: hybrid only works if others are onboard and supportive.

Pre-Production

Fresh from my time at Cheeky Little, I was keen to further practise what I had learned, and make an animatic. This proved to be quite difficult, because Interlude doesn’t have any dialogue. But I did have the musical numbers to work with, plus storyboards drawn by our designer. I stitched these together, feeling distinctly that what I had created was not as polished as I had hoped.

To my surprise, the whole team was excited to see our film “on screen” for the first time. Our writer was encouraged to see that the basics of the story were working, and our composer thought of new musical references that resonated with others on the team. The project felt exciting and doable again. Even though very basic, my animatic served as a morale boost for the team. This would be a recurring theme.

It also made me realise a few things. Below is an excerpt from an email I sent to the director and producer with some of the questions that cutting the animatic had raised for me. The point is not what I say, but how much I say. It was a tad overwhelming, but much better to discover these things in pre-production than on set.

Email excerpt, 15 May 2024.

Production

If you assemble it, they will come.

When our 1st Assistant Director popped over and apologetically asked me to refill the director’s water bottle once again, I joyfully accepted the task, thrilled to be a part of the production process. I found myself doing all sorts of things: dashing through the rain to buy Dr Pepper for our cinematographer, elbowing the continuity person sitting next to me when I heard something relevant they hadn’t, yelling “I got it!” and pulling the smoke machine out of shot. I would then return to my desk and bury my nose in my editing work, still with one ear on what was going on around me.

My onset editing setup was simple. Monitor with the live performance in front of me, and on my desk the laptop where I was assembling the film as we went. I worked with low-resolution footage straight from the camera monitor, which meant all the important information was already watermarked onto the footage, and with only a laptop I could work quickly.

When I did need to focus I could put my headphones on and disappear. But I rarely did. The extrovert in me was ecstatically happy. You may be reading this thinking I was being deeply unproductive. In fact, I’m sure I was cutting faster than usual. I had my animatic open in front of me, slowly replacing the still images with the relevant footage, just like an animator would do.

I wish I had taken a photo of the crowd amassed around my laptop at the end of the first day. I came back from a visit to the snacks table to find at least half a dozen people peering over each others’ shoulders at the assembled first scene. Everyone was very keen to see the film taking shape, and encouraged to see their hard work paying off.

The main actress became my most regular visitor. By the third day, she had basically set up shop at my desk: between takes she would come and sit and watch me edit. Occasionally she would have questions which I would happily answer, about the software or story decisions. Sometimes she would comment on how watching back her performances was helpful in gauging the subtle performance style she was going for. But mostly she just watched. I asked if she’d like to see the assembly edit after she wrapped, and she was ecstatic.

While we packed up the set around her, she sat in the corner and watched the assembly. Some of the timeline was still hand-drawn storyboard, the musical numbers still just master wide shots, but our film was a tangible entity for the first time. She returned the laptop to me with a genuine smile underlining her tired eyes. It had been a long three days, but she could see we had made something special.

Post-Production

This is where my attempt to edit live action like an animation hit a major roadblock. There were, of course, creative problems, mostly relating to coverage, style and story-structure.  The trickiest scene to edit proved to be a songwriting montage in act two. This was meant to be Sam’s moment of decision leading into act three, but the aforementioned problems were making this hard to pull off. Unlike animation, I couldn’t simply request new shots, or amendments to a character’s performance. What we photographed was all we had, and I had to react like a live-action editor. As with Solo un poco Aquí, finding it in post was equal parts fun and hard. It ended up being the scene I was the most proud of.

Note attempt to visualise having an idea.

The scene has now been merged with two other scenes: a dance and a song. Source: Score includes temporary excerpts from Piano Concerto No. 2 (Rahmaninov, 1901)

The scene has now been split into two and moved to separate moments in the story.

Source: Score includes temporary excerpts from Jo Writes (Desplat, 1901)

All in all, deciding to approach Interlude as an animation was a success. I enjoyed being so involved in pre-production, but it also became clear to me that this more iterative approach helped foster greater creative collaboration in our team from script to set. I’m still amazed at what we managed to accomplish in a three day shoot. It wasn’t perfect, and I certainly wasn’t the only contributor, but we collectively achieved something unique.

It was also instructive to discover the limits of this iterative workflow in the live-action context. Reverting back to finding it in post may have felt like evidence that this approach wasn’t working, but with time and reflection, I can see that this hybrid workflow brought out the best of both worlds in terms of editing processes: I had the opportunity to leave my edit suite and collaborate with others, and I was able to solve creative problems in the edit suite.

I have also found myself staying very involved with the project throughout the ongoing stages of post-production, which I didn’t particularly expect. It turns out that wheel hubs pick up a lot along the way, and our producer is often contacting me to help jog her memory about various technical and creative decisions. It’s been very gratifying to be a sounding board for her.

Chapter 5: Conclusion

There is no singular perfect way to edit a film. By asking What are the legacies of film editing which are a hindrance in the digital world? and How did the advent of computers benefit editing workflows? I revealed a bias towards computers. I expected it would be easy to draw a clean line between that which is beneficial, and that which hinders. My early frustrations with the tensions between the digital world and the legacies of film editing led me to expect that learning the history of computers would draw me that clean line.

In truth, by examining live-action and animation workflows, I found more nuance than I expected. But I also found something more surprising: increasing similarities over time between the two mediums, thanks to evolving technologies. Before computers, animation didn’t even have creative editors, just technical cutters. Now, they are creative collaborators engaged in “editing thinking” (Pearlman, 2016, p. 323) from pre-production onwards. Before computers, editors had to wait for film to be developed before post-production could start, but now, for example, I’m able to edit on the set of Interlude giving instant feedback on performance, coverage, and even continuity errors.

One area I haven’t been able to cover in this paper is virtual production, which the Visual Effects Society defines as “a collaborative and interactive digital film-making process which begins with virtual design and digital asset development and continues in an interactive, nonlinear process throughout the production” (Okun, Zwerman, 2021, p. 330). Live-action films are increasingly able to work in iterative workflows where every department can work simultaneously and bounce off of each other’s ideas, largely thanks to developments in video game technology (Baruh, 2020). Whilst this is a fascinating area, it didn’t fall into the scope of my own practice. This technology was used as set dressing on Interlude, but without impacting me and my work. Similarly, Cheeky Little Media has done some work in this area, but not on Kangaroo Beach, so I had no direct involvement. I’m still curious about it, and perhaps opportunities for further research will come along one day.

I am keen to keep exploring this idea of a hybrid editing workflow. But before I conclude, here’s what I’m not saying. I’m not dismissing the importance of the old school. Every day at AFTRS I walk past an old Steenbeck editing desk and its various accessories, wishing I could try cutting on film for myself. Sadly, the Steenbeck is broken.

I’m not advocating for throwing out all this old gear. If reading about it in books has had such an impact on my practice, imagine how transformative actually using the gear could be. As briefly mentioned, being forced to reckon with the contradictions of Avid has made me a more disciplined editor. It has given me a clearer framework for understanding my craft, and taught me the hows and whys of standard practice.

The reality is, the industry as a whole has inherited legacies from film, for good or ill, and it would be foolish of me to take an entry-level job in a well-established company next year, only to refuse to work linearly. That would make me a poor creative collaborator, and certainly not a good employee. Rather, how can I continue to peek my head out of my edit suite out of curiosity and, when helpful, get my “hands much messier in the creative process” (DP/30, 2012), as Lee Unkrich puts it?

I can see that cutting independent films, with small supportive teams like Interlude’s, would be the best way to keep exploring this idea of a hybrid editing workflow. I have observed that it is standard for AFTRS editing graduates to get assistant editor day jobs in the industry, and cut independent films on the side, so perhaps that is a way forward. Ideally with some more on-set editing thrown in!

It’s not that I’m sceptical of tradition, but rather of tradition for tradition’s sake. I’m also not sold on the virtues of change for change’s sake. Catmull writes of Lucas that “his interest in computers began and ended with their potential to add value to the filmmaking process,” (Catmull, 2014, p. 27) in much the same way that the success of Steamboat Willie (Disney, Iwerks, 1928) can be attributed just as much to its “likeable character and a serviceable story” (Vischer, 2006, p. 97) as to the new technology used. Workflow for workflow’s sake would be counterproductive.

Ultimately, moving forward, I just want to tell stories, equipped with whatever editing workflow will serve those stories best. How funny that this all started with something as seemingly insignificant as my computer mouse sending pain up my arm. What are the benefits and hindrances of different editing workflows? It depends on the project.

Reference List

American Cinema Editors. (2017, Feb 10). Editors on Editing: Arrival – Joe Walker, ACE [Interview]. https://www.youtube.com/watch?v=GxHOErwS9Bg 

Baruh, B. (Director). (2020 May 22). Technology (Season 1, Episode 1) [TV Series Episode]. Disney Gallery: Star Wars: The Mandalorian. Lucasfilm.

Batty C., Zalipour, A. (2024). Research, practice, knowledge: introducing the creative knowledges enabling framework. Media Practice and Education. https://doi.org/10.1080/25741136.2024.2384686

Bousquet, M. (2022). The Animation Field Guide [v1.2]. Epic Games.

Bouzereau, L., Duncan, J. (1999). Star Wars: The Making of Episode I The Phantom Menace. Lucas Books.

Buchan, S. (2014). Animating Film Theory. Duke University Press.

Catmull, E. (2014). Creativity, Inc: Overcoming the Unseen Forces that Stand in the Way of True Inspiration. Bantam Press.

Commis, A. (2023 March). Blueback. Australian Cinematographer, 97(3), 10-17.

Crosland, A. (Director). (1927). The Jazz Singer [Film]. Warner Brothers.

Desplat, A. (Composer). (2019). Jo Writes. [Musical Composition]. Sony Music.

Disney, W., Iwerks, U. (Directors). (1928). Steamboat Willie [Cartoon]. Walt Disney Studio.

Dmytryk, E. (1984). On Film Editing. Focal Press, Butterworth-Heinemman.

Douglas, P. (1927). The Modern Technique of Mass Production and Its Relation to Wages. Proceedings of the Academy of Political Science in the City of New York, 12(3), 17-42.

DP/30: The Oral History of Hollywood. (2012, September 25). DP/30: Toy Story 3, director/co-writer Lee Unkrich [Interview]. YouTube. https://www.youtube.com/watch?v=Hw2Dlq5H98A

DP/30: The Oral History of Hollywood. (2016, Dec 6). DP/30: Arrival, Denis Villeneuve [Interview]. https://www.youtube.com/watch?v=v3wV_h5X8Tg 

DP/30: The Oral History of Hollywood. (2017, Feb 13). DP/30: Arrival, Joe Walker (super-sized chat w/ the editor) [Interview]. https://www.youtube.com/watch?v=FjAnVSgxDyc 

Elder, C. (2023). Amazing [Music Video]. Haus Party Pty Ltd.

Finance, C. Zwerman, S. (2010). The Visual Effects Producer. Routledge.

Ford, H. (1922). My Life and Work. Doubleday Page & Company.

Hirsch, P. (2020). A long time ago in a cutting room far, far away…: My Fifty Years Editing Hollywood Hits – Star Wars, Carrie, Ferris Bueller’s Day Off, Mission: Impossible, and More. Chicago Review Press.

Kinder, B., O’Steen, B. (2022). Making The Cut at Pixar. Focal Press.

Kirby, T. (Director). (2010 May 7). Ghosts in the Machine (Season 1, Episode 1) [TV Show Episode] The Genius of Design. British Broadcast Company.

Klingeren, T. (Director). (2014). The EditDroid, Rise and Fall [Documentary]. Manjit Kaur.

Lasseter, J. (Director). (1995). Toy Story [Film]. Pixar Animation Studios.

Laurier, E., Brown, B. (2010). The reservations of the editor: the routine work of showing and knowing the film in the edit suite. Social Semiotics, 21(2), 239-257.

Leland, A. (2021, April 20). The Film Editor’s Role in Creating an Animated Film. Avid. https://www.avid.com/resource-center/the-film-editors-role-in-creating-an-animated-film 

Miller-Zarneke, T. (2017). The Lego Batman Movie: The Making of the Movie. Dorling Kindersley.

The Movie Rabbit Hole. (2024, February 7). “NO CGI” is really just INVISIBLE CGI (2/4) [Video]. YouTube. https://www.youtube.com/watch?v=GdMAEtLrPSc 

Nerdstalgic. (2024, May 28). This Underrated Classic Almost Destroyed Disney [Video]. YouTube. https://www.youtube.com/watch?v=s2IjYRoand4 

Okun, A., Zwerman, S. (Editors). (2021). The VES Handbook of Visual Effects: Industry Standard VFX Practices and Procedures [Third Edition]. Routledge.

Paik, K. (2007). To Infinity and Beyond! The Story of Pixar Animation Studios. Chronicle Books.

Pantenburg, V. (2015). Farocki/Godard: Film as Theory. Amsterdam University Press.

Pearlman, K. (2016). Cutting Rhythms: Intuitive Film Editing [Second Edition]. Routledge.

Pintoff, E. (1998). Animation 101. Michael Wiese Productions.

Rachmaninov, S. (Composer). (1901). Piano Concerto No. 2 in C Minor. [Musical Composition]. Karl Gutheil.

Seastrom, L. (2021, April 2). Lucasfilm Originals: The EditDroid. Lucasfilm. https://www.lucasfilm.com/news/lucasfilm-originals-the-editdroid/ 

Sibley, B. (1998). Wallace and Gromit, The Wrong Trousers: Storyboard Collection. BBC Worldwide.

Skaines, R. (2018). Creative Practice as Research: Discourse on Methodology. Media Practice and Education. https://doi.org/10.1080/14682753.2017.1362175

Spielberg, S. (Director). (1982). E.T. the Extra-Terrestrial [Film]. Universal Pictures.

Sutton, F. (1987). The Ford Foundation: The Early Years. Daedalus, 116(1), 41-91.

The 5-Day Week in the Ford Plants. (1926). Monthly Labour Review, 23(6), 10-14. https://www.jstor.org/stable/41860355

Thomas, F., Johnston, O. (1984). Disney Animation, The Illusion of Life [Popular Edition]. Walt Disney Productions.

VICE. (2014, September 22). Sarah Polley and Great Gerwig on “Frances Ha” – Conversations Inside The Criterion Collection [Interview]. https://www.youtube.com/watch?v=70HXyrodlek 

Villeneuve, D. (2016a). Arrival. Paramount Pictures.

Villeneuve, D. (2016b). Arrival [DVD Extras]. Paramount Pictures.

Vischer, P. (2006). Me, Myself, & Bob: A True Story About Dreams, God, and Talking Vegetables. Thomas Nelson.

Wells, P. (2012) “Validating the animated film. Toy stories, trade tattoos and Taiwan Tigers: or what’s animation ever done for us?”. TMG Journal for Media History, 15(1), 5–24.

Younger, J. (1928). How the Ford Company Gets Low Production Costs. SEA Transactions, 23(1-2), 487-492