View profile

How an Alleged "Space Strike" Beautifully Demonstrates Work-As-Imagined Versus Work-As-Done

The Morning Mind-Meld
How an Alleged "Space Strike" Beautifully Demonstrates Work-As-Imagined Versus Work-As-Done
By Jaime Woo and Emil Stolarsky • Issue #3 • View online
Nearly half a century ago, NASA Mission Control lost radio contact with the Skylab space station for ninety minutes, about the time it takes to do an orbit around Earth. The cause wasn’t due to equipment failure, but rather the three astronauts onboard not responding. “Although they could track Skylab as it shot through the sky, each time they buzzed the crew to try to talk to them, they got no response,” writes Atlas Obscura, in a story about the incident.
Since then, a debate has raged on whether it was an accident—as one of the astronauts stated—or instead “mutiny in space”—a narrative the media enthusiastically latched onto. For us, what’s even more interesting is what led up to the incident, because it beautifully illuminates a concept referenced often in resilience engineering: work-as-imagined versus work-as-done.
The silence was interpreted as a form of protest for the intense workload NASA placed on the astronauts:
Skylab was scheduled for early retirement, and NASA, wanting to squeeze everything they could out of the station, packed the new team’s schedule full. As NASA stated in the mission’s press kit, “Earth resources, solar astronomy, medical and other experiments will fill the waking hours of the Skylab crewmen.” There were few breaks for rest, recalibration, or easing into the zero-gravity lifestyle.
Every minute was tightly scheduled, with such little slack provided that there didn’t seem to be time even for washroom breaks:
Every 10 days, they were supposed to get one free day to shower and kick back, but for the first month, they were so behind that they worked right through those days off.
Beyond scheduling, the astronauts were frustrated by poor design that led to a sub-optimal work experience. Tools didn’t fit inside the astronauts’ pockets, creating potentially dangerous situations. As one astronaut recalled: “Every time I raise my right foot to tie my shoelace, I jab myself in the groin with the scissors.”
The Concept: Work-As-Imagined Versus Work-As-Done
Clearly, work as had been imagined by NASA on the ground was very different from work as it was done by the astronauts in space. But you don’t have to be in a space station orbiting the earth to experience this. The gap is commonplace, and has been extensively studied because it has safety-related implications in fields like medicine, aviation, and the chemical industry. 
Erik Hollnagel, Senior Professor of Patient Safety at the University of Jönköping, Sweden, defines the pair as:
Work-as-imagined (WAI) refers to the various assumptions, explicit or implicit, that people have about how work should be done. Work-as-done (WAD) refers to (descriptions of) how something is actually done, either in a specific case or routinely.
You might be thinking: But, shouldn’t the two be the same? If you prescribe how to do the work, as long as someone follows instructions exactly, then there shouldn’t be any problems. Sidney Dekker, Professor at Griffith University in Brisbane, Australia, and founder of the Safety Science Innovation Lab, shares how this is a common, archaic point-of-view:                     
For some, if there is a gap between how work is imagined and how it is actually done, then this is merely a shortcoming in how we manage and supervise and sanction people. Early on in the twentieth century, Frederick Taylor’s ‘scientific management’ attacked work in exactly this way. It decomposed tasks into the smallest bits. It emptied them of meaning or interpretation, until there was nothing left to imagine. All there was, was work to be done. The way work was imagined by the managers and planners, was the way it was done–or to be done, precisely–by the workers. Layers of supervisors would see to that: it was primarily their job to close the gap.   
This idea that work could be broken the solution is if only workers were somehow more—more alert, more knowledgeable, more skilled, more engaged—is pernicious: work works precisely because workers on the front line are correcting and adapting. Again, from Dekker:
Work gets done because of people’s effective informal understandings, their interpretations, their innovations and improvisations outside those rules.
However, how much of work can actually be broken down so cleanly? This is rarely the case when performing complex work. Returning to the example of the astronauts, while making their schedules, how did NASA know exactly how long each task would take: What if something went wrong? Or, alternatively, what if something went unexpectedly right, sparking a new course of inquiry? The astronauts were scheduled to be in space for seven weeks—they would learn things about the work they were doing that would be impossible to imagine beforehand.
The Problem with Conflating Work-As-Imagined with Work-As-Done
An issue with assuming that work-as-imagined is work-as-done is that then work-as-imagined must be perfect. If work-as-imagined is flawed, then it will negatively impact how work environments gets designed, and ultimately work-as-done itself. Steven Shorrock, Senior Specialist Safety & Human Factors at Eurocontrol notes:
Especially in cases where decisions about the work of others (or that affect the work of others) are made by those whose imagined view of work is incorrect, work-as-imagined can become very problematic. In such cases, inadequate involvement of those who do the work (work-as-done), and inadequate analysis and synthesis of the evolving context of work, often leads to badly designed work and work environments, and unintended consequences, including adaptations to work-as-done to overcome constraints and to work around other unintended consequences.
We saw this with space suits that didn’t fit tools, and work schedules that exceeded human capacity. Even as the astronauts attempted to complete the work, when they fell behind, NASA responded by comparing their performance to the schedule. This isn’t uncommon: evaluation of work is often done against an imagined version.
On Earth, executives, managers, and coworkers imagining our work in ways that are distant from the work we do results in building unreasonable expectations that not only create friction, but also failures. And when things don’t go as expected, they might then react in a manner that exacerbates the situation, cutting timelines, slashing budgets, and pulling resources under the assumption that efforts are being wasted. They cling to imaginary plans, instead of assuming that workers are doing the best they can, and that perhaps there was some flaw in how we imagined the work.
Accepting Work-As-Imagined and Work-As-Done As Potentially Separate Entities
How does this all relate to the radio silence? The gap between work-as-imagined and work-as-done grew so big that it led to a failure event: rather than an intentional retaliation, maybe they were so overworked and exhausted, they simply erred in not turning on their radios. From Atlas Obscura:
Gibson says that the three men simply failed to synchronize their radio response shifts, and that as a result, “one day we made a mistake and for a whole orbit we all had our radios off!”
After the event, NASA took notice and agreed to discuss working conditions.
Firstly, the astronauts laid out their side of things. “I told them everything,” Carr later recalled. “I said, ‘We need more time to rest. We need a schedule that’s not quite so packed. We don’t want exercise after a meal anymore. We need to get the pace of things under control.’” Next, it was Mission Control’s turn. According to NASA transcripts, they had five pages of “comments and status statements” teleprinted on board Skylab. They then gave the astronauts the “straight, unabridged words” they’d asked for. “It’s apparent to us… that the scheduling was too ambitious.”
We must accept that both work-as-imagined and work-as-done are necessary but distinct—we have to imagine work so that we can set plans, but we also have to be open to hearing how that work actually gets performed. The goal then is to close the gap between the two without assuming that work-as-imagined is perfect. Acknowledging the gap and then attempting to reconcile the two actually represents a learning opportunity to improve our models of work.
That acknowledgement by NASA opened the door to a win for everyone involved, including better work performed.
The negotiations were successful. For the last half of the mission, the astronauts of Skylab 4 got mealtimes and evenings off. Rather than being rigorously scheduled, the day’s tasks were added to a “shopping list” posted in the station, which the astronauts then completed when they decided the time was right. “It worked beautifully,” Carr said. “It turns out, when the mission was over, we completed every one of the experiments that we needed to do.”
How to Apply This
Steven Shorrock has three tips:
1. Pay attention to work-as-done. Study recurrent patterns, flows, trade-offs, compromises, change over time.
A persistent interest in work-as-done (implicitly acknowledging it is different from work-as-imagined) is important, especially because work-as-done in complex systems continually changes. From Erik Hollnagel:
Work-as-Done is a moving target because working conditions, demands, and resources rarely are stable. The solution is rather to try to understand what determines how work is done and to find effective ways of managing that to keep the variability of WAD within acceptable limits.
2.Explore work-as-imagined, -prescribed and-espoused. Consider the unintended consequences of wallpaper solutions.
Work-as-imagined and work-as-done are frequently used in literature, but four categories have been suggested as a framework for analyzing work: work-as-imagined, work-as-prescribed, work-as-espoused, and work-as-done. You can read more about each category in this article by Steven Shorrock.
3.Explore the gaps and implications, using blameless (no/pre/post)mortems… discuss, share, reconcile.
As Shorrock says: “When things go wrong, the adaptations, and the gaps between the varieties of human work, are subject to scrutiny.” This shuts down dialogue that could highlight the gap between work-as-imagined and work-as-done.

This is an Incident Labs project, with new issues every two weeks. We’re interested in figuring out the best practices for incident management for software companies. We also produce the Post-Incident Review, a zine focused on outages. If you use PagerDuty and Slack, our software project Ovvy simplifies scheduling and overrides, and is currently in private beta and free to use.
We’re excited to be back for 2020! Our next issue will be February 11, when we look at the difference between Safety I and Safety II.
Did you enjoy this issue?
Jaime Woo and Emil Stolarsky

The Morning Mind-Meld is a chance to build context between the conversations happening around DevOps and SRE, and to hopefully create some inspiration, even—or, especially!—during a hectic week.

In order to unsubscribe, click here.
If you were forwarded this newsletter and you like it, you can subscribe here.
Powered by Revue