This post is by The Maintainers Co-Directors Andy, Jess, and Lee.
A big part of our job involves talking to other people about maintenance and how both maintenance and the lives of Maintainers can be improved in our culture today. Over time, we’ve noticed that different people tend to offer different explanations for why neglect of maintenance is so common. We thought we’d collect these explanations in one place – both for our own use and to get feedback from others.
In this post, we highlight eight factors that lead individuals and organizations to neglect maintenance. Nearly all of these factors influence each other in one way or another. Many of them overlap. Some may even be collapsible into others. Yet, our sense is that there isn’t a single, ultimate cause lying behind all the others. Rather, it is precisely because so many factors point in the same direction—because neglect is overdetermined—that maintenance problems are so pervasive.
We publish this post largely because we are interested in what others think. What are we missing? What could we add? How might we refine these thoughts? What kind of evidence and empirical work would help clarify and support theories of maintenance neglect?
As always, we look forward to hearing from the community. After all, this post is an attempt to echo what we have heard back to you.
1. Human Psychology and “Cognitive Biases”: We are unsure how we feel about terms like “cognitive bias” but the simple truth is that people we talk to often attribute poor maintenance practices to the psychological dispositions of individuals, whether as a product of biology or cultural rearing. These psychological explanations take at least a few forms:
a. Hyperbolic, or Delay, Discounting: Hyberbolic, or delay, discounting is the idea that humans prefer to enjoy rewards now than for one that arrives later, even if the later reward is greater—put another way, humans tend to discount the value of later rewards in light of present gains. Individuals often assert that discounting leads to deferred maintenance and “kicking the can down the road,” when resources can be put to other sexier uses than upkeep. Theorists have given evolutionary reasons for the development of delay discounting, but studies have also found dramatic differences in discounting between cultures, which means there’s more than just raw biology at play here. Rearing is important.
b. Paying attention to the novel: Another argument that we’ve heard several times is that humans are wired by evolution to both attend to novel things in their environments and take the given for granted. Maybe. But we have a hard time seeing this when looking across cultures or back in time. For most of human history, for example, annual economic growth was at 1% or less, and there was little to no fetishization of technological change. So again, whether thinking historically or across cultures, there’s little reason to believe that paying attention to the new is some ahistorical human reality, even if such dispositions play a role today.
2. Cultural Myths: Another set of explanations we hear focus on cultural myths, or discourses - stories we’ve come to tell about the world that end up shaping our values, priorities, perceptions, and actions. Perhaps foremost among these explanations is the idea that the way we’ve come to think and talk about “innovation” leads us to put off and neglect other activities, including maintenance. Note: it’s not that actual innovation—the process of introducing new things and practices in the world—is somehow opposed to maintenance (though it can be). There is, for instance, innovation in the technologies and practices of maintenance. Rather, the ideology of innovation, or innovation-speak, leads us to prioritize creating the new over caring for the old.
How we came to this innovation-speak dominated moment is a complex story. Books like Christine MacLeod’s Heroes of Invention: Technology, Liberalism, and British Identity, 1750-1914, and Benôit Godin’s Innovation Contested: The Idea of Innovation Over the Centuries, examine how invention (once a low-status social role) and innovation (once a concept used to disparage religious heretics) became lauded over the course of the 19th and 20th centuries. Critical works, like John Patrick Leary’s Keywords: The New Language of Capitalism and Maintainers Co-Director Lee Vinsel’s “95 Theses on Innovation” and “How to Give Up the I-Word,” highlight how this change in values have led to a variety of negative consequences.
These cultural myths of innovation have especially come to shape public policy, including around technology and education. In many cases today, “technology policy” just means creating new things, not caring for old ones. And engineering education has increasingly come to focus on the need to train students in innovation and entrepreneurship, even though about 70% of engineers work in operations and maintenance. Our world is upside down.
3. Status Systems: People we talk to also argue that a big part of the problem around maintenance work and occupational roles that focus on maintenance is that they have low social status. This social reality is likely very old, but has also changed a great deal over time, and in general, we believe the topic needs more historical and social scientific investigation. Gender, race, and other identity markers often play important roles here.
Several individuals have noted the Dalit—or “untouchable”—classes in the Indian caste systems do various forms of maintenance work, like cleaning toilets. And we can examine similar work in the lives of enslaved peoples around the globe, including in the United States. And such distinctions around occupational roles clearly do not end when slavery does. In Virginia, for instance, several jobs, such as domestic workers and hotel concierges, are defined out of minimum wage requirements. Brief inspection finds that these unprotected jobs were typically done by black people during the Jim Crow era. Clearly, a great deal of maintenance and care labor—including housework and childcare—is also highly gendered and, for instance, not taken into account in measures of “economic” activity, as feminist scholars and others have been showing for decades.
The statuses of specific occupational roles, especially ones tied to individual technologies, also change quite a bit over time. In Auto Mechanics: Technology and Expertise in Twentieth-Century America, Kevin Borg describes how the occupation of auto mechanic went from being a job that working-class people entered to move up in the world to become a role taken up by “unpromising” individuals who were not “college material.” A similar dynamic has played out with “electrician,” once a high-status aspirational role, now a “trade,” and Clive Thompson has argued in Wired that the same thing may be happening with computer coding.
Such status hierarchies influence other kinds of jobs as well. In Being Mortal, Atul Gawande describes how effective gerontological treatment can be for the quality of life for elderly patients. A great deal of gerontological work focuses on maintaining bodies, or more accurately helping others maintain their own bodies. (Gawande writes lyrically about a doctor looking over an elderly patient’s toenails, checking to see if she is taking care of herself.) Yet, Gawande notes that the field of gerontology is dying in many parts of the US because young people choose to go into specializations that are seen as more innovative and cutting-edge and which receive higher pay.
A recurring question is how we come to know the relative statuses of various occupations, and one answer is that our parents teach us to see the world in this way. In the classic essay, “What Do Animals Do All Day?: The Division of Labor, Class Bodies, and Totemic Thinking in the Popular Imagination,” the sociologist John Levi Martin examined Richard Scarry’s Busytown children’s books and found a correlation between animal species and occupational statuses. Martin argues, for instance, that pigs in Busytown are often stand-ins for “America’s Working Man.” Martin’s essay is fun and funny, but the more serious point is that we are taught status systems and occupational hierarchies, even as very young children, so much that they become a part of the unconscious ways we see and experience the social world.
Now, one might protest all of this business about occupational statuses has more to do with what kinds of jobs people want than whether we neglect maintenance. But people tell us time and again that status hierarchies shape which individuals and what kinds of work get attention and resources within organizations. To give but one example, open source software maintainers often find it hard to get credit for their work, which encourages them to move into other roles where they will get more status and higher pay, which creates an unstable and precarious situation for the software itself.
4. “Growth” and Forms of Short-Termism: Another common explanation is that our culture has become obsessed with economic and other forms of “growth” and that, in a variety of ways, this mindset favors short-term gains that affect maintenance in two ways: 1. It leads us to defer maintenance in the short-term. 2. It leads us to over-build and over-adopt new technologies that increase our maintenance load down the road.
We see this play out in both the public and private sectors. Chuck Marohn of Strong Towns has thoroughly examined how short-term focuses on growth leads localities to take federal money to build new infrastructure even though there is zero reason to believe those localities will ever have the tax base to maintain that infrastructure over the long-term. Similarly, investigations into problems in Washington, D.C.’s trouble-plagued Metro have found leaders responsible for the subway system repeatedly prioritized expanding the system over maintaining it. While some voices in our culture argue that government is often better at long-term thinking than are businesses, the reality is that electoral politics encourage officials to concentrate on growth and ribbon-cuttings over unglamorous and hard-to-perceive maintenance.
In the private sector, scholars such as Mary O’Sullivan and William Lazonick, have examined the ideology of shareholder value as it arose in the 1980s – shareholder value is the assertion that business executives should make choices that enrich shareholders. Often this idea is reflected in a fixation with showing growth and other positive indicators in quarterly reports. While much more research needs to be done, we frequently hear that such short-term thinking leads firms to neglect maintenance and other fundamental tasks because they are not seen as contributing directly to growth. Similarly, Silicon Valley-style startups and the current focus on producing so-called “unicorns” (companies that valued at over $1 billion) emphasize launching sexy ideas without thinking through the sustainability or maintenance of the venture (or even, like, profitability . . . ).
5. Economic Incentives and Collective Action Problems: Economic incentives play several roles in fostering both success in and neglect of maintenance. Some industries, including chemical production, petroleum refining, some manufacturing, and digital services (including Amazon and Netflix), have impressive and cutting-edge maintenance units because uptime is so crucial to their profitability. On the other end of the spectrum, in a quantitative study of work orders, CEO Tom Arnold and others at Gridium, a company that makes software for building management, found that roughly 98% of jobs in building maintenance were reactive, or repair-focused. Only 2% of work effort went into preventive maintenance. While we need much more research into this topic, it appears that different industries have wildly different incentives around maintenance.
Another class of economic maintenance issues arise from so-called collective action problems, especially the phenomenon known as “free-riding.” For instance, citizens like well-functioning infrastructure and will take advantage of it, but may not be willing to pay more taxes to improve or guarantee continued quality of that infrastructure. A great deal of recent thinking has gone into how free-riding and other collective action problems affect open source software, which depends largely on volunteer labor creating a precarious and unsustainable situation, especially since such volunteer work so often leads to burnout.
6. Capability Traps: In his paper at Maintainers II, “Why Aren’t the Wrenches Turning on Preventive Maintenance?” Tom Arnold, the CEO of Gridium, argued that individuals responsible for maintenance often face a difficult tradeoff of better-before-worse or worse-before-better. That is, they have two fundamental options: First, they can use quick fixes to deal with problems and get systems back up and running, while facing (probably more like ignoring) the ultimate failure that lays down the road. Or second, they can take time to develop lasting solutions that improve overall performance, though this often means the situation will get worse in the meantime, for instance, because systems will need to be taken offline.
A wide variety of causes, from the demands of angry users to a sheer lack of time, leads maintainers to run around putting out fires instead of creating more fundamental improvements. The end result? John Sterman of MIT’s Sloan School argues that such capability traps lead individuals and organizations to ignore existing, “low-hanging fruit” solutions that would improve performance simply because they do not have the time or cognitive energy to implement them. This avoidance of worse-before-better problems is likely also a crucial factor in the legacy software space, where organizations put off changing software systems just because dealing with the issue will be such a headache.
7. Interests and Power: As with any other domain of human activity, power and financial interests play important roles in shaping where maintenance is and isn’t done and who benefits. On the one hand, elites and powerful corporations vocally advocate for the maintenance of some public and some private infrastructure. The advocacy organization, Infrastructure Week, which pushes both for the construction of new infrastructure and maintenance of existing systems, has a steering committee that includes groups like the National Association of Manufacturers, the Business Roundtable, and the U.S. Chamber of Commerce. Not exactly purveyors of socialist thinking!
On the other hand, there’s no denying that maintenance, like any other resource or activity in our society, is unequally distributed. For example, news articles often highlight how residents in public housing face terrible maintenance neglect, including broken elevators, which place stress on all residents but especially the elderly, disabled, and ill. We see the same thing in public schools that primarily serve the poor and disenfranchised, as recent reporting by Benjamin Herold at Education Week has examined. A recent study done by the Philadelphia Inquirer found over 9,000 environmental problems in Philadelphia schools, including “mold, deteriorated asbestos, and acres of flaking and peeling paint likely containing lead.”
8. Moral Lousyness: Finally, a near constant theme is that as a society we do not pay attention to maintenance and maintainers but we SHOULD do so. Depending on your view of human nature, this moralizing may just be another way to talk about cognitive biases and economic incentives, but the key point is that often enough arguments are put in moral rather than other terms. The moral “should” is an – at least implicit – point of the idea that infrastructure and some forms of—often gendered—labor are currently “invisible” but should not be.
It’s not just left-leaning thinkers who talk about maintenance and maintainers this way either. In a recent episode of his podcast Econtalk, the libertarian economist Russ Roberts recounted the story of a (probably apocryphal) exam in a business class in which the professor asked one question: what is the name of the woman who cleans this building? Roberts took the story as a lesson about the “minimal human interaction you’re talking about in a place where there’s disparity between people’s prospects and material well-being.” The point is that we should not treat other humans like this, even if differences in social classes make it easy to do so.
Regardless of the overall role of morality in the kinds of maintenance problems we face, it seems clear that a great deal of writing on maintenance today attempts to make a moral case for changing our ways.
These eight categories capture most (if not all) of the explanations for the neglect of maintenance that we’ve collected in our various conversations, readings, and reflections. They cover a wide range of factors, from individual to collective, from moral to material. We’re eager to hear from you about what we’ve missed, how we could collapse or further tease out these categories, and what kinds of generalizations--and recommendations for positive actions--we might draw from them. We are eager to hear your responses and reactions: drop a note to The Maintainers listserv or get in touch with us directly!