The Social Life of Information
By John Seely Brown, Paul Duguid
Harvard Business School Press, 2000
ISBN 0-87584-762-5

Fascinating stories of people using newly invented technology at Xeroc PARC, such as the notebook, the local ethernet network, the word processor, and even the humble photocopiers. Because "The logic of information must ultimately be the logic of humanity."

These stories resonated particularly with the technical writer in me, and I've seen similar situations in my professional practice. For example, the book explains the social fallacy in "no one reads the documentation".

Each chapter of the book is like a wave, weaving a new idea into the ideas developed in previous chapters. I fought to extract structured notes from these waves of understanding, where ideas are revisited in several chapters. I stopped taking notes halfway through the book, probably because the time to read it was running out.

1. Limits to Information

[p12] Theorists talked about humanity's "bounded rationality" and the difficulty of making decisions in conditions of limited or imperfect information. Chronic information shortages threatened work, education, research, innovation, and economic decision making—whether at the level of government policy, business strategy, or household shopping.

Yet a little time in the nether regions of the Web can make you feel like the SETI researchers at the University of California, Berkeley, searching through an unstoppable flood of meaningless information from outer space for signs of intelligent life.

Could less be more?

[p13] most people who complain [about technologies] want improvements, not to go back to life without them.

[p14] If your Web page is hard to understand, link to another. If a "help" system isn't there, then click on through another 1,000 pages. Problems with information? Add more.

Life at Xerox has made us sensitive to this sort of trap. As the old flip cards that provided instructions on copiers became increasingly difficult to navigate, it was once suggested that a second set be added to explain the first set.

The power and speed of information technology can make this trap both hard to see and hard to escape.

[Moore's Law solutions:] more power will somehow solve the very problems that they helped to create. [...] Instead of thinking hard, we are encouraged simply to "embrace dumb power".

Drowning and didn't know it

[p15] [The industrial revolution] was a period in which society learned how to process, sort, rearrange, recombine, and transport atoms in unprecedented fashion. Yet people didn't complain that they were drowning in atoms. [...] while the world may be composed of atoms, people don't perceive it that way. They perceive it as buses and books [...]

[p16] The information reflected in bits comes to us, for example, as stories, documents, diagrams, pictures, or narratives [...]

Infoenthusiasts insist, for example, [...] that such things as organizations and institutions are little more than relics of a distant old regime.

[p17] But it's one thing to argue that many "second wave" tools, institutions, and organizations will not survive the onset of the "third wave." It's another to argue that in the "third wave" there is no need for social institutions and organizations at all.

Origin Myths

Historians frequently trace the beginnings of the information age not to the Internet, the computer, or even the telephone, but to the telegraph. With the telegraph, the speed of information essentially separated itself from the speed of human travel. People traveled at the speed of the train. [but fire and smoke had long been used to convey messages over a distance at the speed of light]

[p18] But it might be time to retreat from exuberance (or depression) at the volume of information and to consider its value more carefully. The ends of information, after all, are human ends. The logic of information must ultimately be the logic of humanity. For all information's independence and extent, it is people, in their communities, organizations, and institutions, who ultimately decide what it all means and why it matters.

Yet it can be easy for a logic of information to push aside the more practical logic of humanity.

[p19] the pencil—whose departure was predicted in 1938 by the New York Times in the face of ever more sophisticated typewriters — the fax, the copier, and paper documents refuse to be dismissed. People find them useful.

By taking more account of people and a little less of information, [futurists] might instead tell us where we are going, which would be more difficult but also more helpful.

Hammering Information

[p19] If you have a problem, define it in terms of information and you have the answer. [...] it allows people to skip quickly from questions to answers.

[p256 note]: the curious enigma of the laptop. Ads for these suggest that laptops can be so "loaded" with communications software that you can travel anywhere and remain a virtual presence in your own office. Yet in suggesting this possibility, they make you wonder why you need to travel at all.

Refining, or merely redefining?

From this viewpoint, value lies in information, which technology can refine away from the raw and uninteresting husk of the physical world.

[p21] Books are portrayed as information containers, libraries as information warehouses, universities as information providers, and learning as information absorption. Organizations are depicted as information coordinators, meetings as information consolidators, talk as information exchange, markets as information-driven stimulus and response.

Can it really be useful, after all, to address people as information processors or to redefine complex human issues such as trust as "simply information"?

6-D vision

[p22] demassification, decentralization, denationalization, despacialization, disintermediation, disaggregation.

These are said to represent forces that, unleashed by information technology, will break society down into its fundamental constituents, principally individuals and information.

[...] They help expose and explain important trends and pressures in society.

[...] they are parts of profound and often dramatic shifts in society's dynamic equilibrium, taking society from one kind of complex arrangement to another [...]

[p23] Ronald Coarse developed the notion of transaction costs. These are the costs of using the marketplace, of searching, evaluating, contracting, and enforcing. When it is cheaper to do these as an organization than as an individual, organizations will form. Conversely, as transaction costs fall, this glue dissolves and firms and organizations break apart. Ultimately, the theory suggests, if transaction costs become low enough, there will be no formal organizations, but only individuals in market relations.

[p26] We've gone from an economy where most people worked in manufacturing—in fairly large companies that were producing manufactured goods and engaged in things like transportation—to an economy where most people work for fairly large companies producing services.
— Paul Krugman, economist

The resilience of the large organization is not all that surprising. Given that information technologies are particularly good at taking advantage of large networks, the information economy in certain circumstances actually favors the aggregated, massified firm.

[p27] Finally, firms are not merely taking power from one another. They are accumulating power that once lay elsewhere. The political scientist Saskia Sassen traces the decline of the nation-state not to the sweeping effects of demassification and disaggregation, but to the rise of powerful, concentrated transnational corporations. The new economic citizen of the world, in her view, is not the individual in the global village but the transnational corporation, often so formidable that it has "power over individual governments".

[p28] disintermediation doesn't necessarily do away with intermediaries. Often it merely puts intermediation into fewer hands with a larger grasp. [...] Moreover this kind of limited disintermediation often leads to a centralization of control.

[p29] If managers are primarily information processors, then information-processing equipment might replace them, and organizations will be flatter. If, on the other hand, there is more to management than information processing, then linear predictions about disintermediation within firms are too simple.

[p30] a clear historical trend whereby information technology centralizes authority. Harold Innis, an early communications theorist, noted how the international telegraph and telephone lines linking European capitals to their overseas colonies radically reduced the independence of overseas administrators. Previously, messages took so long to travel that most decisions had to be made locally. With rapid communication, they could be centralized.

The paradise of shared knowledge and a more egalitarian working environment just isn't happening.
Knowledge isn't really shared because management doesn't want to share authority and power.
— Shoshona Zuboff

[...] it's a problem of management, not technology. [...] [information abundance] is presenting us with new and complex problems that another few cycles of Moore's Law or "a few strokes of the keyboard" will not magically overcome. The tight focus on information, with the implicit assumption that if we look after information everything else will fall into place, is ultimately a sort of social and moral blindness.

The Myth of Information

[p31] 6-D vision, while giving a clear and compelling view of the influence of the 'Net and its effects on everything from the firm to the nation, achieves its clarity by oversimplifying the forces at work. First, it isolates information and the informational aspects of life and discounts everything else. This makes it blind to other forces at work in society. Second, [...] such predictions tend to take the most rapid point of change and to extrapolate from there into the future, without noticing other forces that may be regrouping.

[p32] Today, it's the myth of information that is overpowering richer explanations.

[p33] We do not believe that society is relentlessly demassifying and disaggregating. The social forces that resist these decompositions, like them or not, are both robust and resourceful. They shaped the development of the railroad, determining where it ran, how it ran, and who ran it. And they will continue to shape the development of information networks. [...] to participate in that shaping and not merely to be shaped requires understanding such social organization, not just counting on (or counting up) information.

2. Agents and Angels

[p36] "chatterbots" [...] are software programs that will simulate a human response to typed questions.

[...] bots offer an interesting way to evaluate some of the endisms threatened by information technology.

The road ahead

[p38] in the hands of some futurists, agents appear more as angels offering salvation for encumbered humanity than as software programs.

[p39] The evangelistic view, however, may overlook some significant problems on the road ahead. [...] we seek to ground [agents] in social life rather than "information space".

[...] many accounts simply endow digital agents with human capabilities.

[...] some accounts don't so much make bots sound like humans as make humans sound like bots.

[p40] This kind of redefinition [...] makes human activity—learning, developing taste, wanting, choosing, pursuing, brokering, negotiating — appear principally as different kinds of information processing done equally by software goal-pursuing agents or by "wetware".

[p260 note]: People familiar with the old programs of artificial intelligence (AI) will recognize some of these redefinitions. For classic AI, it eventually became apparent that such activities could not be reduced to notions of mental representation and thus made computationally tractable. The real strength of AI came from recognizing that computers were not like people.

Ranks of agents

[p41] To understand the strengths and limits of agents, let us look primarily at personal assistants and some of the tasks planned for them.

• Information brokering: [e.g. search] [p44] Automatic organization, like automatic search, has its perils. [A sample search of returned 51 results. After 1h, a human dismissed the results as useless].

• Product brokering: [...] involves agents that alert people to the availability of particular items that, given a person's previous track record, might be of interest. [e.g. Amazon, CDNow] [p45] it was revealed that publishers paid Amazon for some endorsements [...] Is your agent, you must ask, neutral, biased, or merely weighted? And can you tell the difference?

• Merchant brokering: [p46] includes agents that can roam the 'Net comparing prices and so report back on "best buy" options. Markets, particularly long-distance markets, work best with standardized measures and standardized products. The integration of the European market, for example, has driven nonstandard varieties of fruit and vegetables out of the market altogether [...] A second problem is that, as part of a process of price competition and standardization, subjective qualities such as quality and service disappear. Both service and quality [...] tend to get squeezed out of markets when low price becomes the decisive factor. [...] They become highly problematic "externalities", not subject to the sales and purchase contract, but imposing heavy burdens after the fact. [...] Many government agencies award contracts to the lowest bidder to eliminate costly corruption. Instead they get costly headaches from inept or deceptive bidding, poor service, cost overruns, supplemental negotiations, "change orders", out-of-their-depth firms going bankrupt, and complicated litigation — all of which must ultimately be paid for. [p48] The 'Net, some economists believe, will become home to perfect information. It will be harder to make it a place of perfect (and trustworthy) informers.

• Negotiating: "situations where your software agent meets my agent and they negotiate for prices and conditions". [...] human negotiating is an intricate process that tends to keep an eye on externalities such as the social fabric as well as on the immediate goal. [...] The first thing to notice is that people negotiate all the time. These negotiations, however, are only occasionally over time. Usually, humans negotiate behavior. [...] for example, the way people negotiate with one another in the flow of collective conversation. [p50] In committees, particular acrimonious ones, such implicit negotiations often don't work. [...] explicit, rule-governed negotiation is clumsy, but necessary when the social fabric will not bear implicit negotiation. [...] In shopping [...] humans have the awkward habit of changing rules and shifting goals in mid-negotiation. [...] Shopping was in many ways a process of discovering or creating underlying preferences rather than acting in accordance with them. [...] In a supermarket, we are often more in negotiation with ourselves than with anyone else. [...] In interpersonal negotiation, people change their immediate goal for the sake of the higher goal of preserving the social conditions that make negotiation possible. [...] Authorities turn a blind eye if they sense that enforcing the law could be more damaging. [...] bots will happily destabilize a market in pursuit of their immediate goals. [...] Social friction and "inertia" may usefully dampen volatility and increase stability. [...] Development will be ill served by people who merely redefine elaborate social processes in terms of the things that bots do well.

How autonomous?

• Delegates: [p53] Human delegation relies as much on sympathetic disobedience when circumstances change as it does on strict obedience. [...] giving orders that have to take account of all possibilities is impossible in all but the simplest tasks. Consequently, rules, contracts, delegation, and the like rely on unspecifiable discretion and judgement on the part of the person following those orders. [...] Judgement and discretion are not features of software. They are products of human socialization and experience.

• Representatives: If bots make decisions and give advice autonomously, who takes responsibility for those decisions? Can Eliza be sued for medical malpractice? And if not, should the programmer be sued instead? [p55] bots end up with autonomy without accountability. Their owners, by contrast, may have accountability without control. [...] One of the remarkable features of the recent failures of derivative traders (such as Barings Bank, Long-Term Capital, and Orange County) has been how little investors, even major financial institutions, have understood the ventures to which traders armed with highly complex mathematical calculations and software programs were committing them.

The brightest fell

[p68] Strange things happen when a gift economy and a market economy collide.

Moore Solutions

[p59] in fifteen years computers will be involved in the solution of most crimes, but they will also be involved in the perpetration of most crimes.

Separate spheres

[p61] The human and the digital are significantly, and usefully distinct.

[p62] Designing bots to imitate or replicate human actions involves both moral and social-institutional questions. Your willingness to put your paycheck into a slot in the wall is, in many ways, remarkable. [...] we look for [ATM's] reliability, both instrumental and moral, in the organizations represented by the ATM and by the institutions regulating those organizations.

Some futurists seem continuously anxious to replace humans with bots in certain tasks without quite appreciating how people accomplish those tasks. In general, it will be better to pursue not substitution but complementarity. (We want ATMs that will supply us with money, not ones that will spend it for us.)

3. Home alone

Just a click away?

[p68] Some people trace a rise in home working in the early 1990s to the explosion of the Internet and related technologies. Other attribute it to the last major round of corporate downsizing. [...] One study of a telecenter found that 25 percent of the participants gave up within the first five months, and 50 percent within a year.

Leaving home

[p69] The cavalier futurists sometimes appear to work with a magical brand of computer not available to the rest of us. It's hard to believe that if they had to deal with the inexplicable crashes, data corruption, incompatibilities, buggy downloads, terrifying error messages, and power outages that are the standard fare for most, they could remain quite so confident about the ease of hot desking and home working.

[p264 note]: an effortless installation [of Internet access software] required two computers, two ISPs, two phone lines, a fax, and significant Internet experience to get the job done.

[p70] by overlooking both the social aspects of work and the frailty of technology, design that attempts to replace conventional work systems may often merely displace the burdens of work. In the transition to home offices, these burdens pass from the social system, where tasks are shared, onto the lap of individuals. The desire to show that with one computer one person can do everything may look not forward, but back to the stage in social evolution before anyone noticed the advantages of the division of labor.

[...] design needs to attend [...] to the ways in which social systems often play a key part in making even frail technology robust.

No room of one's own

[p70] the advertising company Chiat/Day [attempted] to reconceive work in a new way and a new building. The first [Wired, issue 2.07 in 1994] story followed the firm into its new offices. The second, five years later, watched it retreat to a more conventional way of working [Wired, issue 7.02 in 1999].

[p71] [in the building by the architect Frank Gehry] arrangements were based on the simple principle that no one should have a room of his or her own. [...] if they came to the office, employees could check out a laptop and a cellular phone and then look for somewhere to sit. At the end of the day, they had to check these back in again — so there were no tools of one's own, either.

[people] were not supposed to sit in the same place on consecutive days. Anything not stored in digital form on the company's servers [...] had to be put away in hall lockers at the end of the day and taken out again in the morning.

Cooling off

With hot desking, Chiat/Day sought to keep people on the creative edge of business. [...] Chaos rather than dynamism was the most noted result.

[p72] files, documents, and even those wobbly stacks people build on their desks often have a useful sequential order in their disorder (which is why tidying can be so disruptive). Similarly, people customize phones and laptops to fit personal ways of working.

[...] offices keep people together for good reasons as well as bad. Good office design can produce powerful learning environments. [...] For example, people often find what they need to know by virtue of where they sit and who they see rather than by direct communication. [...] With such lottery seating, incidental learning is much harder to come by.

[...] robust work patterns are hard to disrupt. To get things done, people rebelled. They filled their lockers and the trunks of their cars with the files they needed. They refused to turn in their laptops and telephones. And they used various strategies to build "virtual" departments. [...] physical areas of the office that people tried to occupy collectively so that they could re-create the useful connections of conventional departments.

[p73] Students on a college campus don't have offices, the masterplan argued. They move from classrooms, to libraries, to lounges. The same model was expected to work for the office. As some employees pointed out, however, at Chiat/Day the result was more like high school. In particular, the redesign appears to have unleashed the office equivalent of high school bullying. The daily unsettled space created daily turf wars. Departments tried to pull rank on each other, each claiming favored areas because it was "more important." Account executives and creative departments went to battle over pecking order. People with sufficient authority and power pulled rank on those without, seniors shooing juniors away from comfortable spots. Executives with assistants would order the latter in early to hold a place for their boss.

Office space is not neutral ground. Everyone who works in one knows that, for better and for worse, offices are dense with highly charged social relations. Power, tension, authority, and insecurity are all closely interwoven. They can help get work done, and they can hinder it. The troubles at Chiat/Day indicate how the structure of the conventional office, while it may pander to petty fiefdoms and yearnings for a window or a corner office, helps keep social tensions and turf war in check. Any one settlement may appear less than optimal, but it may nonetheless be better than no settlement at all.

[p74] Breaking up such a settlement has different consequences, depending on where you stand (or sit) in an organization. [...] it is usually those in less secure positions who bear the adverse consequences. [...] Middle management, one of the most insecure points in any organization, may find managing increasingly difficult.

[p75] Forward-looking companies are finding that designing a new company and designing its offices are intricately related processes, with each feeding off and into the other.

Given the nature of this resourcefulness, severing the ties that bind people together in work may be as damaging as binding them together more tightly. Finding the balancing point between the mix of centrifugal and centripetal forces needs to be the goal.

Maintaining balance

[p75] In well-supported offices, users often look to the next upgrade with much the same relish with which they greet the annual visit of the winter flu. They know it will precipitate crises and shortages, increase the burdens of those still up and running, and take weeks for the headaches to pass. They also know that just about the time that things return to balance, another disruptive strain will come through. Where there aren't technicians to manage the transformation or peers to share the burdens with, the weight of continuous product innovation can be unsupportable.

[p76] it's easy to believe that everyone except you knows how to use this stuff without a problem.

We saw this pressure at work on a new employee at Xerox PARC. She was intelligent and hard working, but got mired in difficulties with the office computer system. That system came with the usual promises of "usability" and self-explanatoriness, but she found it impossible to use or understand. Being a new-comer, she was reluctant to keep asking for help. Suffering in silence, it seemed daily more likely that she would have a breakdown if she didn't quit.

Then chance moved her desk from an isolated office into the center of a group of offices. There she immediately benefited from the incidental learning that we mentioned earlier. She saw that these "stable" machines crashed for everyone. She saw that there was no more "ease" for experienced assistants, longtime employees, or PARC's hallowed computer scientists than for her. And she also saw that when a machine did crash, its user would without shame look around for help from someone else who, whatever their status, had successfully steered around that particular problem. No one person knew how to handle these temperamental machines. But spread around the office was enough collective knowledge to keep them up and running.

[p77] Office help systems, this story indicates, are not limited to manuals, vendor Web sites, IT departments, or on-line files. The office social system plays a major part in keeping tools (and people) up and running. The "geek" who understands the network, the secretary who knows the secrets of Word, the one colleague proficient with databases, the other who has learned Java in her spare time, and the one who knows how to nurse the server all contribute.

Most systems, amalgams of software and hardware from different vendors, rely on social amalgams of this sort to keep everything running. [...] the facts of office life reveal a combination of technological frailty and social resourcefulness. Infoenthusiasts, however, tend to think of these the other way around, missing the role of the social fabric and assuming that individuals in isolation can do it all.

[...] the 'futz factor' [...] includes "time users spend in a befuddled state while clearing up unexplained happenings [and] overcoming the confusion and panic when computers produce enigmatic messages that stop work." Home office workers usually [...] lack necessary peer support. Consequently, with current technology, money-losing futzing, late at night and early in the morning, is endemic to the home office. Lacking the boundaries and structures provided by office life, work spills relentlessly over into private and family life.

Invisible hands

[p79] there's no IT manager in the home—although much of the technology seems to require one. These cumulative problems may lead to the curious paradox that information technology, by ignoring the role played invisibly by the social system, is keeping people out of the home and in the conventional office, and not the way around.

Something for next to nothing?

[p81] Figures for the 1990s show Microsoft spending $16,000 per annum for each of its workstations on maintenance and upgrading.

Socializing technology

[p87] Too often, information technology design is poor because problems have been redefined in ways that ignore the social resources that are an integral part of the socialization process. By contrast, successful design usually draws on these social resources, even while helping them change.

It is much harder to go into the home, a more isolated social setting. [...] People still have difficulties with much simpler appliances like the VCR. [...] Were there a social context for VCR recording (as there is for watching), the problem would probably disappear.

Consider Alexander Graham Bell's strategy for introducing the telephone, which drew both on the office and on other situations for social resources.

[p88] Early on, he put telephones in hotel rooms and encouraged guests to use them in the familiar task of talking to the front desk. [...] He also promoted their use for office intercoms, drawing in the resources of the office while nonetheless helping it to change. [... Later, the company] put phones near lunch counters. That way, it reasoned, people that didn't know how to use them would be likely to see people who did know how and in this way learn about the phone system.

Though the telephone was a transforming technology, Bell nonetheless worked with the social context of his day, not against it or in isolation from it.

[p89] in order for people to be able to work alone, technology may have to reinforce their access to social networks.

4. Practice makes process

[p92] Taking "a clean sheet of paper", reengineering teams were told to reorganize their organizations around the processes that did add value, making these the center of the new organization. [...] "Forget all you know", managers were told.

[p93] Reengineering seems to have been behind the transformation of several dinosaurs of the industrial age into phoenixes for the digital age.

Did the focus on process, perhaps, overlook the increasing demand for knowledge in modern organizations? We suspect it did.

Perfecting process

[p94] To complete a process, something passes from A on to B ending with C—from, for example, receiving to manufacturing to shipping. In such well-defined processes, it is the "longitudinal" links between each stage that appear to matter. Lateral ties among people doing similar tasks—among, for example, the people in shipping—appear at a heavy discount from a process-based perspective. They are generally regarded as non-value-adding.

[p95] It is not surprising, then, that business process reengineering has had less success in the parts of organizations that are less linear and less clearly defined by process and information. Management, for example, has proved notoriously hard to reengineer. So has R&D. In such areas, life is less linear; inputs and outputs are less well defined; and information is less "targeted". These are, rather, areas where making sense, interpreting, and understanding are both problematic and highly valued—areas where, above all, meaning and knowledge are at a premium.

Meaning and ends

[p95] [Reengineering] focuses most heavily on the input and output of the stages in a process. It is relatively indifferent to the internal workings of these stages—to the particular practices that make up a process and the meaning they have for those involved.

[...] studies of workplace practice [...] reveal tensions between the demands of process and the needs of practice. [...] these tensions are often the result of struggles over meaning. [These struggles] occur throughout, pitting the process-focused need for uniform organizational information against the practice-based struggle for locally coherent meaning.

[p96] many of the problems faced by the health-insurance claims processors [...] could be traced to clashes over meaning and sense making over such things as what a form signifies, why similar claims elicit different reimbursements, and who does or does not qualify for reimbursement. In the end, these problems for the claims processors create problems for the company.

To do their job, processors need to be able to make sense of what they do. The company offers explanations. But, from the processors' point of view, the organization's information and explanations are difficult to use. They explain things in terms of the company's overall goals and processes, but they take little account of the immediate practicalities of the claims processors' work. Indeed, many of the company's explanations have the ring of those parental imperatives that skip explanation and simply say "Do it this way because I say so." Such imperatives make life easier for the company, but difficult for the processors, who need to understand what they do to do it well and also must justify their actions to customers.

[...] in the end it is the practice of the people who work in the organization that brings process to life, and, indeed, life to process. Organizations, then, should not attend to the process and process-related explanations only. They must also attend to practice. [...] we mean the activity involved in getting work done.

Looking the other way

[p97] From outside, people find meaning in functional explanations. They rely on process-based, cross-functional, longitudinal accounts of why things are done. From inside, people take a lateral view. The claims processors, for example, look less to their superiors or to the people to whom their work goes next than to their peer group for explanations of what they do and why. For them, knowledge comes more from fellow practitioners than from cross-functional connections.

[p98] Focusing on individuals, process accounts overlook social resources that people in similar occupations provide one another. [...] In an exemplary piece of partial blindness, for example, British Telecom did notice the damaging isolation of its home workers. As a remedy, however, it decided to pipe the sound of canned background chatter into their home offices.

[p99]: you can't redesign process effectively if you don't understand practice.

Representing process

[p99] An anthropologist, Julian Orr, studied the Xerox technical representatives (reps) who service and repair the company's copiers at customers' sites. [...] The reps might almost be said to succeed despite the company's best intentions. Their success is in good part a triumph of practice over the limits of process.

[p268 note]: these reps are in many ways models of the new knowledge worker. They work alone, with sophisticated technology, supported by extensive information resources.

[p99] Customers having difficulty called the Customer Service Center. This in turn notified the reps. A rep would then go to the customer's site, identify the problem with the help of the machine's error codes, find the right response from the machine's documentation and, having fixed the problem, "clear" the call with the service center.

[p100] [The company] provided training courses [and] documentation. This "directive" documentation provides a map to the repair work, with directions guiding the rep from problems (which appear as error codes in the machine) to repair solutions.

[From the company's perspective, the courses and documentation provides the information needed to see the process from beginning to end.]

Understanding practice

Neither management nor management theorists, [Julian Orr] points out, are adequately "concerned with work practice", by which he means they "do not focus on what is done in accomplishing a given job". [...] what looked quite clear and simple from above was much more opaque and confusing on the ground. Tasks were no longer so straightforward, and machines [...] exhibited quite incoherent behaviors. Consequently, the information and training provided to the reps was inadequate for all but the most routine of the tasks they faced.

[p101] For example, in the process view, machines work quite predictably. [...] Yet large machines, comprising multiple subsystems, are not so predictable. Any one machine may have profound idiosyncrasies, for each inevitably reflects the age and condition of its parts, the particulars and patterns of use, as well as the distinctive influences of the environment in which it sits [...] All this undirected machine behavior inevitably undermines the very premise of directive documentation.

All, too, probably know the frustration of "!FATAL ERROR!" messages that no one, not even the people in customer service and tech support, can explain.

[p268 note]: Our own favorites come from the popular email software, Eudora, which occasionally produces messages that include "No one is listening, so you might as well stop typing" or, following an unintelligible error report, "And a fat lot of good that does you".

[p101] The rep's real difficulties arose, however, not simply because the documentation had lapses. They arose more problematically because it told them what to do, but not why. It gave instructions, but it didn't explain. So when machines did something unpredicted, reps found themselves not just off the map, but there without a compass or tools for bushwhacking. [...] they needed to make some sense of the machine in order to fix it. Directive documentation, however, wasn't designed for sense making. It was designed for rule following.

[p268 note] Reps also had to make sense of the machines to explain problems to customers. Without a coherent explanation, it is hard to keep the customer's confidence in the rep, the machine, or the corporation. "It's not in the documentation" is unacceptable on all fronts. Orr argues that the reps also have to work on the social system of which the machine is but a part. This is common practice for most people in jobs like this.

[p102] So Orr found, when reps fell off the map of process, they went to breakfast.

When the going gets tough

From a conventional perspective, the reps' job was highly individual.

Yet Orr found that the reps were remarkably social, getting together on their own time for breakfast, lunch, coffee, or at the end of the day—and sometimes for all of the above.

At these meetings, while eating, playing cribbage, and engaging in what might seem like idle gossip, the reps talked work, and talked it continuously. They posed questions, offered solutions, constructed answers, and discussed changes in their work, the machines, or customer relations. In this way, both directly and indirectly, they kept one another up to date with what they knew, what they learned, and what they did.

[p103] Chat continuously but almost imperceptibly adjusts a group's collective knowledge and individual members' awareness of each others.

The reps' chatter stood out, however, because the process view assumed that they worked alone and had adequate resources in their training, tools, and documentation. Time spent together would, from the process perspective, be non-value adding. [...] The informal and extracurricular group helped each member to reach beyond the limits of an individual's knowledge and of the process documentation.

The practice in the process

[p104] Having reached his limits, the rep summoned a specialist. But the specialist could not understand what was going on either. So the two spent the afternoon cycling the machine again and again, waiting for its intermittent crashes and recording its state at the time. Simultaneously, they cycled stories about similar-looking problems round and round until these stories, too, crashed up against the peculiarities of this machine. In Orr's account, the afternoon resembles a series of alternating, improvisational jazz solos, as each took over the lead, ran with it for a little while, then handed it off to his partner, all against the bass-line continuo of the rumbling machine until finally all came together.

For in the course of the afternoon, the two gradually brought their separate understandings closer together and simultaneously came closer to a collective understanding of the machine. Eventually, late in the day, the different cycles achieved collective understanding. The machine's previously erratic behavior, the experience of the two technicians, and the stories they told finally formed a single, coherent account. They made sense of the machine and as a result could fix it and satisfy the customer.

We can understand this journey better by approaching it in terms of collaboration, narration, and improvisation.

Collaboration

The reps' practice involved both collaborative problem solving and collective sharing in the solution. [...] Orr later heard a group discussing the story in the course of a long cribbage game.

[p105] In chapter 3, [...] we argued that tasks better shared are often pushed into the lap of individuals, in the name of "empowerment." [...] The result is quite disempowering and inefficient, burdening people with individual responsibility that is better shared by the group.

For instance, the reward structure made it difficult for each rep to hold expensive parts critical for certain types of infrequent failure. [...] To get around this problem, the reps implicitly formed a collective pool of parts, so that each could draw readily on the other in times of need, and the necessary parts were always available.

More important, the reps also developed a collective pool of knowledge and insight on which they drew. [...] All contributed from personal stock, and there was a great deal of overlap, but each had his or her strengths, which the others recognized and relied on.

[p106] In this pool of knowledge, where one person's knowledge ends and another's begins is not always clear. In the example we gave above, it took the collaboration of the two technicians working together to come to a coherent conclusion. But neither had a decisive "piece" of knowledge. Nor was the final solution the property of either one. It was a collective process that created an indivisible product. Thus we tend to think of knowledge less like an assembly of discrete parts and more like a watercolor painting. As each new color is added, it blends with the others to produce the final effect, in which the contributing parts become indivisible.

Narration

Reps tell stories about unsolved problems in an attempt to generate a coherent account of what the problem is and how to solve it. They may do this individually, putting their own story together. Or they can do it collectively, as they draw on the collective wisdom and experience of the group.

[...] stories are a powerful means to understand what happened (the sequence of events) and why (the causes and effects of those events). And so storytelling is indispensable for the reps for whom what and why are critical matters yet often hard to discern.

[p107] The value of stories, however, lies not just in their telling, but in their retelling. Stories pass on to newcomers what old-timers already know. Stories are thus central to learning and education, and they allowed the reps to learn from one another.

Stories, moreover, convey not only specific information but also general principles. These principles can then be applied to particular situations, in different times and places. So the reps find that they carry back what they have learned from their colleagues in the coffee shop to a different site and a different problem.

In their storytelling, the reps developed a common framework that allowed them to interpret the information that they received in a common light. To collaborate around shared information you first have to develop a shared framework for interpretation.

Learning to tell their war stories, then, was a critical part of becoming a rep. It allowed newcomers to see the world with a rep's eyes. And it allowed all to share in their major resource—their collective, collaborative wisdom. [...] Orr concludes,

[p108] This talk shows their understanding of the world of service; in another sense, the talk creates that world and even creates the identity of the technicians themselves. But neither talk nor identity is the goal of the technicians' practice. The goal is getting the job done.

Improvisation

[Reps] have to rely heavily on improvisation to close the gap between the world as they find it and the inevitably limited model of that world embedded in routines and processes. [...] practitioners themselves are barely aware of [that inherent improvisation].

Part of the skill of work, all work, then is routinization, adapting the particulars of the world so that they fit within the general schemas of the organization.

Forms are the crucial means by which an organization brings the heterogeneous world into line with its processes. [...] Everyone knows what it is like not to fit within the standard form that gets things going. And everyone knows, too, the value of the skilled representative who understands how to fit you into the form and the firm without causing problems for either.

[p109] Such gap-closing improvisation is one example of what another former colleague and organizational ethnographer, Lucy Suchman, describes as "endless small forms of practical 'subversion' taken up in the name of getting work done."

Processing

[processing are] are attempts to disguise unauthorized behavior so that it looks authorized, to justify improvisation in terms of routine. [p110] processing provides a screen between what people do and what people say they do. It helps turn unauthorized practice, however effective, into authorized routine, however inept. It makes us all appear "rational" and rule governed to the world, even though a great deal of what everyone does is, of necessity, guesswork and intuition.

Most people [...] simply assume at what they do and what their job description says are one and the same. People thus keep their own skills hidden even from themselves. (Even braggarts usually brag about the wrong things.)

[...] organizations make themselves blind to what lies outside the narrow tunnel of process. [...] by subordinating practice to process, an organization paradoxically encourages its employees to mislead it. Valuing and analyzing their improvisations, by contrast, can be highly informative. Indeed, it's been suggested that Xerox stumbled so badly in the 1970s in part because it failed to gather evidence of poor quality from its field representatives.

Lateral Thrust

[p111] As a result of Orr's work, rather than trying to support the reps with yet more information from outside the reps' community, Xerox turned instead to reinforcing internal ties. The first step was simple. Reps were given two-way radios, which allowed them to continue to talk to one another even when working apart. This intervention both supported and acknowledged the reps' ways of collaboration, narration, and improvisation.

The second step was more ambitious. [...] the insight reps developed in the course of their work tended to have a short reach, traveling primarily in local groups, and a short life, fading from memory even locally. Consequently, reps near and far ended up reinventing fixes that might have been known elsewhere.

[p112] The Eureka project set out to create a database of this useful knowledge, preserving over time and delivering over space resourceful ideas. [...] Such a database must be selective. [...] as with scientific articles, the reps' tips are subject to peer review, drawing on those same lateral ties that make the reps resources for one another. A rep submits a tip, peers review it, and if it stands up to scrutiny—is original, is useful—then it is added to the database.

For the reps, this database has become more than an indispensable tool (reckoned to save the company up to $100 million per year in service costs). It is also a recognition of the value of their own knowledge and knowledge creation, which was previously disregarded by most of the corporation. [p113] And it is a means by which individual reps build social capital and recognition among their peers. At a recent meeting of reps in Canada, a rep who provided highly useful tips was given a standing ovation by his peers.

[...] virtual groups tend to mirror conventional groups, not transcend them. [...] Lateral ties are as significant in cyberspace as in the old world. Cross-functionality is no easier.

Beyond Either/Or

[p114] Practice-driven change, by contrast, tends to be more continuous and continual. For that very reason, overlooking practice risks cutting organizations off from such continuous change. And if they are cut off like this, organizations can only expect to lurch from one top-down "palace revolution" to another.

Of course, practice-based views have their own blind spots. The current interest in virtual organizations, for example, downplays the uses of formal organization and structure, while self-organization abandons it almost entirely.

The process view is important, giving shape and direction to an organization. It always risks, however, binding people too tightly to process, cutting them off from their "lateral" resources, blinding the organization to improvisation and new ideas, which may enter organizations along the lines of these peer groups.

[p115] Practice suffers from the opposing danger—of allowing itself to evolve too independently and so become too loosely "coupled" to the organization.

The balancing act, as we shall see, requires developing coupling loose enough to allow groups to develop their own new knowledge, but tight enough to be able to push that knowledge along the lines of process. The idea that all that matters here is information woefully underestimates the challenges involved, which are ultimately, as we shall see, challenges of organization, knowledge, and innovation.

5. Learning—in Theory and in Practice

Not summarized.

6. Innovating Organization, Husbanding Knowledge

Not summarized.

7. Reading the Background

Not summarized.

8. Re-education

Not summarized.

References

Orr, Julian. 1996. Talking about Machines: An Ethnography of a Modern Job, Ithaca, NY: IRL Press. ISBN 0801483905

Why I bought this book

Altitude Software sent me to a boring Microsoft conference, filled with sales pitches and the occasional disinformation. After a speakers recommended this book on stage, the title picked my curiosity.