Friday, December 08, 2006 adds high-quality HP presses for on-demand books

How many years away are we from printing high-quality books at home? adds high-quality HP presses for on-demand books: "SEATTLE -- has fired up a new line of digital printing presses at its distribution centers around the country, betting that bigger authors and publishers will feed growth in on-demand book publishing.

Amazon will use Hewlett-Packard Co.'s Indigo presses at its fulfillment centers and its BookSurge subsidiary, Amazon spokesman Sean Sundwall said. The new presses offer better quality and higher capacity than the company's previous on-demand printing setup, he said"
Lengthening the Feedback Loop: A History of Feedback Within the Context of Systems Theory

Julia Evans

LIBR 243 Systems Analysis

San Jose State University


Once you start looking for feedback loops, you see them everywhere. As I write this, my refrigerator clicks on, reminding me that negative feedback from its thermostat is responsible for keeping my food from spoiling. The feedback I received from my bathroom scale this morning will determine whether I have dessert after dinner (I won’t). From our everyday machines to our own behaviors, feedback appears in many aspects of our lives. The purpose of this paper is to show how feedback developed from an engineering principle to part of a unifying theory that helps to shape the way we look at the world. I will trace the concept of feedback through history within the broader framework of systems theory, and demonstrate how it is being applied to business, economics, and society at large.


Feedback, whether positive or negative, is essentially a means of feeding output back into a system in order to assert control over a process and achieve a desired result. Positive feedback results in continuation of the current process, while negative feedback results in modification (or correction) of behavior. Even before it was explicitly named, mechanics and engineers have used the principle of feedback to create a variety of machines. Ancient Greeks used the principle to control wine dispensers and water clocks; James Watt used a mechanism called a “governor” to control his steam engine during the industrial revolution. In 1927, an engineer at Bell Labs found that he could reduce the noise in an amplified signal by “feeding back” some of the signal into the circuit in the opposite direction. By the 1940’s, feedback was being routinely applied in commercial applications; but there was little theory behind it (Conway & Siegelman, 2005, 121). Despite the fact that this principle was finding its way into modern lives, no one had recognized its broader applications. This all changed when a man by the name of Norbert Wiener was working, along with Julian Bigelow, to develop an automated gun-control mechanism for American warplanes during World War II. Although the project was never put into production, their collaboration resulted in a shift towards a new way of thinking:

For Wiener, the discovery of feedback was tantamount to the discovery of fire…. Wiener drew the connections between feedback in the technical sense, in the physiological sense, and the innumerable feedback loops wired into the living electrical networks of the brain and nervous system. (Conway & Siegelman, 2005, 122)

Norbert wiener saw feedback as more than a technical idea. When he learned of the principle, he saw an underlying theory that united biology and technology. With the publication of Cybernetics in 1949, Wiener proved mathematically that the principle of feedback was equivalent to the physiological principle of homeostasis. This would have obvious implications for exploring the connection between man and machine; in fact, many books today with “Cybernetics” in the title have to do with artificial intelligence and robotics. Nevertheless, it is a mistake to think of cybernetics as relating solely to cyborg culture. Among those who saw the potential for a broader application was Ludwig Von Bertalanffy, the father of Systems Theory:

Modern technology and society have become so complex that the traditional branches of technology are no longer sufficient; approaches of a holistic or systems, and generalist and interdisciplinary, nature became necessary. This is true in many ways. Modern engineering includes fields such as circuit theory, cybernetics as the study of “communication and control,” and computer techniques for handling “systems” of a complexity unamenable to classical methods of mathematics. Systems of many levels ask for scientific control: ecosystems…formal organizations like bureaucracies, educational institutions, or armies; socioeconomic systems…. there can be no dispute that these are essentially “system” problems, that is, problems involving interrelations of a great number of “variables.” (Von Bertalanffy, 1972, 420-421)

Today, the application of control via feedback mechanisms to biology, sociology and technology has a new name coined by the Santa Fe Institute: complex adaptive systems.

Complex Adaptive Systems

The study of complex adaptive systems is a collection of theories incorporating several disciplines. The Santa Fe Institute describes their research:

Complex systems research attempts to uncover and understand the deep commonalities that link artificial, human, and natural systems. By their very nature, these problems transcend any particular field; for example, if we understand the fundamental principles of organization, we will gain insight into the functioning of cells in biology, firms in economics, and magnets in physics. (Santa Fe Institute, 2006)

The key to understanding how feedback relates to complex adaptive systems is the word adaptive. Adaptation is a way of moving forward towards a goal; and since “all goal-directed organization demands closed circuits” (Guilbaud, 1959, 21), feedback is an important element in studying complex adaptive systems. As we will see, the notion of feedback has become useful in describing and analyzing business, societal, and economic systems.


Not all complex adaptive systems need a goal. Evolution by natural selection is often described as a complex adaptive system, and yet evolutionary biologists maintain that there is no goal involved, at least not organized by an outside “intelligence.” (This sets aside the theory of Intelligent Design, as evolutionary biologists generally consider it the realm of theology rather than science) (Dennett, 1995). This would lead some to conclude that all complex systems are goal-less; perhaps natural systems are. It doesn’t mean, however, that a pattern found in the natural world cannot be applied to the man-made world. Human beings are goal-making machines; and today, feedback is a large part of the systems approach to business management. In Agile Management, David J. Anderson emphasizes the importance of goal seeking in running a business:

Among those interested in running a business for profit, the notion that planning and control are impossible is a little daunting. The thesis of this book is that software is not divergent and can be planned and controlled. The proper jobs of executives are to set the goal for the adaptive behavior of the system and to select the method with which the goal is measured and the system feedback delivered. (Anderson, 2004, 12)

The goal of software development, as Anderson sees it, is to produce working code that is valued by the customer. To ensure this happens, he advocates a series of feedback loops in the form of periodic tests to maintain working code throughout the design process. Multiple feedback loops allow developers and managers to periodically compare their output to the desired goal, and make frequent adjustments. This results in adaptive behavior, or as Anderson terms it in the title of his book, “agile” behavior (Anderson, 2004). It is easy to see why this style of systems thinking is becoming more commonplace in today’s fast-paced and competitive business environment.


Feedback loops can be broader than projects within a single company. In Emergence: The connected lives of ants, brains, cities, and software, Steven Johnson explains how feedback has influenced the news media. Until the late eighties, local news stations depended on national networks to feed them a reel of national news footage, pre-edited by the top executives in New York. When CNN appeared on the scene, they offered a new strategy that changed the way the system was “wired”: to attract local stations, Ted Turner provided them with full access to the CNN news feed, allowing local news stations for the first time to pick and choose what they considered newsworthy. On January 23, 1992, when then-candidate Bill Clinton denied allegations about a relationship with Gennifer Flowers on-camera, all the top executives at the major networks decided not to run the story. According to the old “wiring,” the story would have been dead. Many local news affiliates, however, having access to the same story, decided to air it. The major networks had no choice the next day but to follow along: positive feedback. The decision process had shifted; instead of a “top-down” system, it was now “bottom-up” (Johnson, 2001). This was the first example of a new media structure which still reigns today: positive feedback loops which result in feeding frenzies on sensationalistic stories and entire shows dedicated to the media critiquing itself.

Not all media is subject to positive feedback loops, according to Johnson. Slashdot, the online community for ubergeeks, has created a homeostatic process via negative feedback in the form of ratings and filters. Comments on Slashdot used to have a tendency to be overrun by cranks and spammers, providing an unmanageable volume of endless and useless commentary. Slashdot, however, developed a rating system that provided unworthy comments with negative feedback. Today, users are able to filter comments based on user rating, tuning out spammers and cranks, essentially creating a “clearer signal” amidst a background of noise. Although rating systems have their disadvantages (such as marginalizing minority viewpoints), Johnson believes that a healthy balance of positive and negative feedback can be achieved by tweaking the algorithms that provide user ratings:

As Wiener recognized half a century ago, feedback systems come in all shapes and sizes. When we come across a system that doesn’t work well, there’s no point in denouncing the feedback itself. Better to figure out the specific rules of the system at hand and start thinking of ways to wire it so that the feedback routines promote the values we want promoted. It’s the old sixties slogan transposed into the digital age: if you don’t like the way things work today, change the system. (Johnson, 2001, 162)


Johnson isn’t the only one to use systems thinking as a model for understanding complex societal behavior. Robert Hagstrom uses this paradigm for understanding stock market investment. According to Hagstrom, investors form models that they believe will be successful strategies in the market, such as the discount-to-hard-book-value strategy, the dividend model, or the cash return on invested capital. Based on market performance (feedback), investors will change their strategy if and when they discover it is no longer successful. Investors’ models compete in the market place and are routinely replaced by new strategies. It’s clear that Hagstrom’s view is analogous not only to mechanic systems, but evolutionary biology as well. His comments about the necessity of systems thinking harkens back to Von Bertalanffy:

The new science is connected and entangled. In the new science, the systems are non-linear and unpredictable, with sudden abrupt changes. Small changes have large effects while large events may result in small changes. In nonlinear systems, the individual parts interact and exhibit feedback effects that may alter behavior. Complex systems must be studied as a whole, not in individual parts, because the behavior of the system is greater than the sum of its parts. (Hagstrom, 2000)

Lengthening the Feedback Loop

We have seen how feedback can be modeled from different temporal perspectives: a refrigerator adjusts within minutes; the media can adjust daily; markets adjust year by year. How far can feedback loops extend? Norbert Wiener addressed this very question in Invention: The Care and Feeding of Ideas, a book published posthumously from one of his manuscripts. If feedback is directed toward an ultimate goal, what do we do when the goal extends beyond our lifetime? What feedback mechanisms does a society require to preserve the environment, for example, for future generations? We need a means of passing knowledge down from generation to generation that will not disappear with the business cycle:

As we have seen over and over again, the conservation of the fertility of human thought is as primary an obligation as the conservation of the fertility of the land. Both of these redound to the generations to come, and can only be carried out by one who feels a responsibility, if not to the eternal, at least to the very distant future…. Unless society has in it some institution or at least some traditionally accepted mode of behavior pertaining to the very distant future, the long-time care of the future needs of the human race is something which falls equally on everyone, and hence falls on no one. (Wiener, 1993)


As I said, once you start looking for feedback loops, you start seeing them everywhere. They are not limited to our household appliances, however. Feedback is more than an engineering technique; it’s a way of looking at the world. Once we start looking at the world through systems-colored lenses, we see the application to biology, technology, business, economics, and even ethics. Norbert Wiener was one of the first to see the implications. In the last six decades, his influence has grown—or should I say descended—from the ivory tower to popular culture. If wisdom consists of taking a long-term view of the world, then perhaps lengthening the feedback loop will expand our understanding of the world, and our place within it.


Anderson, David J. (2004). Agile Management for Software Engineering: Applying the Theory of Constraints for Business Results. Upper Saddle River, New Jersey: Prentice Hall.

Conway, Flo, & Siegelman, Jim. (2005). Dark Hero of the Information Age: in Search of Norbert Wiener the Father of Cybernetics. New York: Basic Books.

Dennett, Daniel C. (1995). Darwin's Dangerous Idea. New York: Simon & Schuster.

Guilbaud, G. T. (1959). What is Cybernetics? (Valerie. MacKay, Trans.) New York: Criterion Books.

Hagstrom, Robert G. (2000). Investing: The Last Liberal Art. New York: Texere.

Johnson, Steven. (2001). Emergence: The connected lives of ants, brains, cities, and software. New York: Scribner.

Santa Fe Institute FAQ. (2006). Retrieved November 14, 2006, from http:/ / aboutsfi/ faq.php

Von Bertalanffy, Ludwig. (1972). The History and Status of General Systems Theory. The Academy of Management Journal, 15(4), 420-421.

Wiener, Norbert. (1993). Invention: The Care and Feeding of Ideas. Cambridge, Massachusetts: MIT Press.