SI521 "Open Educational Resources at the University of Michigan" Open Textbook/PeerProduction

From Wikibooks, open books for an open world
Jump to navigation Jump to search

Abstract[edit | edit source]

Open Educational Resources (OER) are produced using a variety of means. Sometimes, they are developed using the traditional educational model where a single person or group of people construct the educational documents. The alternative to this centrally controlled model is peer production, a method of production whereby individuals build something without specific monetary incentives. For some open educational resources, peer production is the primary means of building the resource.

Much of this voluntary collaboration is coordinated using information technology; therefore, virtually all of the open educational resources that are peer produced are electronic. Examples include some OpenCourseWare projects, open textbooks, open datasets, open health information resources and peer produced educational websites, such as Wikipedia. Peer production is a means of production also used in many domains outside of education. Concrete examples of its impact can be seen in domains such as software and artistic work.

In this wiki article, we will first look at the definition, history and philosophy of peer production. With this as reference, we will analyze successful and unsuccessful applications of peer production. We will then analyze an economic and philosophical critique of peer production. We will attempt to frame as much of this article within the context of system theory; a framework used to process a complex system like peer production.

Definition[edit | edit source]

Peer production is a complex system, whereby the largely non-monetary inputs yield often substantial monetary and non-monetary output.

File:System diag.png

The input into a peer production system is human labor, a shared commons (such as a network platform) and output from other systems. Peer production is often a stochastic process, whereby a fixed input may yield different outcomes. This is dependent on both random factors and qualitative factors of components (agents) of the system.

A peer production system consists of internal components on which it operates. These may be as simple, like mailing lists (electronic and paper) or they may be complex hierarchies with vast computational and human resources. The components (or agents) of this system operate on inputs to produce, an output, typically some form of intellectual property: software, learning tools or art, for example.

Modern conceptions of peer production originated out of the early computing movements of the 1970s and 1980s. Over time, the methods and attributes of peer production have been refined in an effort to improve overall efficiency. Today there are many examples of extremely efficient and structured peer production systems. Many of these systems evolved from a single individual's ideas.

Philosophy & History[edit | edit source]

Complexity[edit | edit source]

We define peer production as a complex system with certain attributes that differentiate it from an ordinary system. [1] A complex system has a "large number of diverse agents or actors interacting with each other in a great many ways"[2] Additionally, these agents largely operate autonomously from other agents without necessarily knowing the current or past state of the system. In his (in)famous paper, Eric Raymond applies this idea to peer production, developing the cathedral and the bazaar model.[3] The bazaar model is a metaphor for the open-source and peer production movements; individuals act rationally and autonomously to build value. This is the inverse of the cathedral metaphor, whereby value is constructed using central control.

Within the realm of complexity theory, systems of peer production are categorized as emergent complex systems.[4] Emergent systems are self-organizing; there is rarely any hierarchical or centralized command structure. Additionally, emergent systems often exhibit a "whole is greater than the sum of the parts" attribute. [5] In peer production systems, the components of the system (the agents) self-direct and contribute small parts to a larger project, much like Raymond's bazaar metaphor. Later in this wiki, we will examine different systems of peer production and their means of self-direction.

Peer Production as Anti-Fordism[edit | edit source]

The Industrial Revolution created systems of industrial and information organization that were oriented towards centralized control. This is reflected in many modern political and economic organizational systems:

  • The means of production were largely oriented towards a rival consumer audience; when goods were consumed by one person they were consumed irrevocably. In a peer production system the outputs are largely non-rival; a piece of software, mp3 file or electronic text can be redistributed at virtually zero cost. [6]
  • The production system required a strict division of labor, reflected in Henry Ford's assembly line. Peer production systems require a division of labor, as well, however this division is largely "self-directed".[6]
  • Organizations of firms were largely coordinated by Coasean "transaction costs"[7] which made production more efficient using large and centralized organizational systems. Peer Production systems have dramatically lowered transaction costs, which affords a more diversified and autonomous workforce.

Unix Philosophy[edit | edit source]

Much of the modern philosophy of peer production is derived from the philosophy of Unix, originated in the 1970s. Eric Raymond argues the Unix philosophy centers on "small tools, rapid prototyping and evolutionary programming". The core tenets of the Unix philosophy are:[6]

  • Write programs that do one thing and do it well
  • Write programs that work together
  • Write programs that handle text streams because that is a universal interface

Indeed, virtually any Unix or Linux user will praise the basic and modular tools available: vi/emacs are useful inline text editors, gcc/g++/j++ are software compilers, perl/php are powerful scripting languages, and there are virtually limitless text scripting commands (sed, cat, more, less, etc). This philosophy of strict modularity and granularity has afforded many open source projects the ability to scale, without requiring a hierarchical command structure.[8]

A Networked World[edit | edit source]

In the early days of peer produced software, developers passed copies of software to each other on physical media: diskettes, tapes and hard disks. Most of these developers, at research institutions, shared the software virtually exclusively with others that they knew. Therefore, the network was constrained to a fairly small group of users. With the growing ubiquity of the Internet, transaction costs for sharing software decreased rapidly. This had the effect of enabling a significantly larger group of developers and users than had been previously available. However, this also greatly increased the complexity of the peer production systems; now they had both greater software and social complexity.[6]

Free Software Foundation & Richard Stallman[edit | edit source]

In 1984, Richard Stallman, a computer science professor at MIT founded the Free Software Foundation (FSF). Stallman presented the normative argument that proprietary software is bad; information and software should be free. He did not argue that information should be sold for zero dollars but rather that information should be:[6]

  • Free to use as you see fit.
  • Free to reverse engineer and dissect.
  • Free to copy.
  • Free to improve; provided you redistribute these improvements.

These requirements form the legal framework of the General Public License (GPL). The GPL was one of the early licensing schemes that formed the framework for many early (and current) peer produced projects. The license guarantees to contributors that their contributions will not be locked into proprietary projects or sold without due credit to the contributor. Additionally, the license is viral, meaning that derivative works of the contribution must also remain free. This legal framework is tremendously important; without the legal guarantees of openness, contributors may refrain from working on peer produced projects.

Examples[edit | edit source]

In Software[edit | edit source]

The most publicized example of peer produced software is the Linux Operating System. Started in the early 1990s in Finland by Linus Torvalds, Linux has grown into a massive project; the Linux Kernel alone has over ten million lines of code. The project has established its own psuedo-control structure. Linus Torvalds continues to manage contributions to the project and there are very active mailing lists. They have significant computational and network resources used to build the project and distribute it. However, contributors to the project are largely at-will volunteers and Linux does not direct anyone to contribute or work on any specific project.

Other examples of peer produced software include:

  • Mediawiki, the engine that powers Wikipedia (and this wiki)
  • GCC, The Gnu C Compiler, developed by Richard Stallman and the FSF
  • Slashdot, A popular discussion forum for technology-related news

Open Educational Resources[edit | edit source]

There are many examples peer produced open educational resources.

  • Wikipedia is the most obvious and largest open educational resource
  • Open Textbooks, such as Wikibooks like the "Introduction to Chemical Engineering" textbook[9]
  • Proteome Commons [10] is a large peer produced repository of scientific data for medical researchers.

MIT OCW[edit | edit source]

When it comes to open educational resources and open courseware, the MIT OCW project is often the most referenced. In one sense, the MIT project is peer produced - teachers and academics post their course information online for the public. It is voluntary and contributors act with relative autonomy. However, the MIT OCW project is highly centrally coordinated. They have an office with a budget that directs contributions to the archive. The MIT OCW project is a hybrid peer-produced system, sharing aspects with both true peer produced systems and centrally controlled systems.

Critique of Peer Production[edit | edit source]

Infinite Monkeys on Infinite Keyboards[edit | edit source]

A Sample Random Image Generated by Monkey. Eventually, we'll get the Mona Lisa

Some critics of peer production equate the contributions to "infinite monkeys on infinite keyboards", alluding to the "Infinite Monkey Theorem". This humorous theorem argues that an infinite quantity of monkeys in front of an infinite number of keyboards will eventually produce the complete works of Shakespeare.[11] They argue that peer production is equivalent to handing over institutions to amateurs and the masses. [12] In the domain of open educational resources, they argue that we are giving control of our educational system to misinformed amateurs, corporations with agendas, and are simply doing a disservice to the users. They argue that much of the contributions to peer produced systems is not properly vetted and the community does not have the interest or resources to properly police materials. After all, even this document itself could be altered by anyone. In the end, by relying on peer production and free distribution, society will destroy its market-driven institutions, such as publishers and booksellers.

Higher Transaction Costs[edit | edit source]

Peer produced resources lack a centralized means of control. Therefore, mechanisms are needed to properly allocate resources efficiently and contributors often make incorrect self-assessments.[8] Contributors may incorrectly assess their own skills; a bad programmer may contribute code to a large software project or an educator may unwillingly contribute false information to their open educational resource. Second, the lack of centralized control means that there is no specification regarding who is contributing what and when. Therefore, the project needs to build mechanisms to help individuals self-select into aspects of the project. These mechanisms increases the transaction cost of the project and in some cases may inhibit peer production as a successful means of production.

Failed Projects[edit | edit source]

In 1996, Netscape opened up its web browser product to peer-production. After some initial successes, the project quickly ran into problems - the opened code was "messy" and required some substantial re-writing. Any contributions made by developers would be at the periphery of the project.[6]

Motivations & Incentives[edit | edit source]

Absent Price Signals[edit | edit source]

Prices are information regarding the value of demand for a good. Through prices, economists can infer information: how much quantity is demanded, how much value to ascribe to a good, how much to pay for production and at what price to sell the good. Most peer production systems operate outside of this domain. Without price signals, peer production requires other means for ascribing value to attract inputs (such as labor).[13]

There are many differing reasons that individuals contribute to a peer production system, without these direct monetary signals. There may be indirect monetary incentives, such as reputation and networking. Contributors interested in developing relationships with for-profit firms may wish to contribute, in order to increase publicity about themselves. In theory, these contributors are using peer production as a form of self-advertisement. While research regarding hiring patterns of developers is somewhat lacking, it is worth noting that many of the largest contributors to open-source software are funded by for-profit software and hardware firms. This type of symbiotic relationship may be of value to both the peer production system and the firm.

Weber identifies other reasons for peer production contributions.[6]

  • The joint enemy - This is more notable during the early days of open-source software when many developers shared a common enemy - Microsoft.
  • Ego Boosting - Writing "clean code" and developing a reputation can be an ego boost to developers.
  • Identity and Belief Systems - Individuals like those in the FSF seek to pursue political and ideological goals; particularly contributing software to the greater good of humanity.
  • "Fun of Programming" and Cool Stuff - The ability to use and contribute interesting information is appealing to contributors.

In the case of Open Educational Resources, contributors may be motivated by any of the forelisted goals. Some educators may wish to develop a reputation by which they can later find a good job. Others may simply want to contribute to the greater good of humanity.

Notes[edit | edit source]

  1. Kuwabara, Ko. "Linux: A Bazaar at the Edge of Chaos," First Monday March 2000.
  2. Simon, 1996, Sciences of the Artificial, p. 195; Waldrop, 1992, Complexity,, p. 11.
  3. Raymond, Eric. The Cathedral and the Bazaar. Available at < >
  4. Truscell, Michael. The Architecture of Information: Open Source Software and Tactical Poststructuralist Anarchism. Available at
  5. Corning, Peter A. (2002), "The Re-Emergence of "Emergence": A Venerable Concept in Search of a Theory", Complexity 7(6): 18-30
  6. a b c d e f g Weber, Steven. The Success of Open Source. Harvard University Press, April 2004.
  7. # Benkler, Yochai. Coase's penguin, or linux and the nature of the firm. October 2001.
  8. a b Benkler, Yochai. The Wealth of Networks : How Social Production Transforms Markets and Freedom. Yale University Press, May 2006.
  12. Keen, Andrew. The Cult of the Amateur. Doubleday. 2007
  13. Coase, R.H. The Nature of the Firm. Economica. 1937.