ICT and the Deregulation of the Electric Power Industry: A Story of an Architect's New Tool

Robb Klashner
Information Systems Department, New Jersey Institute of Technology, Newark, New Jersey, USA
Email: klashner@njit.edu; Web: http://web.njit.edu/~klashner/
Key features: References; Figures 1, 2, 3; Tables 1, 2; Audio clips 1, 2, 3, 4
Editor's note: an extended version of this paper is available from the author here

Contents

Abstract

Deregulation of electric power industry in the USA is an effort by the Federal and State governments to exercise power through the control of mission-critical infrastructure. This research asserts that information and communication technology (ICT) was necessary for this deregulatory effort, but ICT by itself is not sufficient to assure the success of deregulation. Using metaphors adopted from Kling and Scacchi's "web of computing", the paper shows how regulators attempted to change the social fabric in the electric power industry by using ICT to alter a complicated set of interdependencies and complementarities. Given the social and infrastructural nature of the research, web models are an effective mechanism to understand these complex relationships. In the paper, the basic web model is extended with architectural aspects to draw out the original connections with "urban infrastructure" and architecture. Ethnographic methods were used to gain a deep understanding of the ecology of the electric power industry. Data were collected over a nine-month period at a large utility's grid dispatch control center because these centers were the primary focal point of deregulatory efforts. Essentially, the paper is another data point indicating rational models are of little value outside tightly controlled circumstances. The paper's primary contribution, however, is intended for policy-makers changing ICT designs at an architectural level using regulatory mechanisms, and for ICT analysts who must map the resulting complex of ecological interrelationships into an integrated design. The resulting integrated ICT infrastructure is used to run the electric power infrastructure that information societies depend on in the most intimate way on a daily basis.

1 Introduction

1.1 Enabling socially constructed deregulation with ICT

Over the past 50 years, a broad and increasing dependence on technology has given rise to a concomitant dependence on information and communication technology (ICT). As social dependency on technology has increased, attitudes in society and policy have evolved towards technology in general with ICT positioned as the agent of change. Thus, system analysis and design is inherently contextualized in a framework of expectations; further, the evolution of social norms and expectations is influenced by the technology itself as users, workers, and policy makers evolve with ICT systems. This research examines how ICT complements general technological dependencies to determine evolution in mission-critical infrastructures (Kling 1992).

The operation of the electric power "grid" (Figure 1) has been complemented or augmented by the Internet, which played a key role in the evolution of the electric power industry. The motivation for "radical deregulation" was to create an electricity spot market premised on "Performance-Based Rate Making" (Navarro 1996) that could be operated over the Internet, 24 hours a day, seven days a week. However, deregulators physically and psychologically dismembered the electric power culture using ICT. Therefore, as noted by Kling (1992), the evolution of electric power generation, distribution, and transmission is not surprisingly a story of complicated complementarities that is played out over an extensive history, in a rich ecology, and recently with the aid of ICT.


Figure 1. The vision ... "Grid 2030". Possible future state of the grid as envisioned by Siemens and the US Department of Energy
(reproduced with permission of Siemens Westinghouse Power Corporation, from Jimmy Glotfelty, Transforming the Grid to Revolutonize Electric Power in North America, January 27, 2004 slides in pdf)

The ICT-enabled deregulation shifted attention to functional aspects during the grid system design. The shift naturally emphasizes some stakeholder assertions based on assumptions of rationality ignoring many of the negative social and political aspects of computing (Kling 1978a, Kling and Jewett 1994). The positive views of ICT held by the electric power industry, policy makers, and the public have been institutionalized over time (Kling and Iacono 1989) as their dependency on reliable and cheap electricity has grown. The unfounded belief that ICT is an effective "agent of change" (Kraemer and Kling 1985) smacks of utopianism (Iacono and Kling 1996) and/or religious overtones (Nobel 1997). However, ICT directives imposed by governments to stimulate innovation and diffusion (King et al. 1994) are likely an ineffective or incorrect prescription for complex socio-technical ecologies like the electric power industry.

1.2 Multifaceted role of ICT in deregulation

This research began as an attempt to understand the multidimensional and contradictory constraints grid dispatch control center (GDCC) (Figure 2) system analysts were facing as a result of the radical deregulation imposed in the 1990s (OASIS NOPR 1996) intended to optimize power industry operations through market-making mechanisms. How are analysts to factor in rapidly emerging and contradictory deregulatory constraints arising from these state and federal efforts? Without a clear understanding of the role of the ICT infrastructure, system analysts fail to grasp essential socially constructed aspects of ICT architectural designs. They lack a model to articulate the complex of complementarities in a manner that would improve their process or designs. Their limited perspective is exacerbated by the industry's vertically integrated structure using a central monopoly model that resulted in a highly cohesive social fabric, ontology, and culture.

Figure 2. Control Center for the California Independent System Operator (Cal-ISO). This is the high-tech electricity traffic controller for 75 percent of the state's load (energy demand) through control of the transmission system previously operated by the three investor-owned utilities: Pacific Gas & Electric, Southern California Edison, and San Diego Gas & Electric
(reproduced with permission of Cal-ISO from http://www.caiso.com/aboutus/infokit/ControlCenter.html)

The pressing question for the regulators and the industry still remains today since the 1990s deregulation failed. Everyone acknowledges ICT is a necessary change agent to deregulate, but is it a sufficient catalyst to ensure deregulatory success given its technological deterministic application in the electric power domain? To uncover the deeper issues, data was collected at a GDCC rich with ICT using ethnographic methods. The paper describes how complex relationships can be analyzed in order to understand the vague complementarities and dependencies that often elude rational actor models. The research extends the web model in order to show how:

  1. The electric power industry is presented as an evolving complex of actors bound together by the physics of the electric power grid.
  2. Stakeholders seek to manipulate their ecology.
  3. The architectural process can be leveraged as an analogy to expound upon the actual complementarities between the grid system and urban architecture (Kling and Scacchi 1982, Kling 1992).
  4. Some structural, psychological, and social confounding factors create complementarities and dependencies.
  5. Misconceptions regarding ICT enabled the failed deregulation attempt.
  6. Kling was correct, i.e.
    1. System analysts in mission-critical domains need to broaden their scope to include social factors in order to cope with ecological constraints arising from powerful and/or deviant actors.
    2. ICT has numerous multifaceted social dimensions that, if ignored during the architecting of civilizations by policy-makers, will result in unexpected permutations of both computational and social structures, which are likely to become institutionalized.

2 Methods

Kling argued persuasively for empirical research (Kling 1978a, Kling and Scacchi 1982, Kling 1987, Kling 1992) and this research reflects his promptings. Kling noted the need to examine the whole "package" that is "not only the hardware and software facilities, but also a diverse set of skills, organizational units and sets of beliefs about what computing is good for" from a symbolic interactionist perspective within segmented institutionalism (Kling 1980). This perspective is used to examine the interactions between ICT and the electric power industry since it has been gradually and completely immersed in ICT over the last 55 years.

2.1 The web model's natural association with architecture

Kling and Scacchi (1982) use web models as a metaphor to illustrate the web of computing. Simmel's (1964) web metaphor was adopted by Scacchi (2004) as an effective means to view "computing in organizations as a web of group affiliations" that sought "to identify, describe, and interrelate the group affiliations that surround organizational computing". Kling (1992) used the web model to:
"draw 'large' social boundaries around a focal computing resource so that the defining situation includes: the ecology of participants who influence the adoption and use of computer-based technologies, the infrastructures for supporting system development and use, and the history of local computing developments".
Computing infrastructure is analogous to "urban infrastructure", i.e. electric power, telecommunications, transportation, or sewage infrastructure (Kling 1992). Infrastructure concepts and structural properties are tightly coupled. Infrastructural architecture is dependent upon emerging scientific discovery of new compounds and/or engineering techniques to apply building materials in order for architects to create new designs. The spider web is a natural phenomenon that is useful when considering architecture, infrastructure, and utilization of compounds. Recent research into spider webs has revealed some interesting structural properties that augment the web model metaphor. Spider webs are presented in this paper to draw out a deeper architectural analogy with the connection to urban infrastructure in order gain additional analytical leverage by providing a visual image of the utility's control mechanism.
  1. Spiders are very territorial (like geographically bounded utilities) and "too anti-social to farm successfully" for their web proteins (BBC 2000). (We mean no negative connotation in these comparisons. The utility personnel we observed/interviewed have very high moral and ethical standards. Their primary focus is to maintain service for the public's welfare.)
  2. According to Hayashi the spider's web has amazing structural properties (Tenenbaum 2001). There are seven types of silk a single spider may utilize as building materials depending on the situation:
    1. A type for wrapping prey
    2. Another type for wrapping its eggs
    3. Five types to construct the web itself.
  3. Scientists and engineers are actively researching spider silk, which is 25 times stronger than steel and has an amazing elasticity (BBC 2000).
  4. However, "a process as complex as spider silk spinning has never yet been replicated. This formidable challenge is now being tackled on three fronts. The genes coding for spider silk are being sequenced, and researchers are trying to work out the molecular architecture of silk threads and the way they are spun" (Magdalena 2001).
  5. The evolution of the spider has resulted in some embedded structural processes reflecting very specific knowledge.
Analogously to the spiders' ability to create the complex from the simple, utilities often utilize their evolved knowledge to combine simplistic components into complex architectures. Architecture is a confluence of art and technology to communicate space and place. "We are looking for the reflection in architecture of the progress our own period has made toward consciousness of itself. Architecture can give us an insight into this process." (Giedion 1967). Spider webs and electric power lines are ubiquitous in the US ecology, thereby communicating their ever-present influence on the place we share. Both spiders and electric power utilities evolve their architectural design by finding new ways to apply existing "technology" to control their place (Table 1).

The spider proteins are complementary in the design of a particular web just as artifacts are common in human created ICT infrastructure. Software that integrates software or hardware components is often referred to as "glue code", a clear connection with the spider web production process or the merging of technologies by utilities. There is yet another set of complementary and reciprocal interactions one can see in the analogy with "urban infrastructure".

2.2 ICT architectural processes in the web model

Web models can be juxtaposed with architectural processes to gain additional analytical leverage over complex ecologies, history, and ICT infrastructure. Software processes are software too (Osterweil 1997), but still leverage knowledge dynamically inferring a socio-technical semi-structured process. Architectural processes for buildings leverage the architect's knowledge of the ecology and emerging technologies (Giedion 1967) to create new design ideas within a well established social process including policy-making through zoning and building codes. However, the actual architecture is brought to life using other human resources who understand these social norms in conjunction with technological resources that include ICT (e.g. AutoCAD). The combination of ICT processes and architectural processes is not meant to infer a Parsonian framework (Parsons 1951), but a socially adaptive and self-aware process.

Conventional building architects respond to new building materials or processes by creating new building architecture designs using new computer numerical controlled manufacturing (MacKenzie and Wajcman 1992) that releases them from many of the traditional space restraints (Perry 2004). "Bringing computing to the scaffolds promises to change the work practices, organizational structures, and productivity of all of the actors involved in construction projects, including architects, contractors, subcontractors, and labor groups" (Boland et al. 2003). These new ICT tools emerged from a series of joint social and technological interactions.

For example, far-reaching possibilities exist for "bucky tubes" (i.e. carbon nanotubes) (Figure 3); due to their amazing properties the materials could be used in the production of cars, roads, or architectures. Bucky tube elevators could reach outer space, thereby replacing cumbersome space shuttles. Besides being stronger than any other building material currently known, bucky tubes can transmit data. Regardless of the feasibility of bucky tubes, the concept of every car being a node on a network comprised of all roadways is thought-provoking. The integration complexity of that sort of socio-technical architectural design is clearly a "wicked" problem (Rittel and Webber 1973) having no real right solutions, only good or bad ones.


Figure 3. Model of a bucky ball (fulleren) and carbon nanotubes
(reproduced with permission of NCCR Nanoscale Science from http://www.nccr-nano.org/nccr/media/gallery/gallery_01/gallery_01_03)

Modern architects attempt to control ecology through the manipulation of the laws of physics using ICT (e.g. AutoCAD) within a socially and historically context-sensitive process (Table 1). A conventional approach to these problems uses rational actors within "discrete-entity" models that are "a-contextual, a-historical, and assume that adequate infrastructure can always be available as needed" (Kling 1992); ignoring the societal or ecological influences.

Hughes (1992) argues that the early electric power industry evolution is a good example of society applying engineering and scientific knowledge in a collaborative manner to difficult problems that are holding back progress. He refers to these constraints as "reverse salients", but lacks the structural view of ICT. Modern interrelationships between data acquisition, storage, and analysis led to an electric power industry-wide adoption of regulations and technologies for voluminous telemetry, which of course created a push for more storage. This cycle of events created the basic building blocks (i.e. metaphorically like the evolution of the spider's algorithm to assemble proteins) driving the evolution of a complementing ICT infrastructure.

Table 1: Corresponding aspects in architecture, spider web, and electric power
Built architecture
Spider web
Electric power
Interlacing beams Interconnected webs Power line networks
Communication via vibration Communication through web strands movements Analog charts reading fluctuating current
Connect to surroundings Connected webs at border Intertie connections
Tensile strength Tensile strength Load capacity
Control of inhabitants Control of passing prey Control of population
Knowledge to analyze and design for construction Context-specific construction capabilities Plant and equipment knowledge used in design
Clear boundaries Territorial Geographically defined
Reflects space and time Reflects space and time Reflects space and time
Context-specific adaptation Environment adaptation Context specific adaptation

2.3 Data collection using ethnographic techniques

Ethnography has also been used because of its direct applicability to interactionist-empiricist research (Kling 1980). Ethnographic research techniques (Spradley 1989, Emerson et al. 1995) were used to collect data at the GDCC through observation and in-depth semi-formal interviews.

The intimate details of the respondents' control center ecology facilitate partaking in the community's ontology. In the past, a self-imposed mandate of community service, open communication among utility personnel, sharing of technology, and an attitude of camaraderie contributed to a stable organizational structure and operational processes pervasive throughout industry. Informants shared how they felt a stable commitment to stakeholders resisted change thrust upon them in the form of negative policy-making.

Ethnography complements the web model (Kling and Scacchi 1982, Kling 1992) since its foci is the ecology of the participants, infrastructure supporting their activities, and history leading up to the current situation. Ethnography allows the researcher to first observe their ecology and interaction with various infrastructures. Semi-formal interviews clarify many of the intricate details observed early in the ethnographic process such as how and why certain ICT infrastructure is used, which often reveals a common history so important to understanding the current ICT architecture.

Ethnographers are a type of instrument who function better as observers if they are ignorant of ecological details, e.g. the richness of a foreign culture is often brought out in more intricate detail before the observer "goes native". The computer infrastructure, rich history, and ecological unity of the field site, the GDCC, were unfamiliar to the observer. Therefore, the staged engagement of observation, casual conversations, and then semi-formal interviews augment the data collection.

2.3.1 Details of GDCC field site research

One field site researcher conducted random interview data collection from September 1996 to June 1997. Management was willing for the research to be conducted without restrictions. GDCC systems analysts were beset on every side with new and contradictory constraints arising from internal and external stakeholders. The GDCC conducted two 12-hour shifts every day of the year. Each shift had five operators and a supervisor with some additional floating operators to cover for vacations, etc. All personnel on both shifts were observed (including information system analyst meetings) and engaged often in informal discussions. All supervisors, upper management, information system analysts, and at least one individual from each duty station were semi-formally interviewed, giving a total of 30 hours of taped interviews. In addition, sporadic observations, conversations, and interviews were conducted with operators acting in the marketing function and support staff such as accountants or forecasters. Also, when an event occurred (e.g. a disturbance in the system causing a fault alarm to be set off) the observer was allowed to stay in the GDCC.

The observer of this extended interaction and analysis of industry data and standard operating procedures became immersed in the GDCC culture. The observer participated in meals on and off site. In essence, the observer went "native" to the degree that at one point a joint 1-day tutorial was conducted at a conference with GDCC personnel. After data collection in the field ceased, the GDCC data was triangulated with the appropriate engineers in academia for clarification and to get a non-utility viewpoint for tangential research activities. IEEE electric power standards-making processes and meetings were also utilized as an external validation of the data.

3 Historical and social context in electric power

The ethnographic interaction with the informants was essential to the study because of the glut of historical and current data available in electric power industry. For example, entire libraries have been established to honor Thomas Edison’s work. The informants provided the guidance necessary to closely examine certain aspects of their domain. They did not knowingly do this as a guide would, but the guidance emerged through the ethnographic process. As each informant painted their worldview of the electric power industry, new correlations emerged between seemingly innocuous events, technologies, people, organizations, policies, or other miscellaneous data.

The importance of the informants' domain knowledge was translated into simple economics near the end of the field research. Both the California Independent System Operator (CA-ISO)and the Power Exchange (an intermediary that is now bankrupt) hired away (i.e. "hijacked") key grid dispatch personnel with lucrative salaries. These organizations had to pay these salaries because the domain knowledge necessary for their operations could only be acquired from these experts. Similarly, the only way for an outsider to truly understand the intricacies of the GDCC was to have the informants share their intuitive and historical expertise over an extended period of time, such as one experiences during an ethnographic study.

The problems of regulating information system (IS) design were evident from the ethnographic data, which acted to contextualize a myriad of domain-specific scenarios that played out during deregulation. The nuances affecting IS design did not show up in the historical or public literature, but emerged from the qualitative data. Nevertheless, understanding some of these historical implications will empower the ethnographic data sprinkled throughout this paper. Table 2 presents an historical timeline showing how the electric power industry -- the largest industry in the USA as of 1994 (Brennan et al. 1996) -- has evolved through effective marketing, social factors, and political maneuvers (Hughes 1983, Hughes et al. 1987, Cowan 1992, Hughes 1992, Hirsh 1989).

Table 2: Historical timeline and increasing ICT dependency
Date
Event
Structural analogy
ICT dependency
1800s Gas lamp dominates Existing distributed infrastructure N/A
1850s-1880s Edison emerges on scene Centralized plant and equipment for service Rudimentary communication, via electric power lines exhibiting behavior, gives rise to early knowledge workers who leverage grid architecture
1880s AC/DC tradeoff analysis Inventions impact network configuration and scope
1890s-1930s Mergers and acquisitions Central plants became substations. Extending lines of service created command and control problems, i.e. a structural tension Introducing tension into system was one driver or precursor for computerization
1935 Public Utility Holding Company Act Created a vertically integrated structure with a natural monopoly foundation Perfect organizational structure to maximize centralized computing architectures
1950-1970s Increasing customer base with the associated need for significant data collection and management Populating nodes shifts emphasis from a "spur" to network configuration resulting in the increasing structural fortification at key points, e.g. lattice structure Incremental growth in mainframe computer facilitates and telemetry to each utility's control center
1973-1974 Oil embargo by the Organization of Petroleum Exporting Countries (OPEC) (Anon 1973) Reconfigured supply chain; economic strain on industry slows construction of oil-based generation facilities Increasing dependence on simulation and modeling to forecast demand and supply
1978 Public Utility Regulatory Policies Act of 1978 (PURPA) forced utilities to purchase energy from unconventional sources Evolved structure in a manner necessary to allow for access to previously restricted parties, e.g. similar to structural modifications to allow handicap access Integration of data with parties previously not required; new types of economic models utilized
1980s Deregulation of other urban infrastructure, e.g. gas (R. J. Karg Associates 1998) and airline industries (Meyer 1999) Changing public, state and Federal opinion created pressure to change structure ICT (e.g. Sabre's airline systems) seemed to enable restructuring
1990s Increasing technological collaboration of utilities, e.g. Electric Power Research Institute (ERPI) budget nears $1 billion Unification of structures due to the common utilization of technologies supported by shared industry research Extensive proliferation of electric power ICT to optimize operations and control structures
1992 Energy Policy Act of 1992 Mandating the use of new building materials and opening the door for unconventional contractors to do the work Internet technologies replace 100 year dependency on telephone for real-time and primary communication

3.1 Short early history of electric power

Thomas Edison's holistic method of electric power innovation reflects the "invention, innovation, and diffusion" process (King et al. 1994). It was based on a "systems approach" (Hughes 1983), which extended to all artifacts and actors in their socioeconomic context. Edison's chose to use technology to drive societal control in a way that was consistent with his vision for radical change (Hughes 1983). Edison and his Menlo Park team were driven, goal-oriented, and shrewd businessmen for the most part (Hughes 1992). They were able to apply technology with the right amount of marketing to advance their agenda (Hughes 1983). Trained disciples of Edison carried his message and corporations supported their agenda. "Menlo Park, Edison's research and development institution, was only an aspect of the beginning of the great transformation brought about by the large-scale, systematic harnessing of science and technology to corporate objectives" (MacKenzie and Wajcman 1992).

3.1.1 Industry evolved through technological dependencies

Standards and protocols for interactions and exchange of electricity were needed as larger utilities made alliances (David and Rothwell 1996). This fact is due to the physical properties of electricity, which travels towards the path of least resistance, thereby instantaneously equalizing or balancing all utilities at the same frequency (i.e. the number of cycles per second). The industry in the USA gravitated to the 60 hertz standard as the industry picked up "technological momentum" (Hughes 1983, Hughes et al. 1987). This choice, as one informant pointed out, subtly drew the industry into developing enormous and extremely expensive generation facilities that were locked into a narrow range around 60 hertz. This approach created a path dependency (David 1985) having extreme implications on today's grid if poor frequency developed [explanation: implications of poor frequency: text image].

A natural outcome of the historical progression of the electric power industry [explanation: text image] was the ongoing confrontations among surviving utilities (Hughes 1983, Hughes et al. 1987). These entities would geographically and metaphorically collide as a result of market pressures. These situations often resulted in a stalemate, since neither party could compromise or acquire their neighbor(s). The stalemate naturally evolved into collaboration and the interconnection of utilities. These "ties", or "interties" as they are called, are a major power grid development, which has both increased reliability and complexity simultaneously. Ties exist physically as large power lines crossing utility boundaries. The amounts and quality of electricity coming across these ties can vary. Conceptually and physically these create one large grid formalized as regions.

A utility's electric power generation and distribution activities were historically run by a small group of operators or dispatchers (because they allocate or dispatch grid resources) in the GDCC. These individuals usually possess years of service in other parts of the company. In their original capacity, operators usually start as lineman (electricians who handle the power cables), station operators (similar to the primary grid control center operators, but closer to the physical problems), or at some other job "in the field". The traditional utility personnel had their entire mindset focused on control with the ultimate goal being perfect reliability through cooperation.

3.2 Technical aspects of the grid center

The GDCC field site is a secure facility located inside a campus of other buildings. The actual command and control center is visually impressive. The walls of the tall, narrow, curved room are covered with their worldview, painted in brilliant colors and punctuated with various size lines as symbols of its importance. As you focus on the familiar geographical names printed where the lines intersect, you realize that the GDCC is the nerve center for thousands of square miles filled with police and fire stations, hospitals, schools, manufacturing and financial industries, residential areas teaming with homes, endless streets, and millions of people. Consider the task: thousands of square miles containing every kind of geographic terrain imaginable and populated by millions of individuals demanding the product 24 hours a day, everyday, consistently, faultlessly. Customer demand creates a "load" on the grid. Customers are in control because they have complete access to the facilities. If the system is operational, a customer can get product in any quantity, at any moment, with no advance warning. The US has grown dependent on that instantaneous power fix (Hughes 1983, Nye 1998).

The informants at the field site repeatedly emphasized that their grid is a real-time production and delivery system with no inventory or stored electricity. This point must not be forgotten. Data created by generation and transmission instantly becomes feedback, which is then used for additional production and transmission. Only the expected amount of energy to be consumed is scheduled for production. Industry conventions and standards regarding the acceptable variance to these calculations have changed due to deregulation and the introduction of an electricity market. The schedules are created by utilities based on estimates of customer usage (i.e. load). Forecasting is based on factors exhibiting a strong correlation with load such as previous hour and/or prior day at-same-hour consumption rates, seasonal historical data, and numerous weather reports. The predictions are a major component of power production. They facilitate the optimal configuration of resources resulting in the greatest profits within a set of constraints, but only during the most common operational situations.

There is not one, but three situations or "states" of production [explanation: text image] that exist within the field site, which must all be considered for security, reliability, and economic reasons. These separate production situations are similar in some respects, but function very differently depending on the implications of the given condition of the grid. Let the three states be referred to as: standard operating procedure, emergency, and abnormal. The production states are all addressed at some level by equipment and ICT. However, the role of ICT changes dramatically in certain state changes. The distinctions between these three situations call for the separate approaches and distinct actions that are socially constructed and evolved. These states and industry culture contribute strongly to GDCC group behavior and are based on expected normative actions when a state change occurs.

The current "generation" of GDCC operators and dispatchers has participated in social, organizational, and ICT developments they interpret as historically significant. An important characteristic of the electric power industry of the past was the free exchange of information, ideas, and intellectual knowledge. Operators could use personal knowledge about the other utilities to offset ambiguity when the operations are disrupted (e.g. blackout). This understanding is not a speculation by the operators, but rather first-hand observations. It was common to have operators take onsite tours of other grid control center facilities to exchange technical information. Those practices of information sharing have mostly changed as a result of policy changes.

3.3 Initiation of deregulation in pre-1990s

The evolution of the system since the Public Utility Holding Company Act of 1935 had been towards stability. The electric power industry's argument for its classification as a natural monopoly had been effective. Natural monopoly status gave it a great advantage as it pressed forward in its attempts to control the reward system (Hughes et al. 1987). The Public Utility Holding Company Act and a 50-year-old tradition among utilities naturally led to further vertical integration. However, the vertically integrated utility had a major problem: command and control of huge inventories of assets used for the simultaneous generation-transmission-delivery of electricity over geography did not facilitate real-time collaboration activities.

GDCC teamwork and communications among the handful of dispatchers evolved to a fluid, nearly transparent, activity because it was needed in mission-critical operations. The intertie management process used at grid dispatch relied on three sources of data [explanation: text image] to monitor and control the grid. The tasks designated for computer support were expanded over time as the grid began to be considered a complete entity. Individual utilities began adopting ICT innovations into their specific operations, thus increasing their complementarities (i.e. via physical and computational interties). The unification of ICT and GDCC operations resulted from the pervasive nature of interorganizational dependencies:

The organizational emphasis on the ICT-managed interties grew over several decades [audio clip 1 with transcript]. This focus was a natural progression because the utilities were able to accelerate their existing management/political plans (Kling 1978a, Kling and Iacono 1984, Kling 1987) of vertical integration. Modern substations that were originally run by operators were incrementally enhanced with ICT until they were fully automated (Kling 1978a, Kling 1978b). The logical result (George and King 1991) was a completely centralized control paradigm reflecting a conscious evolution towards vertical integration legitimized by public opinion.

Only when President Carter "unwittingly challenged the supremacy of utility elites" (Hirsh 1999) by signing into law the Public Utility Regulatory Policies Act (PURPA) (Anon, undated) did the industry begin to suffer significant setbacks. Utilities consume power from several sources: steam-driven generators, hydroelectric plants, nuclear plants, wind generators, and across interties from other utilities with similar facilities. These various generation facilities obviously have different fuel requirements. When the law was enacted, the utilities were forced to consider a complicated algorithm that was computationally and data intensive, which could not have been utilized on the large scope and scale of the USA without using ICT. The resulting supply chain was not envisioned by anyone, and severely curtailed the utility's capabilities.

3.4 ICT-enabled deregulation in the 1990s

The Federal Energy Regulatory Commission (FERC) mandated deregulation in Order No. 888 and No. 889 (OASIS NOPR 1996) after the Congress passed the Energy Policy Act of 1992. Deregulation was predicated on ICT as a change agent, thus dramatically changing the centrality of ICT to the whole electric power regime. The changes FERC mandated had numerous consequences at the state and local government levels. Specific changes were required in the way power industry organizations were to restructure and conduct business in the future. One significant ruling by FERC (OASIS NOPR 1996, p. 145) was that marketing individuals could have no physical or communicative contact with individuals who had knowledge of transmission capacity:
"These standards (of conduct) are designed to ensure that a public utility's employees (or any of its affiliates' employees) engaged in transmission system operations function independently of the public utility's employees (or any of its affiliates' employees) who are engaged in wholesale purchases and sales of electric energy in interstate commerce. Such separation is vital if we are to ensure that the utility does not use its access to information about transmission to unfairly benefit its own or its affiliates' sales."
FERC's anti-collusion regulatory effort used new forms of ICT as the primary purpose of disrupting the established, streamlined industry ICT processes, an obvious attempt at political control through ICT (Kraemer and Kling 1985).

The role of ICT in radical deregulation was to complement the architectural repartitioning of utilities along generation-transmission-distribution lines. Deregulation intentionally altered organizational structure to reduce collusion among the utilities (OASIS NOPR 1996). Deregulators put their new industry architecture in place by prescribing ICT-facilitated processes that acted as new structural extensions to existing configurations, which allowed deregulators to break up vertically integrated organizations and separate internal utility marketing functions.

FERC (OASIS NOPR 1996) technologically deterministic actions using ICT were suggested by two working committees that were populated by industry, public, and marketing personnel. The groups were named the "What" and "How" working groups (OASIS "What" 1995; OASIS "How" 1996). Information systems analysis, software engineering, and requirements engineering have utilized the "what versus how" metaphor for years. The coincidence is too strong. Therefore, the regulatory process has taken on a distinctly ICT architectural design and developmental flavor as regulators mandate the societal adoption of ICT as a change agent with a distinctly rational predilection. FERC went so far as to include ICT design and implementation specifications in its Notice of Proposed Rulemaking (NOPR).

3.4.1 Using ICT to disrupt institutions

The stable operations of the GDCC had become institutionalized because of the numerous constraints associated with the process (Kling and Iacono 1989). A disruptive element was interjected by:
  1. Outlawing face-to-face communication between previous coworkers
  2. Mandating new ICT communication mediums such as dependence on the Internet
  3. Highly structured forms of ICT interaction.
These policy changes forced open communication gaps using Open Access Same-Time Information System (OASIS) Internet sites where non-utility third-party marketers could then enter and participate. This activity created a billion dollar market economy based on the separation of generation and transmission as a bundled product. The mandated usage of ICT by FERC and State Governments are actions accurately described as an "Innovation Directive". In this case, the innovation directive appeared to work at the beginning, but actually created more problems.

Forcing the utilities to reorganize their internal personnel and finely tuned ICT-complemented processes destabilized the routinized work (Kling and Iacono 1989). The reorganizing weakened the effects of technologically enhanced ecology resulting from ICT integration during prior evolution. The deregulation eroded the old functioning ICT model and "de-tuned" grid operations.

Field site informants correctly foresaw that policy makers would have to revoke some of the changes. However, the majority of informants felt that the industry would architecturally evolve into a drastically different structure from the old configuration. They rightfully expected the current evolutionary process that is due mostly to high-profile events such as the collapse of California's electric power market and the 2003 Northeast blackout, both of which are attributable to a great degree to ICT. Their insights alone demonstrate the value of empirical study.

How did such a stable and conventional industry allow its service reliability -- the mantra of the utilities -- to be eroded by the opinions of the uninformed? The answer lies in the seductive nature of ICT and technologically determined value system; i.e. the industry's "can do" mentality. The misconception that ICT is deterministic led the policy makers to become system architects to go beyond safe boundaries in an effort to further optimize systems. Pro-restructuring lobbyists rarely enumerate the social ramifications of ICT adoption. The next section expounds on these reciprocal socio-structural relationships.

4 An ICT story of complementarities

4.1 Structural dependencies

An electric power utility's plant and equipment, scheduling, and human resources have been tightly coupled over the last 50 years. These three resources are connected to other utilities through multiple channels such as: The interplay of these resources demonstrates the architectural aspects of the web model during the evolution of this industry. The schedules were the architecturally weak link of the interorganizational collaborative effort. Over time the consumer population and demand grew, distribution networks became dense and long distance high-voltage transmission lines became more critical to meet demand.

When schedules were in paper form the telephone infrastructure complemented the scheduling process; as volume increased, spreadsheets were used to complement the scheduling process. The Internet replaced the telephone as the primary communication device. This seemingly benign architectural shift of focus consolidated information and communication into one technological medium, effectively transferring one of the primary inter-infrastructural dependencies (i.e. electric power on telecommunications) to a dependency on ICT. However, when viewed through the lens of web models, the system analyst understands that events that include ecology, history, infrastructure, and massive architectural reconfigurations are never benign.

4.1.1 Utilities undermined by their dependence on ICT complementarities

The use of schedules as a key communication artifact is closely tied to the organization's dependency on ICT to enhance mission-critical data. These dependencies are reciprocal. Economic dispatch of the grid without disruptions is mainly based on dependable schedules, which needed to be in place prior to each business day cycle. Sometimes marketers would have to work late into the nightto work out interorganizational inconsistencies. All agreements to purchase or sell energy coupled with transmission were based on the needs and/or additional capacity of their respective utilities.

Schedules have significant weight in the organizational culture and are similar to the "genre" concept, which is typified communicative action in response to a recurring situation (Yates 1989, Yates and Orlikowski 1992, Yates et al. 1995). Having access to the schedule, or being excluded from its ongoing evolution, was closely associated with legitimate power structures. The critical nature of schedules is demonstrated by their pivotal role in deregulation.

The schedules were a primary target of entities such as Enron (an external "marketer") who sought to break into the electricity "wheeling" market ("wheeling" is the process of moving electric power from a point of generation across one or more utility-owned transmission and distribution systems to a retail customer) created by intertie technology and utility collaboration. The schedules were viewed as a chit one needed to enter the wheeling game. This supposed exclusivity was the premise for the deregulation argument put forward by the "marketers".

The marketers' position fostered a belief that deregulation would work and was essentially a "computerized movement" [explanation: text image] (Iacono and Kling 1996). They contended that the spreadsheet-based schedules could easily be moved into an Internet medium, facilitating the creation of a transmission market. The resulting FERC mandate (OASIS NOPR 1996) separated internal utility marketers in line with this movement.

4.1.2 Collusion or social fabric? A deeper look at schedules

ICT facilitated the evolution of the electric power industry by making complex collaboration possible because it lowered the barrier to effective communication. The entire industry is collaboratively operating the grid -- a synchronous machine -- to deliver product traveling at light speed (Unterman, undated). Historically, marketers scheduled energy consumption for months, weeks, days, and hours in advance.

The fine-tuning of schedules must be coordinated because the industry's control centers operate on a customized hourly basis [explanation: text image]. The GDCC marketer purchases or sells power based on importance to the industry and utility, respectively:

  1. reliability concerns stated by the GDCC operators
  2. economical concerns as the marketer understands them
The use of verbal agreements was an integral aspect of the historical energy marketing function. The utility marketers' predictions are based on: An interesting characteristic of marketers was their ability to "read" their counterparts in other organizations in order to gain an advantage -- as  in a game of poker -- as this quotation infers [audio clip 2 with transcript]. Often, these individuals had close friendships because of the tightly knit community and academic-like sharing of knowledge prior to deregulation. After completing their predictions, the marketers would verbally arrange a transaction for the purchase and transmission of the energy between utilities. The constant interactions between marketers from differing utilities via technologies such as the telephone and facsimile machines was complemented by ICT prior to deregulation. The shared utility ontology was the framework for social interactions [audio clip 3 with transcript]. The ICT infrastructure was more than a technological solution; it created dependencies in the existing social network.

4.2 Psychological dependency on control technologies

There exists a psychological dependency on successful technology, which often prohibits society from reverting to an older socio-technical paradigm. The architecture of the grid is not only a result of technological factors, but of social arrangements. The continued use of rational actors and discrete-entity models (Kling 1992) reveals psychological aspects of the electric power industry's evolution through deregulation.

Proponents of deregulation argued in favor of Internet-as-tool to re-architect the industry, thus completely ignoring the underlying social aspects. Experts in electric power domain effectively argued for a technologically driven deregulation solution in spite of obvious historical evidence (Hughes 1983, Hughes et al. 1987, Hirsh 1989, Hughes 1992, Hirsh 1999) that actors are not rational and technological determinism cannot capture the complexity of mission-critical infrastructural ecologies.

A purely technological version of reverse salients (Hughes et al. 1987) could explain the evolution of the electric power industry if technological determinism were correct. But then how could that deterministic model explain why other societies address reverse salients in different ways. Specifically, many first, second, and third world countries have a very different approach to electric power generation, distribution, and transmission even though they have had access to the same technology as the USA. One social explanation of the phenomena may be that it correlates with differing distributions of traditional religions, or religions surrounding technology itself (Nobel 1997).

There are numerous complex, individual psychological dependencies on the skyscraper architecture, e.g. aesthetic, self-actualization, security, etc. Skyscrapers are socially justified in many ways (e.g. urban planning constraints). It would be generally unthinkable for many architectural firms to abandon skyscrapers for architectural, economic, and socio-cultural reasons. We could no more revert to mud huts than a spider could revert to survival without sophisticated webs.

The point is, when one dominates the ecology the result is a certain psychological dependency on the technology facilitating the control. How many of us could stand to use personal computers if we had to regress to a command line prompt, no connectivity, and Intel 8086 processor? We have grown accustomed to flexibility and connectivity, which are means of controlling our ecology.

The electric power industry, regulators, and the consuming public have a psychological dependency on electricity that did not exist 150 years ago. The creation and spread of electricity has been complemented by the emergence and growth of ICT, which maintains the cohesive connection extending from the electric power utilities' control centers over thousands of square miles of wire, switches, generators, and people. This control has allowed the vertical integration, dynamic grid configurations, and economic dispatch of electricity. Electric power organizations are finely tuned to behavioral shifts of the public and policymakers and, for the most part, are themselves tightly knit in an inextricable social web (Kling and Scacchi 1982, Kling 1992).

The shared mindset in the USA is that ICT could fix the grid's problems after each major blackout. ICT automation was built into system components as fail-safe measures to counteract major grid blackouts such as occurred in 1965 and 1977 (Ellis 2003). The lack of tolerance for even nominal electricity disruptions in the USA (when compared to the past or other modern countries) may indicate a deeper evolving psychological dependency influenced by ICT. This psychological dependency on technology was echoed again shortly after the 2003 Northeast Blackout (Ellis 2003) when the public and policy makers expressed the irrational belief that the grid could be manipulated at will, regardless of its socio-technical architecture.

This mentality cultivated regulatory changes in the latter portion of the 20th century that ultimately lead to the deregulation measures of the 1990s, which have failed for numerous reasons (Hogan 1993, Hogan 1994, Hogan 1995, Navarro 1996, Oren 1997). Instead, ICT-complemented market economy has fostered unreliability and high prices. However, the commitment by policy makers to use ICT because it was not at fault has not wavered. This belief flies in the face of empirical data [explanation: text image].

4.3 Socially confounding factors that defy normal analysis

The empirical data from this study indicate conventional rational, political, managerial, or institutional models would fail to capture all of the important nuances associated with ICT. The dependency relationship [audio clip 4 with transcript] between the public, electric power industry, and ICT are complementary in ways not easily explained. Though massive power outages are rare, when they do occur the resulting criticisms in the media and government have become increasingly sharp.

4.3.1 Behavioral expectations based on trust in ICT

The US public has two primary irrational expectations regarding electric power:
  1. no power outages - ever
  2. absurdly inexpensive energy
The electric power industry has somewhat successfully addressed the first expectation with numerous redundant computer systems (Ellis 2003). To address the second, legislators have forced the deregulation of the electric power industry due to the arguments of special interest groups and economists. Informants at GDCC shared the following views: The behavioral miscalculations are not unidirectional. The industry is equally dependent on ICT as the public and policy makers. The utilities incorrectly analyzed the other stakeholders' dependency on energy. The industry incorrectly counted on the US dependency on reliable energy to swing public and policy maker opinion their way because of the rapidly changing perspective on ICT (e.g. Internet usage), which would help their case. Therefore, their total combined rising expectations on ICT were tightly coupled to expectations of extended ICT benefits resulting in further dependency. Fuel price economics and public pressure over environmental issues are just two other major forces pressing upon the utilities.

5 Discussion

5.1 Advent of ICT-driven deregulation

Strong lobbyists raised public and policy-maker expectations by focusing on the evolutionary possibilities of coupling ICT-augmented GDCC processes with the Internet infrastructure. During the 20 years prior to the 1990s, electric power utilities successfully: Unfortunately for the electric power industry, their effective use of technologies like ICT raised the awareness of pivotal stakeholders -- such as external marketing organizations, the government, and the public -- to the economic possibilities through deregulation.

5.1.1 Monopoly merger and acquisition: deregulation

Marketers entering the electricity industry hoped to generate large profits using the existing ICT infrastructure to manipulate integral and complex grid interconnections. The merger and acquisition style of "Marketers" such as Enron emerged in architecturally-specific regulations for ICT expressly aimed at organizational reconfiguration. The possibilities bound up with the electric power industry's use of ICT, complemented by Internet accessibility, resulted in a change agent that set political forces into motion leading to the deregulation efforts.

The computerization movement (Iacono and Kling 1996) started by "Marketers" largely succeeded in convincing the policy makers and public because of deep-seated general psychological beliefs about technology and specific dependencies on ICT. Their lobbying efforts culminated with the OASIS NOPR (1996). The utilities in many states are now divided. The schedules have subsequently largely been moved to the Internet  through the implementation of OASIS Web sites. This development has resulted in effectively changing the interactions between most utility personnel.

5.1.2 Failed deregulation

Long-standing industry processes conflicted with the prescribed deregulated processes and subsequently raised intra- and inter-organizational tensions, which contributed to the failure of deregulation (Hogan 1993, Hogan 1994, Hogan 1995, Navarro 1996, Oren 1997, Hirsh 1999). Proponents of OASIS did not foresee many of the social nuances of ICT use that would emerge after deregulation had begun. Specifically, informants at the field site described situations wherein generators were using the inadequacies of OASIS Web sites to "game" the system. These tactics are similar to those described by DeMarco et al. (1996).

ICT-complemented deregulation could and, therefore, did result in further collusion, but not necessarily by the utilities it targeted. Enron, one of the major proponents of computerization, was able to drive the California electric power market based on its insider knowledge of industry ICT (Kamp 2002) and its organizational predisposition to cross legitimate business boundaries. Enron did not discover a new technology the utility personnel were unaware of, but they did discover how to manipulate the socio-technical ontology of the ecology; something technological determinists refuse to acknowledge.

5.1.3 If only they had listened

Technological determinists, in the electric power industry, regulatory agencies, and governments, made some hard and fast assertions about various aspects of the GDCC ecology and ICT infrastructure, which were codified in OASIS NOPR (1996). The use of the OASIS "What" 1995 and OASIS How (1996) advisory groups aligns with traditional software industry practice, demonstrating the centrality of ICT to the whole electric power regime.

FERC must rely on electric power industry experts who, in turn, must seek out ICT experts. However, because they shared the same technological belief system and neglected the relationships one can find using web models, their combined expectations were unfounded and led to an architecture fraught with reliability, security, and other design flaws.

History, as Kling points out, is relevant for systems analysts. The electric power industry is not an exception. The interactions between government and electric power industry reflect a similar technological movement -- electrification -- from the early to mid 20th century when regulators wanted to maintain just enough control to retain legitimacy while simultaneously riding the wave of positive public opinion (Hirsh 1999). However, industry and deregulator technological determinists have discovered:

6 Conclusion

6.1 Kling's position validated despite critics

Complex ecologies such as the electric power industry have interwoven relationships and multilayered dependencies that are cultivated between historical technology and emerging ICT to empower ongoing public and private social agendas. In contradiction to assertions by technological determinists and naïve policy-makers, this research substantiates Kling's position that technological regimes are inherently social constructs. The implication is that the social aspects of urban infrastructure often determine the structural configurations of resources regardless of the engineering perspectives.

6.1.1 Importance of dependency and complementariness

Rich dependencies and complementariness between urban and ICT infrastructure indicate Kling's insights were correct. Information systems analysts working for vertically integrated monopolistic utilities incrementally added ICT as part of GDCC operations. Schedules were just one case where ICT was used to evolve an artifact in order to complement the utility's inter-organizational agenda. Over the history of electric power, when GDCC knowledge workers fully appreciated newly augmented ICT processes they would naturally evolve away from old Standard Operating Procedures (SOPs).

Process innovation naturally led to ICT dependence in both SOP and technological configuration. Evolution arising from process innovation can be more thoroughly understood when architectural aspects of the complex system are drawn out in order to highlight critical "weight bearing" relationships, which are usually tightly coupled with social mechanisms.

Knowledge workers who have their work processes complemented by technology would seem to be vulnerable to replacement by ICT. However, as seen here with intertie schedules moving from a paper to spreadsheet and then to the Internet, each evolutionary step created new social dependencies that complemented some group of stakeholders.

The new ICT complementariness weakens the old regime (the vertically integrated utility) and strengthens a new one (the "Marketers") for a period. However, inherent in the architectural design are social factors not easily transferred to new structures, especially if they are not expressly sought out. The new marketers had an incredible attrition rate (undocumented estimates go as high as 90%). Also, as noted earlier, supposed utility collusion was replaced by actual "Marketer" collusion (Kamp 2002).

The failure of deregulation was not news to the informants at GDCC who knew the system's dependencies and complementariness. They said: "in 2000 or 2001 everything will fall apart and they'll have to reregulate". In other words, the deregulator's architectural design will collapse when the full weight of the system is placed on the load-bearing beams. They could make such assertions because they knew the industry's shared social network and ontology carried the real system load and not ICT; contrary to what "Marketers" and deregulators supposed.

The added architectural correlations of the web model allow system analysts to view these rich dependencies and complementariness as social relationships across the panorama of mission-critical urban infrastructure using domain knowledge obtained from the ecology's "master builders". Architectural analysis provides a clear mechanism for associating resources and social processes. The spider metaphor serves to demonstrate how seemingly simple architectures that work incredibly well are attained through complex knowledge management and evolution based on ecological constraints. These constructs are meant to augment, and not replace, web models.

6.2 ICT: friend or foe?

A software feature to one user can be a bug to a developer, or an open door to the computer hacker. Such is the nature of ICT. Electric power utilities came to rely on ICT to extend their control over resources. As greater levels of control were attained and maintained over the decades, utilities pushed for greater control. However, when entire society became enamored with the capabilities of ICT, regulators used ICT as the tool to dismember the vertical utility giants: a double-edged sword.

ICT is, therefore, a fickle change agent or malleable construction material demonstrating conflicted behavior dependent on the ecology and social network supporting it. Deregulation leveraging ICT did not result in less expensive energy or more reliable service. When policies, based on rational models of technological determinism, are directed at environments whose operations are already positively enhanced by ICT through social constructs the effects can backfire, i.e. reducing process quality and weakening ICT effectiveness. These dependencies and the loss of these former complementarities result in organizational and group instability, which exacerbates problems during evolution.

ICT architectures used in the future must exceed those technologies of the vertically integrated utility era in order for personnel to reestablish the social coherence that facilitated the old GDCC operations. Current deregulation processes ongoing in Washington, DC, and at state capitals must facilitate new architecturally modified control over the electric power ecology (e.g. the grid and GDCC) in order for industry personnel to have a shared positive reaction to their new "space". Without such universal perceptions, policy makers are unlikely to gather support for their efforts.

Unfortunately, this cycle feeds into the dependencies that already exist. As the "reregulation" foretold by the GDCC informants begins, policy makers are once again likely to depend on ICT as the sculpting tool of choice to mold the next iteration of the evolving electric power industry. But will they again fail? Will positive belief systems in ICT again shape the minds and attitudes of the stakeholders so strongly that "What" and "How" committees will influence all decisions such that the design revolves around perceived levels of ICT-augmented effectiveness? They are likely to unless policy makers take a position that urban infrastructure forms technological regimes, which are inherently social constructs regardless of engineering perspectives.

Kling (1992) pointed out that the "computing infrastructure" depends on technological resources, including electricity. He asserted that infrastructure would often be regarded as inferior or would be ignored without developing an appropriate "Natural Systems model" using a web analysis. People with key control over knowledge and information use innovations to drive their agendas within social, organizational, and institutional constructs. Only by tying together complicated empirical studies can researchers tease apart intricate interactions and socially constructed schemas that facilitate the objectives of some and forestall the success of others.

Acknowledgements

I would like to take this opportunity to acknowledge that Rob Kling was instrumental in my early academic career. Without Rob’s initial encouragement and financial support of my work, this research would never have materialized. John L. King’s encouragement and ongoing support of this research (including a great deal of guidance to prepare this publication) have been indispensable. I appreciate the JoDI editors and anonymous reviewers for their patience, helpful recommendations, and insights. Early criticisms and comments by Ken MacGillivray, Nancy Parmalee, Madhu Reddy, and other friends were helpful building blocks. M. Faizan provided diligent technical assistance. I would also like to thank the many electric power industry informants and organizations who generously gave of their time and knowledge, but for privacy reasons cannot be personally identified.

References

Anon "Public Utility Regulatory Policies Act of 1978 (PURPA)". Energy Dictionary, EnergyVortex.com http://www.energyvortex.com/energydictionary/public_utility_regulatory_policies_act_of_1978_(purpa).html

Anon (1973) "OPEC Oil Embargo Leaves World Economy Threatened". DailyPast.com, October 21, 1973 http://www.dailypast.com/business/oil-embargo.shtml

BBC (2000) "GM goat spins web based future". BBC Online News, 21 August, 2000 http://news.bbc.co.uk/1/hi/sci/tech/889951.stm

Boland, R., Glymph, J., Zahner, B., King, J. L. and Lyytinen, K. (2003) "Computing on the Scaffolds: The Coming Transformation of Architecture and Construction with Digital Technologies". Proceedings of the International Conference on Information Systems (ICIS), Panel/Debate: M2.5, Seattle, WA, December 14-17 (Atlanta, GA: Association for Information Systems)

Brennan, T. J., Palmer, K. L., Kopp, R. J., Krupnick, A. J., Stagliano, V. and Burtraw, D. (1996) A Shock to the System: Restructuring America's Electricity Industry (Washington, DC: Resources for the Future)

Brooks, F. P. (1995) The Mythical Man-Month: Essays on Software Engineering (Reading, MA: Addison-Wesley)

Cowan, R. S. (1992) "How the refrigerator got its hum". The Social Shaping of Technology, edited by D. MacKenzie and J. Wajcman (Milton Keynes: Open University Press), pp. 202-218

Davenport, T. H. and Stoddard, D. B. (1994) "Reengineering: Business Change of Mythic Proportions". MIS Quarterly, 18(2): 121-127

David, P. A. (1985) "Clio and the Economics of QWERTY". American Economic Review, 75(2): 332-337

David, P. A. and Rothwell, G. (1996) "Standardization, diversity and learning: Strategies for the coevolution of technology and industrial capacity". International Journal of Industrial Organization, 14(2): 181-201

DeMarco, C. L., Sarlashkar, J. V. and Alvarado, F. (1996) "The potential for malicious control in a competitive power systems environment". Proceedings of the IEEE Conference on Control Applications, 15 September, pp. 462-467

Ellis, J. (2003) "Debunking Some Myths About the Great Northeast Blackout". Energy Pulse, September 18 http://www.energypulse.net/centers/article/article_display.cfm?a_id=475

Emerson, R. M., Fretz, R. I. and Shaw, L. L. (1995) Writing Ethnographic Fieldnotes (Chicago: The University of Chicago Press)

George, J. F. and King, J. L. (1991) "Examining the Computing and Centralization Debate". Communications of the ACM, 34(7): 62-72

Giedion, S. (1967) Space, Time, and Architecture: the Growth of a New Tradition, 5th edition (Cambridge, MA: Harvard University Press)

Hammer, M. and Champy, J. (1993) Reengineering the Corporation: a Manifesto for Business (New York, NY: Harper Business)

Hirsh, R. F. (1989) Technology and Transformation in the American Electric Utility Industry (Cambridge: Cambridge University Press)

Hirsh, R. F. (1999) Power Loss: the Origins of Deregulation and Restructuring in the American Electric Utility System (Cambridge, MA: MIT Press)

Hogan, W. W. (1993). "Markets in real electric networks require reactive prices". Energy Journal, 14(3): 171-200

Hogan, W. W. (1994) "Efficient direct access: Comments on the California Blue Book proposals". Electricity Journal, 7(7): 30-41

Hogan, W. W. (1995) "Electricity transmission and emerging competition". Public Utilities Fortnightly, 133(13): 32-36

Hughes, T. P. (1983) Networks of Power: Electrification in Western Society, 1880-1930 (Baltimore: Johns Hopkins University Press)

Hughes, T. P. (1992) "Edison and electric light". The Social Shaping of Technology, edited by D. MacKenzie and J. Wajcman (Milton Keynes: Open University Press), pp. 39-53

Hughes, T. P., Pinch, T. J. and Bijker, W. (eds) (1987) The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology (Cambridge, MA: MIT Press)

Iacono, S. and Kling, R. (1996) "Computerization Movements and Tales of Technological Utopianism". In Computerization and Controversy: Value Conflicts and Social Choices, edited by R. Kling (San Diego: Academic Press), pp. 85-105

Kamp, J. (2002) "Reliant Energy Says It Paid Perot Systems For '98 Meeting". Dow Jones Newswires, June

King, J. L., Gurbaxani, V., Kraemer, K. L., McFarlan, F. W., Raman, K.S. and Yap, C.S. (1994) "Institutional factors in information technology innovation". Information Systems Research, 5(2): 139-169

Kling, R. (1978a) "Automated Information Systems as Social Resources in Policy Making". Communications of the ACM, (12): 666-674

Kling, R. (1978b) "Automated Welfare Client-Tracking and Service Integration: the Political Economy of Computing". Communications of the ACM, 21(6): 484-493

Kling, R. (1980) "Social Analyses of Computing: Theoretical Orientations in Recent Empirical Research". Computing Surveys, 12(1): 61-110

Kling, R. (1987) "Defining the Boundaries of Computing Across Complex Organizations". In Critical Issues in Information Systems Research, edited by R. J. Boland and R. A. Hirschheim (Chichester: Wiley)

Kling, R. (1992) "Behind the terminal: The critical role of computing infrastructure in effective information systems' development and use". In Challenges and Strategies for Research in Systems Development, edited by W. W. Cotterman and J. A. Senn (New York, NY: Wiley), pp. 153-201

Kling, R. and Iacono, S. (1984) "Computing as an Occasion for Social Control". Journal of Social Issues, 40(3): 77-96

Kling, R. and Iacono, S. (1989) "The Institutional Character of Computerized Information Systems". Office: Technology and People, 5(1): 7-28

Kling, R. and Jewett, T. (1994) "The Social Design of Worklife with Computers and Networks: An Open Natural Systems Perspective". In Advances in Computers, Vol. 39, edited by M. C. Yovits (Orlando, FL: Academic Press), pp. 239-293

Kling, R. and Scacchi, W. (1982) "The Web of Computing: Computer Technology and Social Organization". In Advances in Computers, Vol. 21, edited by M. C. Yovits (New York, NY: Academic Press), pp. 3-90

Kraemer, K. L. and Kling, R. (1985) "The Political Character of Computerization in Service Organizations: Citizen Interests or Bureaucratic Control". Computers and the Social Sciences, 1(2): 77-89

MacKenzie, D. and Wajcman, J. (eds) (1992) The Social Shaping of Technology (Milton Keynes: Open University Press)

Magdalena, H. (2001) "Web of intrigue". Nature News Service, 29 March

Meyer, J. R. (1999) "Re-Regulating the Airlines: An Idea that Won't Fly". The National Academies, October 8

Navarro, P. (1996) "Electric utilities: the argument for radical deregulation". Harvard Business Review, 74: 112

Nobel, D. F. (1977) America by Design: Science, Technology and the Rise of Corporate Capitalism (Oxford: Oxford University Press)

Nobel, D. F. (1997) The Religion of Technology: the Divinity of Man and the Spirit of Invention (New York, NY: Alfred A. Knopf, Inc.)

OASIS "How" Working Group (1996) Transmission Services Information Network (TSIN) Requirements Document, Version 1.0 (4/13/96) ftp://ftp.nerc.com/pub/sys/all_updl/oasis/tsinreq1.doc

OASIS NOPR (1996) Final Order 889 Open Access Same Time Information Systems, Federal Energy Regulatory Commission, Notice of Proposed Rulemaking, April 24 ftp://ftp.nerc.com/pub/sys/all_updl/oasis/order889.zip; Final Order 888 Open Access and Stranded Costs, (WP5.1) April 24 ftp://ftp.nerc.com/pub/sys/all_updl/nopr/order888.zip

OASIS "What" Working Group (1995) "Suggested Content for an Electronic Information Network to Assist in Performing Transmission Service Transactions". Electronic Information Network "What" Working Group comments to Federal Energy Regulatory Commission, October 16 ftp://ftp.nerc.com/pub/sys/all_updl/oasis/einfina2.pdf

Oren, S. S. (1997). "Economic inefficiency of passive transmission rights in congested electricity systems with competitive generation." Energy Journal 18(1): 63-83.

Osterweil, L. J. (1997) "Software processes are software too, revisited: an invited talk on the most influential paper of ICSE 9". Proceedings of the 19th International Conference on Software Engineering, Boston, MA, (ACM Press), pp. 540 - 548

Nye, D. E. (1998) Consuming Power: a Social History of American Energies (Cambridge, MA: MIT Press)

Parsons, T. (1951) The Social System (New York, NY: Free Press)

Perry, T. S. (2004) "Dennis Shelden: Turning Dreams into Reality". IEEE Spectrum, February, 28-41 http://www.spectrum.ieee.org/WEBONLY/publicfeature/feb04/0204shjob.html

Rittel, H. and Webber, M. (1973) "Dilemmas in a general theory of planning". Policy Sciences, 4: 155-169

R.J. Karg Associates (1998) Deregulation of the Natural Gas Industry: What Has Happened As a Result of FERC Order 636? R.J. Karg Associates http://www.karg.com/dereg%20of%20nat%20gas.htm

Scacchi, W. (2004) Web of Computing, email to R. Klashner

Simmel, G. (1964) Conflict And The Web Of Group Affiliations (Free Press)

Spradley, J. P. (1989) The Ethnographic Interview (New York, NY: Holt, Rinehart, and Winston)

Tenenbaum, D. (2001) "Caught in the web". The Why? Files, University of Wisconsin, April 19 http://whyfiles.org/shorties/077spidersilk/

Unterman, N. A. "Electricity Speed". Ask A Scientist, Physics Archive http://www.newton.dep.anl.gov/askasci/phy99/phy99488.htm

Yates, J. (1989) Control Through Communication: the Rise of System in American Management (Baltimore, MD: Johns Hopkins University Press)

Yates, J. and Orlikowski, W. J. (1992) "Genres of Organizational Communication: A Structurational Approach to Studying Communication and Media". Academy of Management Review, 17(2): 299-326

Yates, J., Orlikowski, W. J. and Okamura, K. (1995) "Constituting genre repertoires: Deliberate and emergent patterns of electronic media use". Academy of Management Best Papers, Proceedings of the 55th Annual Meeting, Vancouver, Canada, August, pp. 353-357

Links

AutoCAD http://usa.autodesk.com/adsk/servlet/index?id=2704278&siteID=123112

Electric Power Research Institute (EPRI) http://www.epri.com/about/default.asp

Energy Policy Act of 1992 https://energy.navy.mil/publications/law_us/92epact/hr776toc.htm

Author details

Robb Klashner is interested in using the architectural metaphor as a communication tool. He uses software architecture styles to study multi-stakeholder software integration projects associated with large, complex, socio-technical systems such as the electric power grid that depend heavily on the flow of information to maintain productivity levels.