|
Article Excerpt
Abstract
Information technology (IT) innovation can be defined as the creation and new organizational application of digital computer and communication technologies. The paper suggests that IT innovation theory needs to be expanded to analyze IT innovations in kind that exhibit atypical discontinuities in IT innovation behaviors by studying two questions. First, can a model of disruptive IT innovations be created to understand qualitative changes in IT development processes and their outcomes so that they can be related to architectural discontinuities in computing capability? Second, to what extent can the observed turmoil among systems development organizations that has been spawned by Internet computing be understood as a disruptive IT innovation?
To address the first question, a model of disruptive IT innovation is developed. The model defines a disruptive IT innovation as an architectural innovation originating in the information technology base that has subsequent pervasive and radical impacts on development processes and their outcomes. These base innovations establish necessary but not sufficient conditions for subsequent innovation behaviors. To address the second question, the impact of Internet computing on eight leading-edge systems development organizations in the United States and Finland is investigated. The study shows that the adoption of Internet computing in these firms has radically impacted their IT innovation both in development processes and services.
Keywords: Internet computing, innovation theory, disruptive IT innovation, IT innovation cores, system development, software management, IT applications
Introduction
In the mid-1990s, computing entered a state of flux that was triggered by the Internet. The flux was accompanied by the creation of tools, techniques, and practices called Internet computing. Internet computing as used here denotes a broad and evolving set of models of distributed computing and related solutions that rely on open, heterogeneous, ubiquitous network services, and associated protocols (March et al. 2000). Many scholars and practitioners have argued that Internet computing represents a new way of conceiving and developing information systems and constitutes a break from earlier approaches to organizational computing organized around mainframes, personal computers, or client-server models (Alter et al. 2001; Baskerville et al. 2001; Lyytinen et al. 1998). (3) Recent research provides ample anecdotal and speculative evidence that Internet computing has spawned a wave of innovations in system development and services (Alter et al. 2001; Baskerville et al. 2001 ; Carstensen and Vogelsang 2001; Glass 2001; Isakowitz et al. 1998; Lyytinen et al. 1998; Lyytinen and Yoo 2002; Turoff and Hiltz 1998), and has changed computing into a global phenomenon that utilizes global infrastructures (Braa et al. 2000; Castells 2001).
These studies, although illuminating, do not formulate a theoretical model of how changes in information systems (IS) development and services depend on antecedent changes in technological capability, and what types of changes are necessary to establish a disruptive information technology (IT) innovation. In short, they identify neither necessary nor sufficient conditions for radical and widespread change in IS development and its outcomes. They also fail to provide empirical validation of the extent to which organizations deploying Internet technologies have experienced significant and lasting changes.
One reason why studies of Internet computing have not relied on prior IS innovation research is because questions about kinds of changes in system development and its outcomes--like the kinds identified in Bijker's (1987) discussion of technological frames--have not entered into the mainstream of IS innovation research (e.g., Fichman 1992; Prescott and Conger 1995; Swanson 1994). In contrast, past IS innovation research has primarily described and explained changes in the volume of technological and organizational change associated with IT innovation. What little work has been done in understanding differences in kind has been narrow in scope and limited to descriptions of changes as either fashions (Newell et al. 2000) or imitations (Loh and Venkatraman 1990).
To address the observed gaps in Internet computing studies and IT innovation research, this paper engages in a cycle of theory generation and validation that seeks to advance our understanding of the dynamics of IT innovation and specific IT innovation types. We use the generated theory to explore the extent to which Internet computing as a new paradigm has led to disruptive changes (introduced here as being both radical and pervasive) in the services (4) that are built, and in the ways those services are built. In brief, the research questions posed are:
1. How can we model disruptive IT innovations to help us identify and analyze qualitative changes in IS development and its outcomes due to architectural discontinuities in computing capability?
2. To what extent can the observed turmoil among systems development organizations spawned by Internet computing be understood as disruptive IT innovations?
These questions were addressed in two ways. First, we engaged in theory generation and developed a model of disruptive IT innovation by drawing on theories of industrial innovation (Christensen 1992a, 1992b; Teece 1986), Swanson's (1994) theory of IS innovation, and Zaltman et al.'s (1977) concept of radical innovations. Second, we use this model to formulate two conjectures that instantiate the disruptive IT innovation model in the context of Internet computing and help clarify the disruptive nature of Internet computing. These conjectures were derived by a careful analysis of the extant literature on Internet computing in light of the disruptive IT innovation model. We conducted a multisite case study by drawing upon the disruptive IT innovation model to validate the extent to which development processes and services had changed radically and pervasively in eight system development organizations that adopted Internet computing. The study lends support to widely published claims that Internet computing can result in radical and pervasive changes in development processes, along with their outcomes. It also shows the value of the disruptive IT innovation model in analyzing complex IT innovations like Internet computing as innovations in type. The paper concludes with a discussion of the implications for IS innovation research and practice, and an epilogue that discusses these implications in light of the end of the dot-com boom.
A Model of Disruptive IT Innovation
Innovation Defined
A widely accepted definition of organizational innovation is that it involves adoption of an idea, material artifact, or behavior that is new to the organization adopting it (Daft 1978; Rogers and Shoemaker 1971). Not all ideas, material artifacts, or behaviors adopted are innovations; an innovation must be accompanied with newness or novelty as a key distinguishing feature. Because defining newness is often difficult, it is argued that it is the perception of newness that counts, rather than whether the idea or artifact is new to the world. Following Zaltman et al. (1977), innovations are always defined in terms of a specific individual, organization, or community.
The innovation literature is voluminous and diverse, but two important streams can be distinguished: (1) theories of industrial innovation and (2) the diffusion of innovation literature. The first stream deals with types of innovations: artifacts and ideas that are new to a community or industry. Industrial innovation research has examined structural characteristics of an industry, product (architecture), market, or organization, and has asked why and how ideas, behaviors, or artifacts that are novel for the industry or community emerge, and what their impacts are (Abernathy and Clark 1985; Christensen 1992a, 1992b). The diffusion of innovation stream has focused on innovations in scope: the adoption of artifacts and ideas that are new to the would-be adopter once they have been discovered. This research stream has focused on the innovation-demand side, and primarily applied diffusion of innovation theory to discern patterns of diffusion (Rogers 1990). Researchers' primary interests have been in discerning factors and processes that can explain adoption outcomes and the infusion of innovations over user populations (Fichman 1992; Kwon and Zmud 1987; Lyytinen and Damsgaard 2001; Prescott and Conger 1995; Rogers 1990).
Disruptive Innovation
Theories of industrial innovation seek to understand what drives innovation in the context of specific industries and products, technology, and market environments (Christensen 1992a, 1992b; Chistensen and Bower 1996; Dosi 1982; Foster 1986; Teece 1986; Utterback 1996). Often these theories relate structural features in products, services, and industries or interactive processes (Slappendale 1996) to specific innovation outcomes. Normally, these models also seek to explain necessary transformations in the industrial organization and/or markets as conditions for, or outcomes of, such innovations.
During the last decade, several seminal studies advanced disruptive architectural innovations as the main engines of industrial transformation (Christensen and Bower 1996; Henderson and Clark 1990; Teece 1986; Utterback 1996). (5) Disruptive innovations are often the outcome of unleashing new product architectures that deviate radically from existing product lines by incorporating novel and unprecedented architectural principles like changing telecommunication service from circuit switching to packet switching, or transforming imaging from an analog to a digital process (Christensen and Bower 1996; Henderson and Clark 1990; Teece 1986; Utterback 1996). Architectural innovations stand out as creative acts of adapting and applying latent technologies or potential to previously unarticulated user needs (Abernathy and Clark 1985). They radically deviate from an established trajectory of performance improvement, or redefine what performance means in a given industry (Chistensen and Bower 1996). They are radical (Zaltman et al. 1977) in that they significantly depart from existing alternatives and are shaped by novel, cognitive frames that need to be deployed to make sense of the innovation (Bijker 1987). Consequently, disruptive innovations are truly transformative (Abernathy and Clark 1985). To become widely adopted, disruptive architectural innovations demand provisioning of complementary assets in the form of additional innovations that make the original innovation useful over its diffusion trajectory (Abernathy and Clark 1985; Teece 1986). By doing so, disruptive innovations destroy existing competencies (Schumpeter 1934) and break down existing rules of competition.
Information Technology Innovation
An information technology innovation can be defined as an innovation in digital and communications technologies and their applications (Swanson 1994). IT innovations result from exponential improvements in computing speed (e.g., Moore's law) and data storage functions that have over time led to radically enhanced functionality in processing, storage, transfer, and display of information. (6) Hence, most IT innovations are inherently linked to a continuously improving computing capability-cost ratio. In their simplest form, IT innovations involve only a technological component--changes in hardware and software that are new to an industry or adopters--but they are often augmented with complementary organizational innovations including new forms of cognition, meaning, work process, business process, or organizational structure. Specific IT innovations involve these elements in different proportions, and consequently affect the content, scope, and organization of IT innovation processes within an organization or the industry (Swanson 1994). As a result, IT innovations normally penetrate organizations through integrating a complex network of interrelated innovations covering discovery of new computing capability, establishment of new development capabilities, and formulation of new services. In addition, other forms of innovation normally help deliver and manage those innovations including virtual teams, total quality principles, and the like.
IT Innovation Types
We distinguish between three types of IT innovations and their interactions in a model hereafter referred to as the three-set model of IT innovation (see Figure 1). The first innovation set (what Swanson calls Type I) involves changes in system development processes, such as new development tools or programming teams. The second innovation set consists of outcomes of development processes (i.e., services). This set involves uses of IT to support the administrative core of the organization (what Swanson calls Type II) such as accounting. It also deals with those innovations in which the uses of IT affect either business functions or core business processes of the organization (what Swanson calls Type III). For many of these IT innovations, changes in computing capability are often a necessary (but not sufficient) precursor. (7) Hence, the third set, noted here as IT base innovations, change available computing capability. They establish a necessary antecedent and an element of many forms of other IT innovations (Friedman and Cornford 1989; Galliers and Somogyi 1987; Sauer 1999; Tsichritzis 1997). (8) IT base innovations include, among others, new software and hardware architectures and services, and new telecommunication capability. The three sets of innovations are mutually dependent in that an innovation in one type may spawn innovations in others. This codependency is created because IT innovation processes are simultaneously driven by both push and pull forces (Zmud 1982, 1984) resulting in what Swanson calls order effects (shown in Figure 1 by double headed arrows).
[FIGURE 1 OMITTED]
Subcategories of IT Innovation Types
The three innovation sets in Figure 1 can be divided into subgroups, depending on the nature and content of the innovation in each innovation set (Swanson 1994). These subgroups are illustrated in more detail in Table 1 and are used to identify key areas of IT innovation in our analysis of Internet computing.
We divide base IT innovations into three subcategories by using the impact of change as a basis for classification. The three subcategories are (1) changes in the base technology as defined by functionality, speed, reliability, architectural principle, or other features (Base 1); (2) changes in IS development as defined by modeling and design principles or by coordination of related processes (Base 2); and (3) changes in services as defined by changes in general service features (Base 3). Subcategories in system development innovations affect either technical (System Development 1) or administrative development activities (System Development 2). The subcategories in service innovations are created by organizational boundaries and task types within the adopting firm (Swanson 1994). Based on Swanson, we distinguish between four types of service innovations: (1) services that support the administrative core (Service 1); (2) services that support functional processes (Service 2); (3) services that expand and support customer interfacing processes (Service 3); and (4)services that support interorganizational processes and operations (Service 4).
A Model of Disruptive IT Innovation
Some IT innovations may become disruptive in that they call for radical shifts in IT focus, capability, and solutions that demand significant investments in complementary assets (Friedman and Cornford 1989). Yet, the IS research community currently lacks adequate models to identify such innovations in kind and thereby help distinguish between disruptive and incremental IT innovations. To deal with this omission in the literature, we propose a model of disruptive IT innovation that is intended to examine the dynamics in kinds of IT innovations. Such a model would recognize those changes within computing capability that would constitute a significant transformation in the architectural principles governing services and development processes. The innovations in the IT base would thus be necessary (but insufficient) to create subsequent changes in development processes and services.
We define a disruptive IT innovation as a necessary but not sufficient architectural innovation originating in the IT base that radically and pervasively impacts systems development processes and services. To avoid technological determinism we use the terms necessary and not sufficient in the definition to clarify the conditions under which specific changes in the technology base can become disruptive. (9) In line with process theories (Markus and Robey 1988) of technological impact, architectural changes in the IT base have to be present in order for a disruptive IT innovation to happen. Hence, an architectural change per se does not create disruptive change in other IT innovations, but requires other (sufficient) conditions to be met as defined by the concept of complementary assets (Abernathy and Clark 1985; Teece 1986). In a disruptive IT innovation, the computing capability undergoes a qualitative change in kind through a set of novel IT base innovations and their application outcomes that radically alter the subsequent trajectory of IT innovation. A disruptive IT innovation knits together a set of interrelated technological and organizational advances involving qualitative IT base innovations and related IT innovations in development processes and services that further the exploitation of the base technology innovation by providing co-specialized assets (Teece 1986).
Significant qualitative changes in computing capability are outcomes of a unique constellation of non-linear growth in underlying computing power. This new qualitative capability eventually overwhelms the current computing paradigm and its associated design space created by available technological frames (Bijker 1987). For example, the http protocol and the concept of a browser within Internet computing (Berners-Lee and Fischetti 1999) drew upon exponentially growing capability in computing speed, data storage, and transfer. At the same time, a qualitative change occurred in the way we think about computing capability. This capability was expanded successively through new models and principles of distributed computing that refined and improved the architecture, solution patterns, and services of Internet computing--thereby accumulating co-specialized assets associated with the new paradigm. Hence, unique moments of discovery in computing capability can introduce a possibility for waves of other IT innovations that push the new computing capability into the mainstream and over time integrate co-specialized assets (innovations) into the dominant computing platform like design methods, performance standards, or alternative ways of organizing the development activity (Christensen and Bowers 1996). In short, as a result of a novel base-computing discovery, IT innovators face a revolution and undergo a paradigm shift (Kuhn 1996).
Our definition identifies two conditions that together define a disruptive IT innovation; specifically, changes in IT innovations that follow a base IT innovation must be both pervasive and radical. The dimension of pervasiveness follows from the concept of disruptive architectural innovations, which states that these innovations create qualitative changes--a kind of systemic shock throughout computing--after which the technology and the existing rules of the game are changed. Accordingly, a disruptive IT innovation is pervasive in that it simultaneously and necessarily spans new services, and new types of development processes, thereby covering all of the IT innovation subcategories listed in Table 1.
The ability to meet the requirement for pervasiveness for sets of IT innovations does not itself result in a disruptive IT innovation. For example, a minor upgrade in one base technology area--for example, a change in the color capability of standard Wintel software that would count as a service capability innovation--is subsequently felt throughout all IT innovation types. However, the overall impact of such an innovation would be insignificant, because this type of a change in computing capability would not substantially affect the configuration of other capabilities in the technology base, and therefore would not qualify as an architectural innovation. A disruptive IT innovation embraces a radical shift in how adopting organizations must view, operate, and utilize IT so that their subsequent use of computing capability will be different after adopting the IT base innovation. Such a change is radical when...
|
