Innovation logics in the digital era: a systemic review of the emerging digital innovation regime

ABSTRACT

This essay reviews the ontological status of digital material in industrial operations and conditions it provides for innovation. It recognises the semiotic quality of digital material and the critical role of von Neumann architecture as to show that digital innovation advances through a three pronged, largely orthogonal process of embedding- a process of interlacing elements of one innovation domain to that of another.  Three types of embedding define digital innovation: operational embedding (code-computer), virtual embedding (real-world phenomena-code) and contextual embedding (performing code-social setting). Each constitutes a ‘leverage point’ for further expanding digital innovation and they interact dynamically while industrial innovation is defined by a static virtual and contextual embedding. The processes that underlie innovation in the two regimes follow differential logics. In industrial regime (1) discover-synthesise and (2) manufacture-distribute activities combine to form the leverage points for innovation. In digital regime 1) discover product or behaviour (ideation); 2) abstract to digital material (virtualise); 3) Implement and replicate digital material (variations in operational/contextual embedding); and 4) Deliver product and/or behaviour (perform digital material) activities form the leverage points for innovation enabling wider, open, and faster innovation with a distinct value logic.

KEYWORDS:

• Digital innovation
• industrial organisation
• scale
• scope
• digitalisation
• von Neumann architecture

Acknowledgments

I am grateful to Robin Gustafsson and Kimmo Karhu for lively discussions, where many of the ideas originated and were debated and shared. Special thanks to Janis Kallinikos for moral support and shared interest and sharp comments on the topic. I thank the guest editors of this special issue, Llewellyn D.W. Thomas, Youngjin Yoo, and Raghu Garud, for their insightful comments and suggestions for improvement. All mistakes, errors, and misunderstandings remain solely mine.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. There are several excellent historic analyses of how such models emerged and were perfected during a period from 1850 to 1970 (see Chandler, 1990; Hughes, 1989).

2. This argument is simplified in purpose somewhat. Industrial innovation in can be also nonlinear in that manufacturing capabilities may already be available from which new innovations emerge, for example, by repurposing some existing capabilities for COVID-19 equipment. Or modular platforms can be in place before ideas come into fruition, as exemplified by investments in traditional technology platforms for automobiles, or in knowledge platforms such as the ones in 3M (Karnøe & Garud, 2012).

3. How such digital technologies were organised in car industry at the height of this period is reported in King and Lyytinen (2004).

4. We call digital representation any string of 0 and 1 that has a shared agreed meaning in any given context. A digital object is a digital representation with clear definition and use purpose in a setting like a payroll record or a CAD/CAM model.

5. Just a change of one bit’s value can fundamentally change the meaning of the code.

6. Note that in Figure 1 the direction of the embedding does not match with the direction of the arrows, that is, up or down, but, instead it conveys whether it is the digital (grey) being embedded with physical (white), or vice versa.

7. Both examples have a different and evolving operational and virtual embedding. In the first phase, those cover agreements and digital representations of stocks, prices, money, trades, and so on. In the latter they involve representations of car geometry, hit strength or point, weight, speed, and so on.

8. Recently, this embedding has become a subject of increased virtualisation (called cloud computing), whereby the connection between the digital and hardware has been rendered dynamic and non-transparent by adding a separate digital layer that coordinates and controls how the digital material hits the hardware in a distributed way during a dynamic embedding of the operations.

9. These presentations follow strictly Ludwig von Wittgenstein’s famous last proposition in Tractactus: ‘Whereof one cannot speak in ‘0ʹ and ‘1ʹ, thereof one must be silent.’ (Wittgenstein, 1922). In this sense, virtual embedding has a fixity and comes with social inertia related to agreements. But the form and structure of representation can be changed relatively easily as well as what the agreement means (a common trick in repurposing data fields for other uses in many applications).

10. Recently, the device is rather any product because cars, refrigerators, and faucets may share computing capacity.

11. Consider rules and conventions around how we are expected to use mobile phones and related services in different settings in work and family life and at different times of the day.