Sectoral Systems of Innovation: A Socio-Technical Approach
Summary 14: From Sectoral Systems of Innovation to Socio-Technical Systems: Insights About Dynamics and Change from Sociology and Institutional Theory
In the last decade, *sectoral systems of innovation* have emerged as a new approach in innovation studies. This article makes four contributions to the approach by addressing some open issues. The first contribution is to explicitly incorporate the user side in the analysis. Hence, the unit of analysis is widened from sectoral systems of innovation to *socio-technical systems*. The second contribution is to suggest an analytical distinction between systems, actors involved in them, and the institutions which guide actors’ perceptions and activities. Thirdly, the article opens up the black box of institutions, making them an integral part of the analysis. Institutions should not just be used to explain inertia and stability. They can also be used to conceptualize the dynamic interplay between actors and structures. The fourth contribution is to address issues of change from one system to another. The article provides a coherent conceptual multi-level perspective, using insights from sociology, institutional theory, and innovation studies. The perspective is particularly useful to analyze long-term dynamics, shifts from one socio-technical system to another, and the co-evolution of technology and society.
The first contribution is to include both the supply side (innovations) and the demand side (user environment) in the definition of systems. The sectoral systems of innovation approach has a strong focus on the development of knowledge and pays less attention to the diffusion and use of technology, impacts, and societal transformations. Sometimes, the user side is taken for granted or narrowed down to a *selection environment*. Hence, I propose a widening from sectoral systems of innovation to socio-technical systems. This means that the fulfillment of societal functions becomes central (e.g. transport, communication, materials supply, housing). This indicates that the focus is not just on innovations, but also on use and functionality. The need to pay more attention to innovation and users has, in fact, already been identified by a range of scholars in innovation studies and evolutionary economics. So the paper aims to link up with an identified *open issue* in the field.
Second, with regard to the kinds of elements, I propose to make an analytic distinction between: systems (resources, material aspects), actors involved in maintaining and changing the system, and the rules and institutions which guide actors’ perceptions and activities. I suggest such analytical distinctions are useful because some current literature groups together too many heterogeneous elements. For instance, Malerba (2002, pp. 250–251), wrote that “the basic elements of a sectoral system are: (a) products; (b) agents: firms and non-firm organizations (such as universities, financial institutions, central government, local authorities), as well as organizations at lower (R&D departments) or higher levels of aggregation (e.g. firms, consortia); individuals; (c) knowledge and learning processes: the knowledge base of innovative and production activities differ across sectors and greatly affect the innovative activities, the organization and the behavior of firms and other agents within a sector; (d) basic technologies, inputs, demands, and the related links and complementarities: links and complementarities at the technology, input and demand levels may be both static and dynamic. They include interdependencies among vertically or horizontally related sectors, the convergence of previously separated products or the emergence of new demand from existing demand. Interdependencies and complementarities define the real boundaries of a sectoral system. They may be at the input, technology or demand level and may concern innovation, production and sale. The (d) mechanisms of interaction both within firms and outside firms: agents are examined as involved in market and non-market interactions; (e) processes of competition and selection; (f) institutions, such as standards, regulations, labor markets, and so on”. Although these elements are all important, it is somewhat unclear how they are linked. This article aims to make progress on this issue.
The third contribution links up with another *open issue*, which has also been identified in the field, i.e. to pay more attention to institutions. Sometimes institutions are a *left-over category* in analyses. It also happens that institutions are wrongly equated with (non-market) organizations. See, for instance, Reddy et al. (1991, p. 299), “examples of non-market institutions include: professional societies, trade associations, governmental agencies, independent research and coordination organizations, and public-service organizations”. Anyway, there is a recognized need to better conceptualize the role of institutions in innovation. In particular, it is useful to explain how institutions play a role in dynamic developments, rather than explaining inertia and stability.
A fourth contribution of the article is to address the change from one system to another. This is relevant, because the main focus in the systems of innovation approach has been on the functioning of systems (e.g. a static or comparative analysis of the innovative performance of countries). If there was attention for dynamics, it was usually focused on the emergence of new systems or industries (e.g. Rosenkopf and Tushman, 1994; Van de Ven, 1993). Not much attention has been paid to the change from one system to another. In a recent discussion of sectoral systems of innovation, Malerba (2002, p. 259), noted that one of the key questions that need to be explored in-depth is: “how do new sectoral systems emerge, and what is the link with the previous sectoral system?” This question is taken up in the article. This means the focus of the article is not on (economic) performance, but on dynamics and change.
These four contributions are made by describing a coherent conceptual perspective. This means the paper is mainly conceptual and theoretical, using insights from different literature. Insights from sociology of technology and institutional theory are combined with innovation studies, science and technology studies, cultural studies, and domestication studies.
Summary 22: A World Run on Algorithms?
Mostly without our awareness, algorithms now run much of our lives. In the future, they will likely be even more ubiquitous in ever more aspects of our personal and work life. They will increasingly shape our choices and delineate our options. Even more pervasive but less transparent, algorithms will be used to mine and exploit data about us that is collected and stored daily in ever increasing quantities by business and government. Algorithms are also taking many of our jobs.
Mysterious Algorithms
In their simplest and oldest form, algorithms are sets of rules for processing data to produce outcomes. They are “provable, well-defined (and generally well known) solutions to a specific problem set” that can be carried out using the same set of instructions each time, although the number of instructions required depends on the data input.
But although algorithms are not new, they were put on steroids decades ago by computers—all software is built on algorithms—and they are used in every digital device in existence. Now big data, combined with nearly free and unlimited computing processing power and storage, is adding a massive boost to the power and impact of algorithms.
In short, anything software-driven is part of the digital ecosystem that runs our lives and will be running even more aspects of our lives in the future.
Algorithms to Run Even More of Our Lives
While for many years algorithms have influenced decisions in our personal lives such as guided “matches” on dating web sites, in the future our work lives may also be increasingly run by algorithms.
Big Data Powered by Algorithms
These vastly improved algorithms have been a key factor powering the age of “big data,” a concept with no official definition but which refers to huge volumes of data that are created and captured but cannot be processed by a single computer—instead requiring the resources of multiple machines or even the cloud to store, manage, and parse.
While the amount of information individuals create themselves—writing documents, taking pictures, downloading music, making searches, posting tweets and “likes,” etc.—is huge, it remains far less than the amount of information being created about them in the digital universe, often without their knowledge or consent.
“If you’re keeping track, algorithms already have control of your money market funds, your stocks, and your retirement accounts,” Christopher Steiner notes, adding “they’ll soon decide who you talk to on phone calls; they will control the music that reaches your radio; they will decide your chances of getting lifesaving organ transplants; and for millions of people, algorithms will make perhaps the largest decision in their life: choosing a spouse.”
Helping Watson Win Jeopardy! and the NSA Mine Data
Robots run on algorithms and they are becoming increasingly connected to the web, feeding data into “big data” streams, and using the computational power of the cloud to enable their artificial intelligence as they become increasingly ubiquitous, moving out of the confines of the factory to work directly with human beings.
Algorithms are being unleashed on massive data flows from potentially billions of sensors on natural and human-created objects as well as on humans themselves, who are carrying data generators such as smartphones. But the big data produced will be useless without algorithms to parse it and take actions based on the analysis.
Risks of an Algorithm-Run World
While algorithms, big data, and the Internet of Everything (IoE) are making important contributions in science, health care, efficient use of resources, and smart cities, there are concerns about a “dark side” of algorithm-driven big data:
- While businesses may gather massive amounts of personal data that can be used to sell products to you, government could potentially use that data to spy on and prosecute you;
- Algorithms unleashed on big data can lead to misuse at the expense of innocent individuals;
- Algorithm-driven decision-making can take humans out of the loop with disastrous results;
- Algorithms can also be used as the basis of autonomous cyber weapons capable of creating physical destruction;
- Algorithms are eliminating many jobs, and this is expected to be a long-term, structural trend;
- Algorithms can be vulnerable to malicious hacking with potentially catastrophic effects.
Benefits of the Algorithm-Powered Internet of Everything (IoE)
This will result in many cost efficiencies. A few examples:
- A reduction in energy wasted on inefficient shipping and warehousing;
- Supply chain efficiencies. A reduction in the cost of the “Cold Chain” in particular would result in huge energy savings;
- A reduction in the cost of repairs for many larger appliances and machines;
- Smart (very smart) homes that would know to manage energy;
- New heights of factory automation;
- Far safer food supply allowing nearly instantaneous sourcing of tainted goods as well as the ability to detect and report spoilage automatically;
- Automated surveillance that will be much easier and cheaper;
- Tracking virtually everything in the health care system, including patients.
Failsafe and Safeguard Systems
A more vigilant approach to the construction and maintenance of critical infrastructure will be required in the future. Key systems relying heavily on software and its underlying algorithms are becoming progressively more capable and complex. To keep pace with the potential of failure, we need to increase investments in both up-front testing and subsequent monitoring of these systems.
Given the potential for sizeable damage, key systems should always require a separate infrastructure to instrument and monitor them.
The financial system especially requires failsafe and safeguard systems.
Conclusion
- Without algorithms there would be no computers, no Internet, no modern communication, transportation, or energy and water systems. In short, our modern world would grind to a halt.
- Algorithms are essential and mostly reliable, but the world is highly vulnerable should they be misused, hacked, sabotaged, or simply fail.
- There may also be need for regulation of the use of algorithms, especially in financial systems and the exponentially-increasing collection and analysis of personal data by a growing number of businesses.
- We should not lose sight of the huge benefits of algorithms, not only in running our basic systems but in pushing the frontiers of science, especially in understanding our natural environment, improving our health, making our cities more energy and resource efficient, increasing access to knowledge and education, widening our personal networks, and enhancing productivity.