practices. create interventions through the design of artefacts. But how exactly can design research support the creation of change? In this paper we present the approach of Socio-Informatics which we have developed over the past two decades, and reflect on the diferent ways in which this research and its results contribute to changing social ∗Corresponding author.
Many forms of design research can be understood as action research (see e.g. [
Action research seems to have gained a poor reputation, but design-oriented research remains one of the few domains in science and research that is explicitly considered action research or at least action-oriented. Yet creating or supporting change in whatever domain is not a straightforward process. Therefore, the question remains: how does design and design research create social change?
In this paper we try to answer this question, by examining a specific form of design-oriented research
we have been developing for several decades, which we call socio-informatics (SI) [
Socio-Informatics (SI) emerged in critical interaction with existing academic traditions, seeking to overcome narrow, management-oriented or techno-centric conceptions of computing. It is an approach that explicitly acknowledges that design research is always interventionist, but argues that these interventions must be grounded in deep, situated understandings of social practices. By foregrounding this orientation, SI challenges deterministic or purely formal approaches to design that ignore social contexts or treat them as secondary concerns.
The paper is structured in the following way. We begin by outlining the research program of socioinformatics, including its epistemological foundations and its position on the spectrum between natural and social science, positivist and interpretivist research. We then explore the diferent ways in which socio-informatics aims at creating change, including brief examples from our own research history. Lastly, we will discuss our approach in relation to other forms of action and design research and outline future directions for research. (V. Wulf)
CEUR Workshop
ISSN1613-0073
Socio-Informatics is an approach focused on both the design of technology and the study of its use. It is
rooted in the practice-based approach of human-computer interaction (HCI) and computer-supported
cooperative work (CSCW). This practice-based approach, in turn, is rooted in the “turn to practice” of
the social sciences [
Developed as a critical response to the formal, technically oriented traditions of German computer science and the management-focused approach of business informatics or information systems 1, SI emphasizes the importance of ethnographically informed, participatory design. This commitment is not only methodological but also deeply normative: SI argues that researchers have a responsibility to support and improve social practices in ways defined together with those afected.
The central goal of Socio-Informatics is therefore not simply to analyze existing practices but to contribute to their development through carefully designed socio-technical interventions. It rejects views of technology as neutral or universally applicable, insisting instead on the situated, contingent, and negotiated nature of change. Design outcomes must be collaboratively developed and evaluated in the context of their actual use, acknowledging the “double character” of technology as both shaped by and shaping social practices.
Moreover, Socio-Informatics adopts a reflective, critical stance toward its own interventions. It aims to make the world “a better place”, not in a naive or unilateral way, but through participatory processes that empower local actors to define what “better” means for them. This requires researchers to remain aware of their own positionality, to navigate power asymmetries, and to invest in long-term, trustful engagements. Such an orientation aligns SI with traditions of participatory design and critical systems thinking, while demanding ongoing reflexivity about whose interests are served by any intervention.
This commitment to supporting practice improvement shapes the methodological approach of SI, which we describe next.
It usually begins with an initial pre-study or context study, to gain an understanding of exisiting practices and the challenges members face. This inquiry is typically participatory and qualitative, employing interviews and (participatory) observations. The resulting data form the foundation for the next step.
The pre-study is followed by participatory design activities. First, the collected data are analysed to define a case and identify a challenge in the practices or an opportunity to support or enhance them. This is followed by the exploratory task of asking: “what could work?” to address the challenge or opportunity of the case, recognizing that design requirements and solutions cannot be directly derived from this analysis. The study’s results are necessarily partial: they can describe a challenge, but not dictate how to addressed or solved it. Deciding how to go forward is thus always a creative and explorative process. After these two decisions are made, an artefact is designed. This entire process is (ideally) participatory, involving members of the practice context in the decision, to ensure that the case is worth addressing and that the manner in which it is addressed aligns with their wishes, needs, and ideas. Participation continues through design decisions made during the actual development and design of the artefact.
Socio-Informatics also insists on a critical, reflexive stance toward its own design practices. Researchers must examine how institutional settings, funding sources, and power asymmetries shape design outcomes. It is not enough to claim participatory design; there must be systematic analysis of how such participation is structured, whose voices are included or excluded, and what values are negotiated in the design process.
Lastly, the artefact is introduced into the practice context for members to work and appropriate. This
is understood as a creative process: Use does not just follow design. Members need to give the artefact
meaning and integrate it into their practices, which requires changing those practices. Crucially, this
process reveals shortcomings of the artefact , which then needs to be addressed and adapted as the
practice itself evolves in response to its introduction. Studying this process of appropriation employs
the same methodological approach as the pre-study, relying on interviews, observation, and interpretive
analysis. Introducing a technical artefact into the practice thereby also changes the initial practice: the
challenge to be addressed becomes another one as a result. In line with Rittel & Webber [
These are only the most central elements of the socio-informatic approach, and they do not necessarily
occur in that order, nor are they necessarily as distinct from each other as described here. Depending on
the context and situation where research takes place, members might be more or less involved in each
stage. They might already have specific and elaborate ideas for design interventions that are voiced in
the pre-study. Studies do not need to complete all elements to produce valuable knowledge, nor do
they need to stop there: knowledge can also be generated by comparing several design case studies
(e.g. [
As a design science that produces technological artefacts as well as knowledge about social contexts and change, socio-informatics sits between natural science and social science. It builds on the physical nature of computing technology, rooted in deterministic natural science, to create specific artefacts that function according to natural laws. Of equal if not more importance, however, is its goal to understand social practices and the roles digital applications play in them. This is its social science dimension, and while fields such as experimental psychology or economics might blur the border by following positivist traditions of the natural sciences, socio-informatics is firmly on the interpretivist side.
The knowledge created in socio-informatics about social practices, and the role of digital applications is always tied to the specific contexts in which it emerges. But the fact that it does not uncover universal laws does not mean this knowledge is inapplicable elsewhere. Generalization to other contexts is both possible and desired in the SI program, but it follows diferent ‘rules’ than the experiment- and statisticsdriven approach typical of psychology, which claims to produce universally applicable knowledge.
There are several ways in which knowledge generated in this program is applicable elsewhere; we briefly list three.
1) Following the ethnomethodologist approach outlined, e.g., by Crabtree et al. [
2) Socio-Informatic knowledge furthermore reveals possibilities, in a way similar to Harold Blumer’s
idea of sensitizing concepts [
3) The knowledge generated through the socio-informatics program is embodied in the designed artefacts and the design decisions behind them. The artefacts themselves carry therefore knowledge, or they are knowledge, without fully determining how they are to be used or what efects they will have. Their application wherever physical conditions allow is therefore a form of generalization, even though they do not carry absolute knowledge about their use 2.
It is important that we do not resolve to this form of research and its more limited or humble truth
claims only because we like to, but because we see no other option. Positivist research of course makes
broader and more general claims to knowledge and its applicability or generality, but such a stance has
been criticized by many and from many diferent angles, including authors such as Haraway [
Nevertheless, and to conclude the main take away from this section, even though socio-Informatic knowledge is generated in specific and strictly limited contexts the size of a social practice, it remains applicable and generalizable in specific ways to other contexts. Following from this, we will next explore how socio-informatics research can support, create, efect social change.
Our research program potentially supports change largely through three means: 1) Knowledge Creation, 2) Artefacts and 3) Participation in Design. We will look at each one in turn.
Socio-Informatics produces several forms of knowledge, that are applicable by others in diferent ways and for diferent ends.
Firstly, it generates knowledge about how practices are structured. This can inform or inspire design
activities and/or investigations into other practices. A well-known example in CSCW (though not our
own) is the sensitizing concept of “articulation work” ([
Secondly, socio-informatics creates knowledge about the specific problems encountered in the practice under investigation. This knowledge can serve as the foundation for design activities to address similar problems in the same or other contexts. It therefore informs design work or alerts designers to design opportunities.
Thirdly, it generates knowledge about the use of technology and the kind of efects it creates before an intervention. This helps to inform research and design work.
Fourthly, it creates knowledge about the role of technology after an intervention, revealing what worked or did not work in achieving a specific change in a practice. Such insights highlight possibilities which can inspire and inform other design activities.
Fifthly, by reflecting on the research and design process itself in meta-research, socio-informatics
potentially informs or inspires the research practices of others. SI aims to contribute to the broader
scientific discourse by developing and refining methodological concepts. Examples include Grounded
Design, Design Case Studies, Integrated Organization and Technological Development, and PraxLabs
[
Artefacts themselves create change - that is indeed the goal of their design - even if there can be no guarantee that the intended change will occur. They are a directed intervention into practices, designed with a specific change in mind. When an artefact is applied and appropriated in a specific context it produces efects, even though the exact nature of those changes cannot be fully anticipated because they are contingent on local conditions. Nevertheless, appropriation by members of a practice context inevitably changes that practice in some way.
Socio-Informatics views artefacts not as static products but as part of ongoing processes of “infrastructuring.” This concept highlights that design continues into use, as users appropriate, adapt, and reconfigure artefacts over time. Infrastructuring emphasizes that users are co-designers, shaping the artefact and their own practices in a mutually constitutive way [22].
An example from our own work illustrates this. In [23] we reflect on a project in which we designed a tool to support maintenance practices at an industrial manufacturing plant. The tool was designed to facilitate the reporting and prioritization of maintenance issues, which would help organize the maintainers work but also allow to adapt production planning to maintenance issues and machine or tool failures. The study followed the process described above, but when it was rolled out, management of the plant only gave access to specific people in the organisation’s hierarchy, which was not planned or agreed on prior to roll-out. It therefore created unintended and unanticipated efects in the organisation’s hierarchy. Nevertheless, it did address some of the challenges it was designed to do, and the company rolled it out to other departments. (See [23] for a more elaborate description of the design case study).
Furthermore, the artefact itself also embodies knowledge beyond its application and appropriation. It demonstrates to others designerly and technological opportunities as well as the design decisions that led to the final design. This, in turn, can inform other design and development work.
Lastly, the participatory nature of our design approach creates an additional opportunity to support change. Participation is not simply a means to gather requirements, but a transformative practice in itself. Through participation, stakeholders develop new skills, build mutual understanding, and renegotiate roles and responsibilities. SI research emphasizes that this process is often messy, conflictual, and contingent [24], but precisely in this complexity lies its potential for meaningful change. Participation in design generates learning efects for participants. While it helps researchers better understand practices and the design requirements of the participating members of the addressed context, it also enables participating members to learn about technological possibilities during the design process. Participants can also develop a clearer understanding of their own requirements, needs and demands for an application addressing the agreed-upon challenge. Finally, they may gain new skills related to technology design and introduction processes.
An illustrative example comes again from our own work. In [25] we describe how, together with an industrial SME in the region of our university, we designed a tool intended to serve as a checklist in the quality control process of large machine pieces, welded at an external plant. Members of the quality management department, whose challenges the tool actually addressed, were deeply involved in the design process, to the extent that one employee took on some of the coding work. When the moment for roll out in practice had arrived, the IT department of the company blocked the application due to security concerns3. Nevertheless, through the design process the participating members learned about their own requirements which enabled and encouraged them to select suitable commercially available options, which they had been afraid of before, and enabled them to create a new role for the participating employee involved in coding, whose new role it became to create similar projects himself in the company. (See [25] for more information on this example from our research history.) 3The department had so far deliberately been excluded by the participating members exactly for the fear that they would hinder the project
With this, we highlight three potential ways in which socio-informatics can create change. We do not claim there is any guarantee that projects will work exactly as described here. A further limitation lies in the wish to expand or fully realize the efects described above, which is to some extent outside the scope of the actual research and design work. The following chapter describes the challenges of expanding our work and reports on several experiences we have had in trying to overcome these challenges.
While above we sketched potential means for creating change that the socio-informatics approach
ofers, this does not mean that these changes actually occur or occur beyond the immediate research
context. To achieve this, further activities might be necessary [
The efects of the knowledge we created are limited by their reach, by the number of people who actually have access to this knowledge, even if it is theoretically applicable in other contexts. This is of course not unique to socio-informatics but a challenge for all scientific endeavors. Distributing scientific knowledge, making it accessible especially to practitioners is a task beyond the usual activities of research of course, which at most include the publication of scientific articles, written in jargon and often behind paywalls. It requires journalistic work and science communication, or other forms of knowledge sharing such as presentations or workshops. Ensuring that “practitioners” outside of one’s immediate circle of colleagues find and have access to knowledge (and see value in it) does not follow immediately from research and other academic work “as usual” but requires specific activities.
Similarly, there are limits to the realm in which artefacts can be efective. The artefacts designed as part of the research process of socio-informatics are rolled out firstly in the context in which they were developed. They might also be included in other research projects, but by themselves they will not be found by other practitioners, and, depending on the level of development, might not be immediately applicable. Bringing them to other contexts where they could be of use is also not always part of research projects, but requires “transfer work”, for which perhaps transfer ofices of universities are responsible. Common avenues to achieve broader reach is open-source publishing of the code and design, on publicly accessible repositories, or founding start-up companies and achieving a wider reach with economic activities. There are other avenues of course, but none of them follow immediately from the research and design work which we have described above. Researchers who want to achieve or support wider change with their work will have to deviate from “research as usual” and engage in activities that are not their daily bread and butter (or they need to partner with people whose bread and butter it is).
In our group we have made several observations as to how the results of our work can spread and difuse throughout the region of our university, beyond the immediate contexts in which the work took place [25, 26]. A crucial element in what we observed was the notion of “spillover”, that efects of a design intervention spilled from one context to another. This did not happen by itself, neither was it the result of only directed efort, but it took place through serendipity and certain kinds of activities. It is also not best thought of as ‘transfer’ but as co-construction of knowledge, in which designed artefacts take on a supportive, mediating role. An example illustrates the observed process.
After developing the checklist tool designed to support quality control processes, outlined above in section 3.3, a person employed by the administration of the region to support local companies often worked in close collaboration with us. She heard of the project and independently saw a potential benefit of the application for the care sector of the region. She invited members of our research team and from there a design collaboration developed with various actors in the care sector. Together with a care institution a research and design project was carried out. A pre-study revealed that care staf faces significant workload collecting and monitoring health data about patients but also about the temperatures of fridges where medicine is stored, etc. This data was usually collected with pen and paper. An application was envisioned that would consist of a checklist similar to the quality management application, but in a smart watch format. Later it was discovered that much of the data could be connected automatically via interfaces in the various devices already used to measure data, such as oximeters and fridges. Instead of the smart watch application, a dashboard and IoT system was developed that allowed the automatic collection and display of the necessary data. The example illustrates potential means and conditions for broadening the area in which socio-informatic can become efective. Such spillover requires actors that see the potential for spillover between diferent contexts and that have the personal connections to make it happen. It also requires a similarity of practices that an application addresses, e.g. in our case regular activities of monitoring data (of welded machine parts in one example, health data in the other). The artefact is not simply carried over from one sector to another but provides an opportunity for collaborative knowledge creation appropriate for the context. The artefact itself carries knowledge, and illustrates opportunities and decisions, but this knowledge needs to be appropriated for the new context, which is an active and creative process. Supporting such spillover efects also requires the construction and maintenance of social networks, as spillover depends on interpersonal connections, which can be a significant amount of work beyond central research and design activities [25, 26].
We have also noted the success and necessity in some projects of extensive network-work and press work. In larger projects that include several disciplines and institutions we have created specific roles of “network managers” and public relations managers located in the specific project instead of the university as a whole. The positions are responsible for creating, maintaining and deepening relations with various actors in the region and ensuring that the results of the work as well as opportunities for collaboration are spread. These connections are crucial for spillover to occur, as such personal connections and awareness of opportunities are in some sense the ‘infrastructure’ along which spillovers occur. We have reflected on these activities also in [ 26].
With this article we give an overview of the socio-informatics research program and the diferent ways in which it realizes its commitment to action research and to creating change. It does so principally by engaging with practices as its primary research focus, but also as the focus of its design interventions. While we outline above how socio-informatics intends to support change, socio-informatics does not necessarily imply the direction of the change. Its commitment to practices implies a certain commitment to people and what they actually do, the sense and meaning they perceive in their practices as well as the challenges - not what other people think they ought to be doing. This carries with it certain implications for the kind of change socio-informatics supports. For example, it would be dificult to design for the full automation of jobs within the research program of socio-informatics, unless the workers in question wish for it, including the consequences this would have for their employment. SI is nevertheless adaptable to a broad variety of contexts, of which the varied output of our group over the past decades provides ample proof.
In our deliberations we also deal with the range or arena in which socio-informatics usually or efectively supports change. This arena is characterized by several limitations which are not unique to socio-informatics but apply to other disciplines and varieties of design research as well. The impact of knowledge is limited by the number of people who have access to this knowledge, the impact of design interventions is limited to the context in which they are carried out. We have nevertheless made experiences with deliberately broadening this arena for our work, and have published initial observations and reflections about our experiences. However, more research is required to develop a better understanding of how the arena in which change is supported under the socio-informatics program can be increased.
Ultimately, SI frames itself as an interventionist design science with a clear normative orientation: to support and improve social practices in collaboration with those who enact them. It rejects the search for universal, one-best-way solutions, arguing instead for context-sensitive, participatory, and reflective approaches that respect the complexity and contingency of social life.
ings of the 1st Conference on Designing Interactive Systems: Processes, Practices, Methods, & Techniques, Dis ’95, Association for Computing Machinery, Ann Arbor, Michigan, USA and New York, NY, USA, 1995, pp. 55–64. doi:10.1145/225434.225441. [22] V. Pipek, V. Wulf, Infrastructuring: Toward an Integrated Perspective on the Design and Use of Information Technology, Journal of the Association for Information Systems 10 (2009) 447–473. doi:10.17705/1jais.00195. [23] C. Kotthaus, N. Vitt, M. Krüger, V. Pipek, V. Wulf, Negotiating priorities on the shopfloor: A design case study of maintainers’ practices, Computer Supported Cooperative Work (CSCW) 32 (2023) 141–210. [24] M. Krüger, A. Weibert, D. de Castro Leal, D. Randall, V. Wulf, ”What is the Topic of the Group, Please?” On migration, Care and the Challenges of Participation in Design, Proceedings of the ACM on Human-Computer Interaction 5 (2021) 1–29. [25] M. Krüger, M. Gerbracht, N. Vitt, M. Kudic, M. Ahmadi, A. Boden, F. Ofergeld, M. Stein, C. Kotthaus, D. Unbehaun, et al., Travelling artefacts: Lessons learned from interventions in a regional innovation ecosystem, in: Proceedings of 20th European Conference on Computer-Supported Cooperative Work, European Society for Socially Embedded Technologies (EUSSET), 2022. [26] M. Kudic, M. Krüger, M. Gerbracht, M. Ahmadi, A. Boden, M. Stein, C. Kotthaus, N. Vitt, D. Unbehaun, P. Tolmie, et al., Knowledge spillover: Between serendipity and strategic planning-lessons for practice-oriented interventions into regional innovation systems, Proceedings of the ACM on Human-Computer Interaction 8 (2024) 1–26.