In megaprojects, as in many other interorganisational settings, several interests are at stake simultaneously, causing tensions that need to be resolved. In this paper, the authors take a closer look at compromises as one possible path for resolution. This study aims to make a theoretical contribution to the literature on business networks regarding the phenomenon of compromises and their effects in a compromising process.
Based on the Industrial Network Approach, the authors investigate a case where a big science facility is created. The case focuses in particular on the compromises that were important for the creation.
The authors show that the creation process is characterised by a need from involved actors to continuously balance different goals and expectations through strategies that often include a need to compromise. A compromise is, hence, best understood as a process phenomenon where there is a pre-compromising stage, a compromising stage and a post-compromising stage where the performative process of creating the facility continues with the compromise as a mutual commitment that directs and restricts the continued performance of the involved actors.
The paper contributes to the Industrial Network Approach by pinpointing the concept of compromise, as well as by demonstrating and discussing some of the benefits of studying tensions as a source of business network dynamics. In addition, the authors contribute to different interorganisational theories that use the concept of compromise by focusing on compromise as process.
1. Introduction
In a multifaceted network setting where various actors from different spheres of society interact, multiple interests are at stake simultaneously. Due to divergent interests and goals of the involved actors, tensions between opposing and interrelated demands may arise and need responses. One way of dealing with tensions is to find a compromise between them, which can be regarded as a pragmatic way of beneficial advancement. The most general view of a compromise is that by giving a little, a settlement can be reached. For instance, a compromise can be defined as “an acceptance that you will reduce your demands or change your opinion to reach an agreement with someone” (Cambridge Dictionary). Inspired by Jones and O’Flynn (2012), we view a compromise as a two- or multisided deal made between two or more engaged actors who all give ground through some sort of compromising process. We also argue that the phenomenon of compromise has its origin in a tension (Mitzinneck and Besharov, 2019) and that a compromise is something that the parties commit and adhere to.
In this paper, we take a closer look at the compromise phenomenon and focus on the effects of compromising. To do so, the Industrial Network Approach (Håkansson and Snehota, 1995; Håkansson et al., 2009; Håkansson and Snehota, 2017) is adopted. Compromises have been thoroughly discussed in political science (see e.g. Gutmann and Thompson, 2012; Van Parijs, 2012), but in business studies less so, even though compromises have been investigated to some degree in consumer marketing (Boldt and Arora, 2017), in studies using gaming theory (He and Wu, 2020) and in ethical business reasoning (Graham, 1998; Tangpong, Li and Hung, 2016). Surprisingly little has been written on compromises from a business network or relationship approach (cf. Håkansson, 1982; Håkansson and Snehota, 1995), despite the fact that a compromise can be viewed as a focal point for coordination between involved parties (He and Wu, 2020).
According to van Bommel (2014, p. 1181), “a pluralistic environment typically makes a compromise a more likely and preferable solution”. In this paper, we study the effects of compromises made in a pluralistic environment involving many different actors. The chosen setting is large projects, often referred to as megaprojects (see e.g. Davies et al., 2009; Flyvbjerg, 2014), which “represent an extreme setting with inherent uncertainty and complexity” (Romestant, 2020, p. 535). Research in megaproject settings also shows that tensions are commonplace in these (Brady and Maylor, 2010). A specific type of megaproject is big science facilities [1]. The type of project we focus on in this study, i.e. the creation of a big science facility, consists of actors from different sectors in society. The various actors control different resources and perform contrasting activities (cf. Håkansson and Snehota, 1995) in the process of jointly creating the facility.
Earlier research on interaction related to big science facilities has shown that tensions may stem from various sources. One tension arises from the goal to create value for the society financing it, while simultaneously conducting the unpredictable and non-linear process that characterises basic science; “scientific research is regularly promoted not just for the potential scientific discoveries, but also for the perceived societal benefits” (Åberg, 2013, p. 27). Other tensions stem from the goals and interests of different funders. When these are nation-states (referred to as member states) involved, they all need to show value for their own taxpayers (Åberg and Bengtson, 2015).
Working with the construction of a big science facility means dealing with tensions and, as a consequence thereof, a continuous process of making compromises between divergent goals of the different member countries, but also of various actors with an interest in the values that can come from the facility. The tensions found are created by the plurality of actors involved and the divergent and often conflicting goals they have for the facility. There is a duality between, on the one hand, providing value to society by financing it and, on the other hand, being involved in the unpredictable and non-linear process that characterises basic science. We argue that this type of tension is persistent over time and, as such, unsolvable (Clegg et al., 2002; Smith and Lewis, 2011). In the investigated case, the people involved, therefore, need to find ways of meeting competing demands or considering divergent ideas simultaneously to identify various possibilities in the shape of compromises. Smith and Lewis (2011) state that compromising between poles is important as they are mutually reinforcing to enable long-term success (despite short-term competition for resources). At the same time, compromises “entail considerable work, as they have to be created, solidified and justified” (Oldenhof et al., 2014, p. 55).
Based on a process perspective (Langley, 2009; Pettigrew, 2012), we view a compromise as a pragmatic strategy and a critical event that helps initiate and direct (but sometimes restrict) a specific process. We argue that optimal outcomes are created rather than found, and that making compromises between the interests of involved actors can be viewed as a pragmatic way of moving forward for beneficial advancement in a certain process. We perceive a compromise as a focal point for coordination that alleviates conflicts of interest. This is in accordance with He and Wu (2020, p. 217), who claim that “While a compromise option may serve as an effective coordination device in the short run, how its presence may affect people’s decisions in the long run remains unknown”. To determine the long-term effects of compromising, process studies of the kind represented in this paper are necessary. By studying how involved actors navigate compromises caused by tensions, we aim to make a theoretical contribution to the literature on business networks regarding the phenomenon of compromises and their effects in a compromising process. The paper contributes to the Industrial Network Approach by pinpointing the concept of compromise and showing the effects of compromises on business network dynamics. We answer the question:
How can compromises be understood from a processual network approach?
The paper is structured as follows: First, in Section 2, we provide a theoretical base for the paper by discussing what compromises are and how they arise through tensions, grounding our reasoning in the Industrial Network Approach. Thereafter, we describe our method for the study and show how we analysed the data in Section 3. Third, we present our narrative in Section 4, illustrating the tensions and how they are managed in the creation of the big science facility, followed in Section 5 by an analysis of the material. Finally, we draw conclusions from the study and provide theoretical and managerial implications, as well as some thoughts on avenues for future research in Section 6.
2. Theoretical underpinnings
2.1 What is a compromise?
As a theoretical concept in the Industrial Network Approach, compromises have previously only been dealt with scantily. We, therefore, begin by identifying how compromises have been defined and theorised in the limited studies available and in other research areas. Some researchers dealing with compromises take a pragmatic view of the phenomenon, for example stating that it is “a sense of instrumental necessity that subordinates [the parties] to compromise” (Tangpong, Li and Hung, 2016, p. 83), or that the compromise serves as a focal point for coordination as it “alleviates conflicts of interest” (He and Wu, 2020, p. 216). Consumer marketing research, however, has extensively studied what has been termed the compromise effect. Boldt and Arora (2017, p. 436) explain that, for a consumer, “an alternative will gain shares when it becomes the intermediate option in a choice set”, since it is easier to justify (Simonson, 1989) and leads to less aversion (Simonson and Tversky, 1992). Gutmann and Thompson (2012, pp. 16–17) differentiate between compromising and uncompromising mind-sets, where “The compromising mind-set displays what we call principled prudence (adapting one’s principles) and mutual respect (valuing one’s opponents)”.
Others interested in the phenomenon are more sceptical of compromises and argue that it is necessary to stay aware of their effects. For example, Lehtimäki and Virtanen (2020) state that there is an inherent instability in compromises, since the different principles tend to remain distinct despite the fact that the compromise functioned as a means to solve the tension situationally. They argue that compromises “include the possibility of collapse and turn toward critical relations” (Ibid., p. 157). In the same vein, and based on a rather narrow perspective on the phenomenon, Graham (1998) questions the efficiency of compromises, arguing instead for wider solutions that integrate the interests of all parties, as they lead to “long-term, more productive and happier relations” (Graham, 1998). Other researchers, such as Gutmann and Thompson (2012, p. 14), claim that integrative agreements, or “win-win situations”, differ from “classic” compromises, because both sides gain and neither side has to make any sacrifices. In general, and in sharp contrast to the view in political science where compromises are perceived as something valuable, there seems to be a negative connotation often associated with compromising in business studies. Tidström (2014), for example, describes how compromise in some theoretical models (e.g. Thomas and Kilmann, 1978) is associated with situations where parties surrender something for a decision that is mutually acceptable (cf. Rahim, 1983). Pruitt (1983) even associates compromise with lazy or half-hearted problem-solving.
Many researchers focusing on compromises seem to study different interpretations of agreements or documents. For instance, van Bommel (2014, p. 1157) studied integrated reporting and found that “a compromise requires a common interest, avoidance of clarification and maintenance of ambiguity”. In another study, Trevisan and Mouritsen (2023) differentiate between compromising and compromise. They analyse valuation practices and claim that “[c]ompromising is a process of arriving at trade-offs between values, while compromise is a situation where different values simultaneously exist” (Trevisan and Mouritsen, 2023, p. 1). The authors see compromise and compromising as two modalities. In compromises, “actors create ambiguous associations […] that suspend critiques and requests for clarifications”. In compromising, on the other hand, actors “mobilize relations of equivalence that enable them to trade off, negotiate and mediate different principles” (ibid.) In this study, we view compromise as process, thus, close to compromising as described above, but expand on the time frame to see what actors make of the compromise, i.e. its effects, as part of a more holistic compromising process.
2.2 The industrial network approach
This study is grounded in the Industrial Network Approach (Håkansson and Snehota, 1995; Håkansson and Snehota, 2017), where one of the basic tenets is that industrial networks consist of long-term, relatively stable relationships (see e.g. Håkansson, 1982; Håkansson et al., 2009; Håkansson and Snehota, 1995). Whereas early studies using the approach focused on understanding interaction between business actors from an embeddedness perspective, more recent studies have added interaction and relationships between business and non-business actors into the equation (Bengtson et al., 2009; Leite, 2022; Leite and Bengtson, 2018). The relationships, be it between business actors or between business and non-business actors, evolve over time through interaction when firms collaborate, coordinate activities and adapt their respective resources (Wadell and Bengtson, 2023).
Even though it is well understood and extensively researched within the Industrial Network Approach (Håkansson and Snehota, 2017) that business actors find constructive solutions to various challenges and are able to create value through their cooperation with others, little is written on the matter of compromise in business-to-business settings. It is argued that conflict is inherent in all kinds of relationships, and this does not imply that the relationship is malfunctioning; it has even been stated that conflict can be necessary for a healthy relationship (Håkansson and Snehota, 1995, p. 9). However, solutions to coordination problems are “especially difficult to achieve in the presence of conflicts of interest” (He and Wu, 2020, p. 216; cf. Alimadadi et al., 2019). The predominant sentiment within the relationship, therefore, has to be one focusing on constructive solutions; “It is the concern with cooperation and value-creating which is what makes a relationship worthwhile for the parties” (Håkansson and Snehota, 1995, p. 9). Cooperation resulting in constructive solutions involves, by necessity, the art of compromising.
2.3 Divergent demands lead to tensions
One way of dealing with compromises and compromising in industrial networks is to see them as involved actors’ responses to tensions and, hence, as network dynamics (cf. Gadde and Håkansson, 2023; Alimadadi et al., 2019). According to Smith and Lewis (2011, p. 395), “The juxtaposition of coexisting opposites intensifies experiences of tension, challenging actors’ cognitive limits, demanding creative sensemaking, and seeking more fluid, reflexive, and sustainable management strategies”. In their research, Smith and Lewis (2011, p. 381) explore how organisations can simultaneously attend to competing demands, which is vital for sustainability. They argue that continuous efforts are required to meet multiple, divergent demands. Cunha and Putman (2019) argue that researchers need to examine how organisations respond to the tensions, rather than simply treating them as a problem or a tool. They explain that responses in this setting refer to actions and reactions that emerge as part of contextually embedded, ongoing organising processes when actors decide to move forward in the middle of contradictions (Cunha and Putman, 2019).
Extensive research on tensions, but also on duality and paradoxes (see e.g. Farjoun et al., 2018), has a view on interaction between actors that is aligned with the Industrial Network Approach, as all focus on interaction and resolution rather than hierarchical control. As explained by Cunha and Putman (2019, p. 100), “Responding, however, differs from controlling or resolving contradictions; rather, it centres on how tensions emerge, evolve, and transform in the midst of (inter-) organizational events”. Long-term sustainability, therefore, requires continuous efforts to meet and handle these multiple and divergent demands (Lewis, 2000), because there are underlying tensions that are based on elements that exist simultaneously, are inconsistent and that do persist over time (Lewis, 2000). Smith and Lewis (2011) explain that tensions surface between the differing and often conflicting demands that various stakeholders have. Researchers using a duality perspective argue that opposing forces are fundamentally interdependent and, hence, both contradictory and mutually enabling (Farjoun, 2010). In a similar vein, Åge and Eklinder-Frick (2017) show in a study on negotiation processes that the negotiator is continuously balancing opposing, and seemingly contrasting, forces. They argue that the core of successful negotiations is finding processual and relationally oriented negotiation processes.
2.4 Analysing compromising processes
The construction of a big science facility involves, by necessity, a large number of actors from different sectors in society. These business and non-business actors control different resources and perform contrasting activities needed in the joint creation process. Throughout the process, there will be tensions that surface due to the large plurality of actors involved and the divergent and often conflicting goals they have in relation to the facility. Therefore, there is a continuous need for compromising, which makes this an ideal setting to learn more about the phenomenon.
To learn more about compromising and its effects, we adopt a view on compromise in accordance with the suggestions of Jones and O’Flynn (2012), who state that a compromise has some essential features related to interaction, process and commitment. Accordingly, we argue that:
A compromise is made between two or more network actors;
the compromise has its origin in a tension, i.e. a conflict of interest, between the actors due to divergent preferences, interests, principles, beliefs or judgements;
the compromise is a two- or multisided deal, in which all parties are engaged and give ground, that;
reaching the compromise involves some sort of pre-compromising stage, often in the shape of a discussion or a negotiation between the compromisers (Åge and Eklinder-Frick, 2017; Eklinder-Frick and Åge, 2020), and, most importantly for this study, that; and
a compromise has a post-compromising stage, with effects that the parties may commit and adhere to as the compromise becomes solified (Oldenhof et al., 2014).
In this study, we will focus especially on the post-compromise stage, which is in accordance with He and Wu’s (2020) request for more studies on the long-term effects of compromising.
3. Method
This paper is based on an in-depth qualitative study of the process of creating a new big science facility, the European Spallation Source (ESS). ESS is a multinational big science facility under construction in the south of Sweden, in a process where several tensions that impact both its creation and continued operations have been faced. As part of data collected for a larger project, we study the process of creating ESS and focus on tensions that surface and compromises reached to handle these tensions. We investigate the process of creating ESS both physically and mentally, starting from the time when the decisions of building ESS were made, involving, e.g. decisions on financing, location, etc.
ESS is a multidisciplinary research facility under construction in the Swedish city of Lund by 13 European member countries, with Sweden and Denmark serving as host nations. This means that part of the facility, the data centre, will be located around 70 kilometres away, in the Copenhagen area in Denmark. The creation of ESS is taking place at present, and gives us a unique chance to investigate the unfolding of critical events (MacKay and Chia, 2013). A spallation source, such as ESS, consists of a high-powered accelerator that brings protons with high energies into a heavy metal target, such as mercury or tungsten. The metal “spalls off” free neutrons in response to the impact. Once constructed, ESS will constitute the world’s most powerful neutron source. When the user program starts in 2027, two to three thousand researchers from academia and industry are expected to use the facility for experiments each year. ESS as a case is interesting for several reasons:
it is a huge financial investment, meaning that it will be perceived as constraining and important for actors involved;
it involves many different actors with various wants and needs; and
it takes a lot of time and effort to complete, meaning that there will be several twists and turns to the events happening in the process.
3.1 Focus and study design
The focus of this paper emerged from one of our interviews, where the respondent – when asked about what it meant to have one part of ESS located in another country – stated that it was a bit of a compromise. This remark resulted in the aim of the paper, to make a theoretical contribution to the literature on business networks regarding the phenomenon of compromises and their effects in a compromising process.
We ground the phenomenon of compromises in the Industrial Network Approach (Håkansson and Snehota, 1995; Håkansson and Snehota, 2017). Given that tensions are explained to be sensitive to contextual circumstances in both time and place, it seemed a suitable choice to design the study as a case study in real time (Cunha and Putman, 2019). A case study design helps us to follow the interaction between and among the involved actors (Van de Ven and Poole, 1995) and to capture these circumstances in understanding organisational responses to tensions. We investigate how interaction between actors involved in the creation of ESS unfolds (Bizzi and Langley, 2012; Langley, 2009) in its contextual setting (Pettigrew, 2012). It has been stated that “the complexity of large projects requires pluralist approaches […] that recognize diversity in beliefs, values, and interests among stakeholders” (Boonstra, van Offenbeek and Vos, 2017, p. 19). We believe that the approach chosen will help us understand how compromises play out in the empirical setting of the study.
3.2 Data collection
The data collection for the ESS case began as early as spring 2014, when one of the authors participated in a one-day conference introducing the imminent start of ESS construction in autumn 2014. During this conference, a number of perspectives on ESS and its value for Sweden as a nation, for research and for industry were laid out. The data collection officially ended in 2024, but the authors are still following up on the developments taking place with and at ESS (by following the ESS website, reading news articles and documents being published and talking to people involved in the creation of ESS). We have based the data collection on a method of systematic combining (Dubois and Gadde, 2002, 2014), where we have alternated between data collection and theoretical grounding. The Industrial Network Approach formed a basis for the interviews from the start, but our findings from these interviews gave us further clues to interesting phenomena, such as compromising, to dig further into theoretically before returning for further interviews and other data collection. Data have been collected from interviews with actors involved in the creation of ESS in various ways, as well as information from conference attendance and secondary data sources (e.g. archival materials, press articles and annual reports). More than 60 interviews have been carried out, and extensive secondary data concerning the initiation and creation of ESS have been collected, including over 7,000 (Swedish and Danish) newspaper articles, and a large number of reports and other documents. In Table 1, the different data sources used in this paper are outlined.
Overview of data sources
| Type of data | Content | Amount of data |
|---|---|---|
| Conferences | Conferences focusing on ESS and industry | Seven one- and two-day events |
| Interviews | Interviews with industry, ESS and political actors | Approximately 20 interviews (of 60) used in this paper |
| Newspaper articles | Swedish and Danish daily press | Around 30 Swedish newspaper articles (out of 7,000) used in this paper |
| Reports | Studies and reports on ESS in Lund Memoires | 8 different documents explicitly used |
| ESS documentation | Progress reports Annual reports | Approx. 10 different documents used |
| Videos and podcasts | Presentations by ESS staff, presentations by Skanska |
The people interviewed fell into one of three categories:
central scientific actors from the ESS project perspective;
firms that are current or potential suppliers to ESS; and
regional and national (Swedish) political actors involved in the project.
The interviews focused on understanding the respondents’ engagement with ESS, including the reasons for their engagement, as well as the various twists and turns of events that occurred during the process of creation from their respective perspectives. Parties were added to the list of interviewees, and new interviews with people we had spoken to earlier were conducted, as we are following the ESS in the making. While all interviews carried out within the project were vital for understanding the context and the process, 20 interviews were especially important for the development of the case in this paper. These interviews are presented in Table 2.
Overview of interviews
| Respondent | Position | Date/duration of interview | Relevance for current case study |
|---|---|---|---|
| Carlile, Colin | Retired: Former director of ESS, ESS Scandinavia, currently Science Village Scandinavia | 29 September 2022 (2h10) 22 November 2022 (2h10) | Background on the creation of ESS, the Swedish negotiations, incl. the compromises |
| Carlsson, Patrik | Former ESS-S, co-director and Swedish ILO, Big Science Sweden | 5 September 2022 (40 min, morning) 5 September 2022 (40 min, afternoon) | Information about endeavours to place ESS in Sweden, relations between Swedish industry and big science facilities |
| Ekelöf, Tord | Professor Emeritus, FREIA Laboratory Uppsala University | 6 April 2022 (50 min) 10 May 2022 (1h30) 13 May 2022 (2h30, including site visit FREIA) | Background on big science facilities in Europe, Swedish industrial deliveries, in-kind vs in-cash aspects |
| Hall, Anna | (Former) Director big science Sweden and ILO ESS | 22 March 2021 (50 min) 26 March 2021 (1 h) | Swedish perspective on construction of ESS, industry contracts and in-kind |
| Holm Rod, Thomas | ESS (data centre), group leader of data analysis | 12 March 2021 (1h15) | Background on the ESS data centre and the effects of dual locations |
| Kindhult, Pia | Head of host country relations, ESS | 14 January 2020 (1h50) | Background on ESS and host country relations (Sweden & Denmark) |
| Larsson, Allan | Retired. Former finance minister, Sweden; former chief negotiator ESS, ESS-S | 22 June 2022 (1h30) | Swedish government inquiry on ESS, negotiations concerning ESS |
| Lefmann, Kim | University of Copenhagen, professor. Involved in early stages of ESS and the ESS data Centre | 16 April 2021 (1h45) | Background on the ESS data centre and the effects of dual locations |
| Leijonborg, Lars | Retired. Former minister for higher education and research, Sweden; former chief negotiator ESS | 15 September 2022 (1 h) | ESS negotiations to locate ESS to Sweden and to get other countries on-board |
| Melander, Fredrik | Former senor advisor, Danish ministry of higher education and science. Head of R&D relations, Science Village Scandinavia | 4 March 2022 (1h50) | Denmark’s negotiations to join ESS; negotiations between Sweden and Denmark |
| Nilsson, Christoffer | Head of the municipal civil service, Lund, SWE | 9 March 2022 (55 min) | The importance of ESS for the municipality of Lund, local and regional effects |
| Rennie, Adrian | Uppsala University, professor | 6 October 2020 (1h45) | Background on negotiations and construction of ESS (science perspective) |
| Taylor, Jonathan | ESS data centre (DMSC), head of DMSC | 11 March 2021 (1 h) | Background on the ESS data Centre and the effects of dual locations |
| Vieweg, Michael | (Former) CEO, Scanditronix AB | 29 November 2019 (1h30) 27 June 2022 (1h10) | Company perspective on delivering to big science facilities (and particularly ESS) |
3.3 Analysing the data
Just like the data collection, the data analysis has been carried out abductively (Dubois and Gadde, 2014). Using the Industrial Network Approach as a broad theoretical base, the empirical material was collected, combined and written up in an preliminary and very broad case narrative (Langley, 1999, 2009), from which themes could be found and further developed as we moved between previous research and empirical findings (Dubois and Gadde, 2002).
In the next step, we went back and coded the interviews based on the themes. We also turned to theory again to find tools for deepening and further analysing the findings in the material. It was at this stage that we discovered studies on tensions and duality and recognised their value in understanding the process we are studying. Once the events that were important for the focus in this particular paper – tensions and compromises – were found, we triangulated these by adding additional data sources around important facts. Hence, through the material, we became acquainted with the overall process studied in the paper, as well as found explanations and resemblance to it in theory (cf. Gioia et al., 2013). This analysis process helped us to narrow down the focus to the logic of a duality thinking to deal with tensions and a specific focus on understanding the act of compromising in this process.
The final step of coding and analysing, after we decided on the focus of compromise, was to zoom in on a few important compromises to reach enough depth in the circumstances surrounding these. The compromise surrounding the co-host arrangement and the compromise concerning the in-kind procedures were chosen, partly since they were brought up and described by many informants and partly due to their importance for the overall process. We decided to describe the process chronologically to show the processual nature of the creation of the big science facility.
4. Compromising in the creation of European spallation source
On 2 September 2014, the Swedish and Danish research ministers put the shovel to the ground on a Greenfield site in Lund in the south of Sweden (SvD, 2014 -09-02). This was the symbolic physical start of a huge research infrastructure project by the name of the ESS. Once finished, ESS will be the world’s most powerful neutron source (ESS website), and the biggest science infrastructure ever built in Scandinavia. ESS will work with neutrons emitted from a target to gather information on various materials, substances and objects. Behind the project initiative lies years of planning and negotiation between the 13 European partner states, who in May 2009 decided that the facility should be located in Sweden. As will be further described, there are quite a few actors from the science community, industry and society at large that would like to have their say in its design and progression.
4.1 The idea of a new big science facility
The idea of a Scandinavian big science facility had been simmering among researchers in Denmark, Southern Sweden and Norway for some time when the European idea behind ESS, which was started in the nineties, surfaced around the millennium. The idea of a new spallation source of the kind that ESS represents, however, was a German suggestion and the first large gathering to discuss the prospect was held in Bonn in 2002 with representatives from Germany, France and Great Britain, among other nations. Sweden had a delegation from Lund, including the university principal and several politicians from the municipality (Aobadia, 2002), taking part. Nothing was decided at the time, much due to the fact that Germany had other priorities and had decided not to support the project.
The Scandinavian researchers stayed with the idea, however, and started developing the design concept, while in parallel beginning to involve the political side. As described by Patrik Carlsson from the organisation Big Science Sweden:
There has to be support from the host country to pull something like this off. One has to understand what it means that a single country cannot pull it through but needs to build relationships with other countries. And that one can expect some problems, but also great advantages.
After a couple of years, the grassroots project grew, as the former Swedish Minister of Finance, Allan Larsson, was asked to investigate the possibilities for Sweden to host a European facility of this magnitude. Looking back, Larsson describes his interest in the project:
It is very fascinating to me to see this as an idea, that researchers can identify something that no one has done before. No one had thought this technology through, and how you put it together and understood how it works. And copes with pushing it to a design and use phase […] that to me is huge.
After completing the first investigation, Larsson was sold on the idea and worked with the project group of researchers to get acceptance for the ESS project internationally, a task that, after the victory of the conservative parties in the Swedish election in 2006, was handed over to Lars Leijonborg, the new minister of research and higher education. Based on the work already done by Larsson, Leijonborg made further efforts to convince Swedes that this was worth investing in. He had two main arguments, one being that:
A facility of this magnitude, to the largest part paid for by other countries, would have a positive effect on the growth of Sweden.” But also that: “A large research facility on Swedish ground would put research in Sweden on the world map and entail a profoundly needed and positive stimulus for Swedish research.
4.2 An agreement to construct the facility is reached
In 2009, an agreement was signed stating that ESS was to be built in Sweden and its data centre in Denmark, and, hence, to appoint Sweden and Denmark as the host countries for the facilities. This joint partnership did not come without both technical and economic difficulties, however, and some people even describe it as a (necessary) compromise. The funding of the endeavour, i.e. the member countries’ contributions, either in-cash (paying money directly to ESS) or in-kind (providing materials and equipment) to the project, was barely solved when the construction started in 2014.
For the practical operations of the facility, it was decided in 2012 that Swedish and Danish Governments would own the facility jointly (Sweden contributing with 35% of the funding in-cash, while Denmark contributing 12.5% essentially in-kind), and employ around 150 researchers and administrators. According to some researchers (Kaiserfeld and O’Dell, 2013), the location of the facility and its data centre in the Öresund region were conditioned on the opening of the Öresund Bridge. The construction of the Öresund Bridge started in 1995, and the almost 8-kilometre-long bridge that connects Malmö in Sweden with Copenhagen, Denmark, was opened on 1 July 2000 (Berg et al., 2000).
For the more specific arrangements between Sweden and Denmark as the two co-hosts, it was decided that Denmark would be in charge of the ESS data management and software centre (DMSC). The division into two separate locations has been described in terms of a “metaphor that the heart of ESS is located in Lund, but its brain is in Denmark” (Fredrik Melander, interview). To support ESS, Denmark wanted the data centre, because:
[…] the major development lies to a great extent in data management, and not just data but visualization and analysis and things like that […] so it is clearly an exciting technology development area which they would like to see Danish research and Danish development connected to (Fredrik Melander, interview).
In simple terms, every research facility has to provide computing support, and in the ESS case, this is arranged through the data centre located in Denmark. Professor Rennie, who was initially hosting the advisory panel for the centre, describes that the Danish government were negotiating quite hard early in the process in return for supporting the ESS project. Fredrik Melander, who was working for the Danish government at the time, took an active part in all of the negotiations. He says that Denmark wanted a lot of specifications clearly spelt out:
I wrote the whole part of the ESS statutes that has to do with it [the data centre] and we watched the development closely to make sure that the data centre would get the role, functionality, and size that we counted on. At that time, we had almost weekly host meetings with the ESS management in Sweden, so it is clear that it was very important for us and for Denmark. (Fredrik Melander, interview).
Not everybody was keen on the idea of dividing the experimental facility and the data centre up, and the Danish demand of specifying everything regarding the data centre in the ESS statues was also considered out of the norm:
The arguments were typical; partly that this type of function should be located in the main facility, and partly that these issues should not be specified in an international agreement, it should emerge from the operational side. […] Other countries’ interests are not outlined in the statutes in that way. (Fredrik Melander, interview).
In the end, however, the negotiation resulted in a co-host agreement between Denmark and Sweden, with the facility’s data centre infrastructure located in Denmark, which, when it is up and running at full capacity, will employ over 60 people. The plan was for the data centre to be located on site at the campus of Copenhagen University in Denmark.
4.3 The coordinating and dividing roles between centre and facility are formed
According to Rennie (interview, 2020): “This [the co-host arrangements] actually means that the Danes will always get their money back”. However, having a data centre remote from a facility of this type always causes some challenges: “There are some challenges that are technical, some that are almost political and economic” (Rennie, interview 2020). The technical concerns have to do with logistics and safety. The data from ESS is planned to be transported by the fibre optics installed across the Öresund Bridge. However, the capacity of the fibre optics is not infinite, especially if the ESS neighbour facility, the MAX IV laboratory, will use the same data infrastructure, the capacity might be too low. As explained by Professor Rennie, “Basically, the fastest way of shipping data is still to take hard discs in lorries” [2]. There are also some safety concerns about having data analysis remotely, which indicates that some things will be duplicated with server rooms on site in Lund, leading to costly backup solutions. To these technical difficulties can be added political issues, such as the fact that Sweden and Denmark have different tax rules, making it difficult for personnel working in both countries. As explained by Rennie; “Now this is subject to some intergovernmental discussions. I don’t know if it has been fully resolved at the moment”. Summing up the interaction between the ESS in Lund, Sweden and the data centre in Copenhagen, Denmark, Rennie argues that the Danes’ part in it and the obstacles faced due to remoteness was the price Sweden had to pay to get the Danish contribution; “if it was the difference between [that and] not having the facility, you know […]”
The Group Leader for Data Analysis at the data centre describes that it has taken time for Danes to realise that there is an ESS centre located in Denmark:
[…] it actually took some time before people realised that now there is actually a data centre […] I still face questions, like, talking to people about being involved in ESS. And they didn’t realise that actually I was located in Copenhagen. I still hear questions about that.
At present, during construction, the data centre has a less important role to play for ESS as a whole than it will in the future, as described by Jonathan Taylor, who has been the head of the data management software centre division since the beginning of 2017. He describes that:
It becomes more important when things go into operation, when there are users and there is actually data being produced. And you have to process the data to publish it. To get scientific impact, you need to use data processing.
As indicated previously, the model for data handling at ESS actually involves two data centres. There is a data centre in Lund and there is a data centre in Copenhagen, where the one in Lund deals with everything concerning the experiments performed on the instruments in the facility on any given day (Taylor, interview, March 2021):
They all in principle have users visiting the facility – about 20 to 30 visiting scientists during any given day during operations. Doing experiments, acquiring data, doing data processing, processing data, data reduction, data analysis.
The data centre on site in Lund deals with the immediate need for computing and storage. That then syncs through the data centre in Copenhagen, which does the archival storage. According to the ESS data policy, the data is stored for a minimum of 10 years. After three years, the data becomes open access (Taylor, interview, March 2021):
For that to work, of course, you have to make a commitment to store the data and make it available. And that’s one of the reasons we have a data centre in Copenhagen. For that capacity.
Scientists who come to ESS to do experiments at the facility will stay for something like three days, and then they go back to their university. During their experiment, they can take something of the order of 10, 20, 40 or 100, maybe even 200 terabytes of data. This means, explains Taylor, that: “That is a big amount of data for someone who works in a university research environment to deal with. Obviously, it’s not going to fit on anyone’s laptop”. The data centre provides the data and the data services out to the community when they are doing their data analysis after their experiment: “The real nitty gritty of the science comes after the fact. Because the neutron data gets combined with other data”.
Another role of the data centre is to store data for combined analysis from several experiments. As explained by the managers:
There’s not so many high-impact publications which just have neutron data in it from a single experiment. The really high-impact research has got lots of data from lots of different experiments and theoretical calculations, and so on and so forth. It all has to be pushed together, into some like coherent, well, in principle, a coherent story.
The whole idea with the data centre can be summarised, as Taylor does, stating that:
It’s the same model in all the large neutron facilities. People come, they do experiments, they go, and then they go back to the facility’s infrastructure remotely to do data processing. So that’s why we built this thing in Copenhagen, really for that.
The construction of the data centre started in 2011, and it had 34 staff in 2021, plus some postdocs working on European projects dealing with the European open science cloud. Between 2013 and 2018, the centre was occupied with lower-level design work and starting a lot of collaborations to get in-kind contributions. During this time period, there were many discussions with in-kind partners concerning what they could contribute and how that would work. With software, especially, there can be a win-win situation, because code written for ESS can be used elsewhere as well. Organising in-kind for the data centre is not as challenging as organising it for accelerator components or instruments, and the value of the data centre’s in-kind contracts is much lower. According to Taylor (interview, March 2021), the largest in-kind package was approximately 2.2m euros, and there was no need for people to travel to Denmark – software can be written anywhere. In-kind contributions are further described in the next section.
4.4 Constructing European spallation source using in-kind procedures
One way to get countries to contribute to a big science facility is to create a system with in-kind contributions. Put simply, in-kind entails the possibility for a country to contribute technology instead of cash to the facility (Åberg, 2013). As described by one of the managers: “The ESS has an in-kind model of funding where something of the order of 60% to 65% of the budget is held by the partners, and then they do in-kind contributions” (Taylor, March 2021). The technology is procured in the specific country responsible for the in-kind contribution, thus, increasing the likelihood of benefits to industry in that country.
When ESS was agreed upon in 2012, Sweden was the only country which would only contribute in-cash. According to our respondents, this arrangement means that Sweden has essentially lost the chance to sell many components and systems, with a loss of opportunities for knowledge transfer and relationship development as a consequence. The Swedish CEO for the Danish data centre explains:
Denmark was much quicker in these regards and do deliver things in-kind […] Sweden on the other hand lacked at the time an organisation that could take this type of discussion.
The fact that Sweden delivers in-cash has the consequence that the more interesting high-tech deliveries are performed by international supplier groups, whereas Swedish firms deliver more basic infrastructural needs at the facility. As explained by Anna Hall, former Director of Big Science Sweden and Industrial Liaison Officer for ESS: “ESS is delivered by in-kind partners and we are bad at grasping this competence development opportunity, the internationalisation, and actually build the structures”. She pinpoints the fact that it is the upstream activities that are most important from an industry perspective, given that only around two percent of the experiments will be run by industry. She argues: “Yes, we can talk use, but we need as well to talk about technological development, and knowledge transfer, the pieces that come from constructing the facilities”. Similar comments were made by the CEO of a Swedish high-tech company (Vieweg, interview, June 2022):
Sweden got its place and we have the whole construction. Most of it is built by Swedish companies, so there Sweden got a lot back, and it is placed in Sweden. But most of the high-tech has been divvied up between other countries. So that is regrettable. We get a super-top-of-the-line facility, but the capabilities for advanced technology have perhaps not benefited Sweden so much. It is what it is.
In 2023, ESS was working under the in-kind model with over 40 European partner institutions and over 130 collaborating institutions worldwide (ESS website). According to ESS (ESS website), “The majority of the instruments, the target station and the accelerator are delivered as in-kind. Overall, about 30% of ESS’s construction budget comes from in-kind contributions from European Partners”.
The same year, in 2023, there was a shift in the in-kind agreements, making Sweden eligible for in-kind contributions for the first time since the start-up of ESS. On 19 January 2023, there was a launch ceremony at ESS for the first Swedish in-kind contract, where representatives from Lund University, Uppsala University, the Royal Institute of Technology (KTH) and the Swedish Research Council were welcomed by the involved staff from ESS. On the ESS website, the following quote was published:
To strengthen the engagement of the Swedish scientific community in ESS, Sweden has now officially become part of the ESS in-kind community with a number of contracts agreed under the Initial Operations funding phase.
In this initial agreement, four different projects, all related to the neutron instruments and support facilities, are included. The agreement covers Swedish in-kind contracts for a total value of 2.2 million euros.
4.5 Current issues at European spallation source (2023 - )
Some interesting turn-of-events connected to tensions between scientific and societal values and needs in the creation of ESS happened in 2023. The first one was the inclusion of Sweden as an in-kind contributor, which took place in January 2023. This has been a sore topic among Swedish actors since the ESS agreement in 2012, that Sweden was missing out on potential knowledge transfer to industry (Ekelöf, interviews).
In the spring of 2023, it became publicly known (see e.g. Sydsvenskan and News Øresund, 31 May 2023) that the ESS data centre has to move from its location in central Copenhagen due to the contract running out at the end of 2024. The newspapers (Ibid.) reported that the staff were unhappy about the lack of information, and the unions launched complaints. In November 2023, it was reported on the ESS website that the data centre was to move to the Lyngby Campus of the Technical University of Denmark, which is situated north of Copenhagen. The data centre started moving in February 2024. With the move, the compromise of two locations, thus, causes further tensions, as the time of travel between ESS in Lund and the data centre is already long today, and the move increases the travel time between the sites significantly (News Øresund, 16 November 2023).
5. Analysis
We stated in the theory section that a compromise involves two or more network actors, which is one reason why we chose to study compromising in the context of a big science facility. Big science facilities, such as ESS, are, as described in the case, highly resource-intensive, complex and multi-levelled, with many actors from different spheres of society involved. We also claimed that compromises have their origin in tensions, why we need to take the plurality and conflicting goals present in the network of involved actors into account. From the case description given, we can identify at least three actor groups: actors representing the political sphere (who have the mandate to represent the civil society actors), actors from the scientific community and actors from the business sector. Each group of actors can be present on different contextual levels, regionally, nationally or internationally, and there are large variations within each group concerning interests, goals, etc. due to differences in possession of resources and network positions (Håkansson and Snehota, 1995).
From our data, it becomes obvious that (at least) two opposing but interrelated forces for big science facilities are working simultaneously in the process of creating the facility, and that these elements at times create tensions between involved actors. These forces entail the divergent wishes of providing value to society, financing it and being involved in the unpredictable and non-linear process that characterises basic science. The case description has demonstrated that these two opposing forces cause goal tensions. It is these tensions that the actors engaged in the process of taking ESS from idea to reality try to handle by acts of compromising on various aspects throughout the construction process. We see that the actors need to continuously and simultaneously manage multiple elements, e.g., both handling the pressure of being valuable to (different parts of) society and the scientific requirements and demands.
The first compromise that is zoomed in on in the case concerns the location, with the facility in Sweden and the data centre in Denmark. This compromise stems from the fact that a facility of this magnitude is an international matter, impacting countries and their economies, as well as the scientific community. The initiative behind ESS came from the scientific community and was originally a German idea, based on the demand for a new and up-to-date facility that could complement and eventually replace the old ones. Due to the large budget needed, however, this was not a decision that could be made by the scientists themselves, but had to be negotiated on an international political level between and among European countries as part of the pre-compromising stage. The case shows that the pre-compromising stage, related to the compromise on location, engaged many actors and took considerable time and effort to agree upon. However, once in the compromising stage, and once the compromise was reached, the focus of all actors was on joint collaboration and value-creation (Håkansson and Snehota, 1995).
The fact that Sweden and Denmark were able to host the facility was perceived nationally in Sweden by both scientists and politicians as a victory, but the case illustrates that the Swedes saw the co-hosting arrangements as a necessary compromise in this, albeit that it entailed considerable pre-compromise work (Oldenhof et al., 2014). Due to the pluralistic context of the big science facility, however, the compromise was both the (only) likely and preferable solution (van Bommel, 2014). The difference between the compromise of ESS location and many other compromises discussed in literature is the fact that, unlike compromises on interpretations or documents (see e.g. van Bommel, 2014; Trevisan and Mouritsen, 2023), once the compromise is settled, then the physical resources dedicated to the facility (Wadell and Bengtson, 2023) mean that it is difficult to go back on the compromise. Once decided, it becomes a factual circumstance and a joint commitment of involved actors that needs to be managed in the continued construction process (cf. Jones and O’Flynn, 2012). This was a compromise that they were willing to make, but the case also shows that it did not come without difficulties in the post-compromising stage for the running of the operations at the facility.
The second compromise described in the case, the one related to the in-kind contributions, was a compromise made to get enough funding to make the investment in the facility. In the pre-compromising stage, we witness a tension between the need for funding of the facility – i.e. what is needed for research, and the desire of nation states to get a return on the money spent – i.e. what is economically and technically desirable for industry and nation states. This led up to the in-kind compromise as described in the case. This compromise relates to the first compromise on location and co-host arrangements in the sense that location influenced the hosting arrangements and, thus, who was responsible for the funding arrangements. To attract other nation-states to invest in the facility, a guarantee of the return on investment was crucial. However, it has also been perceived as negatively affecting Swedish business actors and the national industry in Sweden, which Sweden, as the host, delivers in cash. The chances for the facility to become the local growth engine that many Swedish politicians had hoped for decreased, relating to more negative descriptions of compromises (Tidström, 2014).
We see in the post-compromise stage how Swedish companies, due to the compromise arrangements, had a much harder time bidding on the most technically interesting parts, as these were procured directly in other member countries. Instead, a Swedish perception is that they were stuck with parts that were not announced as in-kind packages, often containing fewer innovation and development aspects. The tension behind this compromise resurfaces in the pre-compromising stage as part of the goal problem (what the goal of the construction is and who will benefit from it). In this case, it becamea compromise affecting the Swedish business actors and a compromise that was perceived as necessary to get funding (a problem both national and scientific). In contrast to the first compromise, however, we see that the arrangements in this compromise were not irreversible, but could be shifted in the post-compromise stage, and after 10 years, Swedish in-kind contributions were allowed.
One reason for the tensions that surface and the need for compromises in our case study has to do with the plurality of actors and the various goals that they each have, which is a situation that has to be embraced in the creation of large scientific facilities of this type. The creation of ESS could not be managed without the connection to parties other than the purely scientific ones. It is shown that there is a need to legitimise the facility and its large budget based on taxpayers’ money, both in the host country and towards other countries, in order for them to become members and thereby provide part of the overall budget for construction. Hence, we see an aspiration to make this into something larger than the facility as such, or in the wordings of the former CEO of Big Science Sweden (an interest organisation with the mission to promote the interaction between big science facilities and Swedish industry): “A cool eco-system that can attract competence and prosperity to the region”.
6. Conclusion
Our case study has demonstrated that a lot of interaction between various actors is needed throughout the creation process to progress in a megaproject. Due to the long time span and the intensity of work for several, a network of relationships between involved actors is developed in the process (Håkansson and Snehota, 2017). The case also shows that the process is characterised by a need from involved actors, controlling various resources that are central for the creation, to balance and manage different goals and expectations on the project and the resulting facility. These goals are formulated by different actors with varying involvement at various stages in the creation, and at times, there are tensions between goals that surface and need to be managed. The case illustrates that compromising is one way that the involved actors use to deal with these tensions.
The compromise reached at a certain time, like the ones dealing with co-host arrangements and in-kind procedures, might vanish in importance, however, the tension between competing demands will persist and resist closure (Gaim et al., 2022) forcing the actors to keep living with it (Wenzel et al., 2019). Given the plurality of network actors in the processes of this sort, we argue that more compromises can be highly expected in the continued process, making the phenomenon an important area for further studies.
6.1 Implications for research
This study focuses on compromises as process. It is argued that compromises become a necessity due to network tensions stemming from different goals of the various actors involved, in our case of actors from the political, societal, scientific and industry sides (cf. Halinen and Törnroos, 1998). The study has shown that the different actors involved had no choice but to live with the goal-related tensions found in the process of realising the facility. One way of living with these tensions, and the one that we have been focusing on here, is how the actors were compromising on various matters where tensions surfaced in the process. Hence, we conclude that it will at times be necessary for involved actors to make compromises in order for a certain process to continue.
We show in the case description that conflicting goals are handled continuously by the involved network actors through a strategy that often includes a need to compromise, and a need to live with and make the best of the consequences of various compromises. Based on this reasoning, we conclude that a compromise is best understood as a process phenomenon. As such, it should be studied through time, where there is a pre-compromising stage that helps understanding why various actors have certain interests, a compromising stage that has been studied previously by researchers interested in negotiations (see e.g. Åge and Eklinder-Frick, 2017; Eklinder-Frick and Åge, 2020) and a post-compromising stage where the performative process of constructing the facility continues with the compromise as a mutual commitment that directs and restricts the continued performance of the involved actors. Our study has shown that in the interaction among actors in industrial networks of plurality and conflicting goals, an important task becomes knowing when and how to make compromises, and how to best live with the effects of these compromises.
The compromises we studied were pragmatic compromises (Jones and O’Flynn, 2012) made out of necessity. Whether compromises are good or bad, fair or unfair, is, in that sense, uninteresting. The compromises that were made needed to happen. What becomes much more productive is investigating how the effects of the compromises made can be dealt with moving forward. Our third conclusion is, therefore, that the important question concerning compromises becomes how to deal with the act of compromising over time. We suggest that taking both (all) sides of a goal conflict into consideration is most often a must to find a workable way forward, and evaluating the effects not just at the time of compromise, but in the post-compromising stage is vital. Finding ways to compromise and to deal with the long-term effects of compromises in a productive manner is, therefore, a managerial task that needs to be prioritised.
6.2 Managerial implications
Pluralistic settings like megaprojects often contain tensions and persistent, both competing and complementary, forces. The act of compromising seems, therefore, to be highly important for managers to understand and handle. Managers need to prioritise finding ways to live with and handle tensions through means such as compromises.
Our reasoning on compromises as process implies that interaction, communication and knowledge about the perspective of different actors are valuable assets to understand the demands of various parties and how a functioning solution that takes different goals into consideration can be found. The managerial mindset gains from being one where compromises do not necessarily need to be either/or, but can incorporate considerations from several parties using more of a both/and, or happy-happy, thinking (cf. Åge and Eklinder-Frick, 2017). In practice, this means listening to different actors, interacting with them and finding solutions that are at least acceptable for everyone.
Another important managerial task is knowing how to deal with the continuation of the process, given the direction it takes based on a certain compromise. The compromise as such directs the process, but much room for manoeuvring is still there, leaving space for involved managers to influence the continuation of the process.
A third task, given the fact that new tensions are bound to arise in this type of setting, concerns knowing when further compromises need to be made and handling the process of compromising in ways that impact the continuation of the overall process positively from a project perspective. Finding compromises that can be accepted by the mutual parties creates commitment from these parties and is, therefore, positive for the continuation of the creation process.
6.3 Suggestions for future research
One area for further research concerns the heterogeneous nature of the involved actors in megaprojects. In the paper, this aspect has been handled as a reason for tensions, but it has not been further investigated. Non-business actors of various types have, however, been studied in previous research using the Industrial Network Approach (e.g. Åberg, 2013; Leite, 2022; Romestant, 2020), wherefore it would be interesting to build on these studies to further investigate the nature and resource control of different actors, how they impact the process of creating a megaproject in different ways over time and how their goals differ and change through interaction with other actors in the process.
Another interesting finding in this study deals with the perception of compromises. We have seen in our interviews that the compromises described by some of the involved actors are not perceived as compromises by others. These differences in perceptions have consequences for the continued process of creating ESS. Actors that were involved early in the process view, for example, the joint host arrangement as a compromise, whereas actors that join later just see it as a reality, the way it has been designed. These differences will impact future action and ultimately what performing stability will be reached in the process. We stumbled on these results, and more research on compromise as process is needed.
More research could also be conducted on the aspect of commitment and compromise. Jones and O’Flynn (2012, p. 129) point out that the stability of a compromise depends on the mutual commitment, because “the parties have a greater obligation, as well as a greater inclination, to comply with a compromise to which they have consented”. Compared to compromise, commitment is a much more investigated phenomenon in business network research (see e.g. Hadjikhani and Thilenius, 2005), and interesting avenues can most likely be found. It would be interesting to study how committed various actors are to a certain compromise, what impacts their level of commitment and what the consequences of breaking a commitment of this type are. Related to dynamics and to time, one might even ask if it is possible to break a commitment of this type, and if so, for how long?
This study has focused on compromises in a setting where basic science is heavily involved. It would be interesting to test our findings on processes involving plurality, but in another project setting within a different focal area, be it health care or industry collaboration concerning sustainability. Our empirical investigation concerns actors involved in a project setting. Future studies could investigate compromises in change processes in more long-term and stable contexts. Also, since this is a single case study, there is a need for future studies of compromises using other methods and study designs.
Notes
Large-scale research facilities are often referred to as “Big Science”, a concept that emerged in the 1960s (Weinberg, 1961; Price, 1963). In short, it refers to scientific research that needs big investments, big facilities, big machines and many people involved. More recently, the political aspect has been added to list, i.e. big science also needs “big politics”.
This quote refers to a quote from Computer Networks (3rd ed, 1996, p. 83) where the author Andrew Tanenbaum paraphrases Dr. Warren Jackson (Director, University of Toronto Computing Services (UTCS) circa 1985), stating that you should: “Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway”.
