Chapter 7 – Creating Shared Understanding in Research Across Distance: Distance Collaboratiion across Cultures in R&D
Planning and managing globally distributed teams presents exceptional challenges in most circumstances. In innovative work, such as research and development (R&D), where questions are often emergent and processes cannot always be clearly defined in advance, these challenges are exacerbated. A difficult balance between in-person and technology-mediated communications must be achieved when R&D teams are spread over multiple locations. On the one hand, the complexity of research projects requires resources, skills and knowledge that are not found within any single location. On the other hand, research and, more generally, innovation require cooperation that has traditionally been facilitated by participants meeting in person around a table. In particular, it is not clear to what extent the discovery and interpretation of new research problems requires the co-presence of researchers. As projects become more complex and are distributed over greater distances, in-person meetings are often not feasible or practical. They are increasingly being replaced by the use of communication technologies. While these technologies allow more frequent communications between distant collaborators, they also reduce the “bandwidth” of that communication. Facial expressions, body language, tone of voice, etc. may be lost. However, successful collaboration, especially for innovative research, relies to a large extent on intense interaction to create a shared language, a common understanding of problems, and the trust required for members of a group to suggest original and untried solutions. These crucial factors are not readily transferred across electronic networks. How, then, do corporations and researchers deal with the challenge of collaborating across geographic distance and organizational boundaries using technology-mediated communication? How are differences in education and national culture bridged? Through analysis of a series of qualitative interviews in international R&D centers in Bangalore, India, these questions are addressed in this chapter.
As globalization and the impact of worldwide communications networks expand, scholars are increasingly debating the role of physical place in the world. Predictions have ranged from utopian (the “death of distance,” “anytime, anyplace communications”) to dystopian (rampant global capitalism and the disintegration of societies and traditions). This chapter discusses the role of shared space in innovation, specifically in corporate research and development (R&D). In particular, it explores the role of intangible factors, such as tacit knowledge, shared language, trust and conventions for researchers who are separated by large distances and several time zones as they collaborate on projects.
Collaborative innovation and creativity arise from interactions between people. New ideas are developed by grappling with and rethinking existing ones. Once the ideas are developed, their implementation may be partially achievable through existing technologies, processes or heuristics. In the initial stages, however, innovation requires a willingness to engage with new perspectives, hunches and seemingly absurd suggestions. These are only partly evaluated through rational decision-making. Judging the potential of a new idea relies as much on experience and intuition as it does on logical analysis. While a variety of different perspectives, backgrounds and life experiences is helpful for a group in generating new ideas, expressing and evaluating those ideas collaboratively presumes a shared language, a shared understanding of the problem at hand and a shared set of experiences. These are developed through close and rich personal interactions between group members. They are particularly hard to develop across distance.
This chapter discusses researchers’ and managers’ everyday experiences and strategies as they engage in collaboration across thousands of kilometers and multiple time zones. It is based on a series of interviews with researchers based in one of the world’s new R&D “hotspots,” Bangalore, India. The analysis of daily, individual interactions reveals the limits of innovating across distance. Barriers to collaboration and innovation can lie in the simplest effects of distance: differing time zones or strange accents. They are also found in a simple lack of physical presence: not having shared a meal or a drink together or missing facial and physical feedback that cannot be verbalized. Most importantly, they are found in the uncertainty that people have about someone they have never met, never looked in the eye, shaken hands with, or answered in the affirmative: “Can I trust this person?”
WHY LOCATION (STILL) MATTERS
Studies of regional innovation systems (Freeman, 1991) highlight the importance of collocation for innovation. Many recent studies stress the importance of intangible factors for successful collaboration in innovation. Saxenian (1994) describes the role of personal and professional networks in fostering the flow of tacit knowledge and the generation of trust. Storper (1997) argues that innovation centers on processing uncodified information that cannot be understood outside the context of the agents transmitting it. This information can be scientific, social or political. For successful innovative collaboration, “dense and multiple understandings of what is being transacted are required, that is, ways of reading between the lines, of verifying in multiple ways the possible meanings of what is inherently uncertain formal content” (Storper, 1997: 37). This understanding can be achieved in two different ways: it can be facilitated by a long-standing personal relationship based on trust between the collaborating individuals, or it can be based on less idiosyncratic relational assets. These last are defined as “taken-for-granted mutually coherent expectations, routines, and practices, which are sometimes manifested as formal institutions and rules, but often not” (Storper, 1997: 38).
All collaboration for innovation involves risks. Collaborators must work through repeated phases of uncertainty, surprise and failure before they reach a solution. They must be open to new ideas, but also willing to give and accept criticism. And, they must have faith that all team members are contributing their share and will receive adequate recognition if the solution is successful. Legal instruments such as employment and alliance contracts or patent laws can provide some reassurances. However, successful collaboration requires a much more personal form of trust, a minimum level of goodwill between collaborators. Fear of being ridiculed or taken advantage of inhibits the flow of new ideas.
Traditionally, this kind of trust has been based on shared membership in the same social group. It relies on repeated shared experiences (Child, 2001). Collaborators working across large distances are unlikely to be members of the same social group and will have a very limited basis for sharing experiences. Nevertheless, trust can be established between collaborators located far apart from each other who rely on various forms of technology to communicate. Jarvenpaa and Leidner (1998) demonstrated “swift trust” in dispersed teams that worked together on a project temporarily. These group members assumed the trustworthiness of their collaborators (for the duration of the project) because this was in the group’s interest of achieving its project goals. Swift trust works best in situations where collaborators rally around a clear set of goals and tasks.
Unfortunately, innovative collaboration precludes a set of clear objectives, much less a task list that can be handed out and crossed off. However, creating a higher-level awareness among group members about the purpose and importance of their collaboration can contribute to fostering a basic sense of trust. Repeated collaborations can then lead to predictability and firmer forms of trust over time (Davenport, Davies, & Grimes, 1998; Dodgson, 1993). Video and audio conferencing are generally seen to be better suited to establishing trust than text-based communication. But even with the use of these richer media, trust is likely to be slower to fully develop and to be more fragile than between collocated group members (Bos, Olson, Gergle, Olson, & Wright, 2002).
Mistrust can also arise because technology-mediated communication gives a very limited insight into the events at a distant location. An audio conference participant has no way of knowing whether other conference participants who are sitting in a room together are quietly signaling each other. Once mediated communication ends, participants to a conversation have no recourse to observations or chance encounters that might help them gage the sincerity of the other participants. Collocated subgroups can severely impair the overall functioning of a dispersed group. The effect is exacerbated when each subgroup is composed of members of the same nationality (Polzer, Crisp, Jarvenpaa, & Kim, 2006).
Access to tacit knowledge
Sharing tacit knowledge and passing on knowledge through socialization or learning-by-doing are important aspects of innovation (Nonaka & Takeuchi, 1995; Von Krogh, Ichijo, & Nonaka, 2000; Nonaka & Teece, 2001). Tacit knowledge is not wholly specifiable: the sum of the parts does not fully describe the whole; an observer must refine his understanding of the whole through experience and intuition.[i] As a result, we “can know more than we can tell” (Polanyi, 1961: 30). Tacit knowledge is knowledge that has not been expressed in language or symbols (codified), either because it cannot be fully codified or because it has not been necessary or expedient to codify it. It includes the beliefs and assumptions that people hold without being aware of them; it is available and in use, but difficult to share; and it is often a source of technological and competitive advantage. Nonaka (1998) describes knowledge creation as the process of making tacit knowledge explicit: creating explicit models of previously tacit know-how, assumptions and beliefs. This process requires intense commitment, challenges, discussions, brainstorming and even distractions. It is very difficult for a group to achieve without sharing space and experiences.
The flow of tacit knowledge is particularly important for activities that cannot be uniformly structured, such as innovation and creative work, or work that requires the interpretation of physical or social cues (the two are obviously not mutually exclusive). In research and other innovative work, problems and solutions emerge as hunches or fragments of ideas long before they can be fully and unambiguously expressed. Passing on this kind of knowledge requires, for the most part, face-to-face interactions.
Shared understanding represents a “collective way of organizing relevant information” (Hinds & Weisband, 2003: 21) and lets a group collaborate more effectively. It encompasses ideas about group goals, the task itself, social aspects of the group (such as roles, interdependencies and communication patterns) and the characteristics and activities of individual group members (Hinds & Weisband, 2003). These ideas are “understood” within a group and therefore information about them does not need to be transmitted with every interaction. They can help group members to anticipate the reactions of other members and allow them to coordinate their actions implicitly.
Shared understanding is “developed through a history of communication, past coordinated action, and/or other common experience (such as professional socialization)” (Dickey, Wasko, Chudoba & Thatcher, 2006, introduction). It is easy to see how it might be created through frequent, media-rich, face-to-face communication (Daft & Lengel, 1986). According to Hinds and Weisband (2003), physical presence increases collaborators’ unintentional, informal and social information exchanges and their awareness of each others’ work. It also provides a shared context for the interpretation of new information.
Nevertheless, it is sometimes possible to achieve shared understanding through archived text-based communication (e.g., an on-line bulletin board) since, unlike most face-to-face communication, a central text repository can stabilize meanings across space and time (e.g., Dickey, Wasko, Chudoba & Thatcher, 2006). Many complex, distributed, collaborative projects have been carried out through technology-mediated communication. The French aviation company Dassault designed an entire business jet in a purely virtual environment, including a database of 40,000 parts and 200,000 fasteners that was shared between workers at some 30 different firms. (The Economist, 2005, 16 June) Even pure text-based collaboration has been successful despite its lack of media richness, for example, in developing open source software. Dickey and colleagues (Dickey, Wakso, Chudoba & Thatcher, 2006) propose that it is not the technological characteristic of the medium, but rather the shared understanding (or lack thereof) in a group that determines its ability to deal with complex, ambiguous tasks across distance.
Despite these advances, face-to-face meetings remain extremely important in establishing shared understanding as the basis for subsequent collaboration. Salter and Gann (2003) found that engineers preferred face-to-face communication for gathering ideas during the initial and most ambiguous stage of their projects. Kaufmann, Lehner and Todtling (2003) find evidence that technology-mediated communication is more suited to maintaining and managing existing relationships than to creating new relationships. In their review of the literature on proximity in work groups, Kiesler and Cummings (2002) find that face-to-face discussion increases cooperation by strengthening bonds, social contracts and group identity. In addition, they find that a high frequency of spontaneous, informal communications plays an important role in creating strong social and work ties between group members and thereby facilitates the exchange of complex knowledge. Tools such as instant messaging facilitate spontaneous communication for distributed groups, but may not suffice.
As the distance between collaborators increases, it becomes more likely that they will be working in different regional or national cultures. Cultural differences complicate the picture because of a priori dissimilar norms and identities. Collaborators who work in different cultural settings are also more likely to be demographically diverse and to have different educational backgrounds. Their perspectives diverge more, and they have a narrower base of shared understanding on which to build. They are more likely to face language difficulties, which in turn inhibit participation in technology-mediated collaboration (Kim & Bonk, 2002; Sarker, 2005). On the other hand, there is some evidence that shared national, ethnic or organizational culture can enhance collaboration across distance: Agrawal, Kapur and McHale (2006) found that shared ethnicity can partially substitute for collocation in facilitating the flow of tacit knowledge.
In summary, shared understanding and the flow of tacit knowledge are key elements of innovative collaboration; and trust, shared context, and frequent interactions (both formal and informal) are central to establishing these elements. Despite proclamations of the “death of distance” (Cairncross, 1997), physical presence and shared space are still important in today’s world, especially for people who are highly interdependent and faced with ambiguous situations. This study was designed to explore the potential and the limitations of distance collaboration in R&D, given the communication tools available.
The study is based on a series of individual interviews conducted in 2005 and 2006 in R&D centers in Bangalore, India. It is also informed by several additional interviews and countless informal discussions that took place during that time. Interviews were chosen over a written survey to elicit richer and more open-ended responses. Also, several academics in India discouraged me from conducting written surveys as they had found response rates to be extremely low, and questionnaires were more often than not filled out by a secretary, rather than by members of the target audience.
Interview method and limitations
The format of the interviews was a guided conversation, conducted in a meeting space at the respondent’s workplaces. The interviewer guided the discussion to include most or all of a pre- determined set of topics, but allowed the respondents to concentrate on or add topics relevant to interest or expertise.[ii]
Respondents were asked to describe the geographical distribution and internal organization of their teams, their use of and experience with different communication technologies (incl. travel/face-to-face[iii]) within their teams, the role of cultural and educational differences between team members, and their collaborative networks outside of their team. As often as possible, they were asked to describe specific instances, using current or recent projects as examples. All respondents remained anonymous, and any respondent could refuse to answer a question. Even so, some topics were automatically excluded from the interviews. For example, due to concerns over protection of intellectual property, respondents could not share details about the content of their research. Some managers agreed to be interviewed on the condition that I not ask them about investment numbers. These restrictions were not a direct concern, since the goal of the interviews was to uncover personal work experiences. However, it does render the discussion of work practices very abstract. Many statements would be easier to understand with more specific technical context.
Thirty-one qualitative interviews were conducted in 17 R&D centers (all engaged in some form of international research collaboration). All centers were focused on information technology or telecommunications research, although some belonged to parent companies in other industries. Seventeen respondents were researchers and team leaders in these labs; fourteen were senior R&D managers. Ten additional expert interviews were conducted to gain context and background knowledge. The experts also served as sounding boards for some initial, rough conclusions.
Respondents were identified through the interviewer’s personal contacts and through references by other respondents. This approach had two advantages: personal references greatly facilitated access to respondents, and the (indirect) personal connection allowed for a more frank and open discussion. For the purposes of this study, these advantages were seen to outweigh the obvious disadvantage that such a sample cannot strictly claim to be representative of the larger population.
The results of all interviews together are presented in the following sections. Direct quotes are from the core set of 31 interviews conducted in R&D centers.
RESPONDENTS’ EXPERIENCES OF COLLABORATING ACROSS DISTANCE AND CULTURES
I begin this section by providing a general overview of respondents’ experience of R&D collaboration across thousands of kilometers, different national cultures and multiple time zones. The subsequent sections relate to respondents’ experience of the issues of tacit knowledge, building trust and creating shared understanding.
Separated in time and space
At first glance, one might think that there is not much to differentiate long-distance and face-to-face collaboration. Respondents repeatedly answered, “No,” when asked whether they could recount a specific episode of long-distance collaboration that was particularly difficult or frustrating. A hesitancy to acknowledge problems to the interviewer may play a role here. However, work across distance and the difficulties it presents are also taken for granted by many people who have spent most of their careers in an MNC subsidiary or an outsourcing firm. In this sense, respondents may simply not have striking examples to recount.
As the interviews progressed, respondents’ strategies for dealing with or avoiding everyday difficulties of long- distance collaboration became apparent. The use of these strategies suggests the existence of difficulties and indirectly sheds light on the limitations of technology-mediated communication. The prevalence of air travel also belies the “absence of problems” with technology-mediated communication. Almost all respondents had been or were planning to go abroad to meet some of their collaborators.
The most frequently cited problem of work across distance was adjusting for different time zones. Several respondents specifically mentioned that geographical distance was not so much of a problem as time, as reflected in the following remark:
Being in India can sometimes make work with product units difficult. The
distance is not that important, more the time zones. The time zones make it very
difficult. We can really only schedule calls at 8:30pm, which is 8am for them. So
it’s a bad time for us and it’s a bad time for them.
For another respondent, time zones were the only major difference between his previous collaboration in the US and his current intercontinental collaboration: “There’s not much difference between Texas-Boston collaboration or Texas-Bangalore collaboration. Only that it’s night in the US when it’s day in India.” The difference of 9½ to 12½ hours between India and the United States results in very little or no overlap in the official working hours in both places. It also means that there is a limit to the duration of synchronous communication. During regular working hours, each person can only transmit one round of messages per day and has to wait until the following day for an answer. Respondents in this situation missed the immediacy of feedback that they could get from someone working within the same space:
The main difficulty is if you want to explain something, you can’t use the board.
It takes more time to write something in an e-mail. If you’re working in the same
location, you drop by the other person’s office, put something up on the board
and together you can immediately figure out whether it works or not. This way,
you talk on the phone, e-mail a document, and talk on the phone again.[iv]
There was a strong tendency in most labs to reduce interdependency with researchers in distant locations as much as company policy would allow. In some cases this meant defining clear boundaries and responsibilities within a project; in other cases it meant conducting entire projects independently of labs in other countries. This approach defeats the R&D strategy of those firms that use distributed teams in order to make optimal use of experts and knowledge located in different areas of the world (Gassmann & Von Zedtwitz, 1998).
Respondents sometimes struggled with language issues when they collaborated with non-native English speakers, or even when both sides were fluent in English. Dialects, the choice of words and technical language caused difficulties or misunderstandings. Accents were another source of confusion. Some language issues, however, were alleviated through the right choice of communication technology. E-mail allows people to take more time to understand or compose a message; both e-mail and instant messaging even out accents and eliminate some cultural markers. Some respondents thought that asynchronous, written communication was the best alternative to face-to-face meetings in these cases because synchronous communication exacerbates language problems and the resulting embarrassment or discomfort on both sides.
Access to tacit knowledge
A few R&D centers were specifically set up in Bangalore to provide their parent companies with a better understanding of the Indian market. However, most labs were still focused on American or world markets. Their researchers lacked a direct, personal connection with end-customers and considered this a severe disadvantage compared to their Western counterparts. As expressed by one respondent, these counterparts “had an advantage in terms of proximity to the customer. They had access to more information, a refined understanding.”
Multinational companies increasingly locate their R&D labs close to production facilities. The software research labs in this study benefited somewhat from their proximity to Bangalore’s IT industry. However, most had no direct, physical access to factory floors, semi-conductor fabs or sometimes even software engineering and maintenance facilities. In these cases, training one on-site expert is sometimes not enough. It is more difficult for someone who has recently acquired knowledge to pass it on coherently than for an expert to teach it. Teams also take time to discuss new insights in order to relate abstract ideas to their specific tasks and goals: “Now, there are meetings for technical work. Because of the larger team size. It became harder for one person to pass on everything. It takes much longer for people to learn second-hand what they would learn by going out.”
Politically sensitive information proved the most difficult to access across distance. Often, the “real story” isn’t available by e-mail because people don’t want impolitic statements traced back to them. Even telephone conversations require a basic level of trust that is hard to establish without sitting in the same room:
I need to understand the evolution. I want to understand why decisions were
taken. If I have ideas on how to improve something, I need to know why the
decision wasn’t taken in the first place. Often the reason was political, so it’s not
documented and people won’t put it on e-mail.
Building trust and personal relationships
A frequent and important experience for many respondents was that collaboration became more “comfortable” after in-person meetings. A variety of mental barriers to intense and open collaboration were reduced. As the examples below show, what respondents describe as comfort can be interpreted as basic trust in a collaborator’s good will. Predictability and knowing how to “read” a person’s response are particularly important:
Meeting someone makes a difference even though science is supposed to be
objective. It increases your comfort level; you know how someone responds in a
certain situation. It also helps when you have non-technical discussions.
This is an important part of building trust between collaborators. Innovation, by definition, involves taking risks – not just for an organization, but also for the individuals who contribute new, untested ideas: “It makes it easier that I’ve met them. It’s easier to just say things. Without worrying about looking stupid and things like that. To say, ‘I wrote this code, what do you think?’” Informal social activities play an important role in increasing the comfort levels. Eating and drinking together or meeting someone’s family provided a broader base for the relationship and more insight into the other person’s character. Again, this helped to reduce uncertainty and build trust.
Researchers were also aware of the effect that in-person meetings had on relationships with superiors. Through a general conversation, it was possible to establish a personal connection and lay the groundwork for addressing future, more specific issues. The following two comments reflect this:
Also, it’s important for discussions about the future direction of your work. You
can discuss with management, get their buy-in.
When I am in Bangalore, I don’t have any contact with managers in the US. But
when I went there, I had meetings with them. That helps when there’s a project
proposal that needs funding. It’s easier to send an e-mail and ask the US for
Several respondents mentioned that in-person meetings were better suited for sensitive issues. General outsourcing fears played a role here. Respondents were highly sensitive to their counterparts’ possible fears of losing their jobs to Bangalore-based labs or simply of having to work with people whose qualifications they couldn’t judge. More specifically, respondents mentioned that misunderstandings between team members were more easily cleared up at meetings, where it was possible to bring everyone together at the same table.
For collaboration involving language difficulties, face-to-face communication and the increase in comfort levels that results from meeting a collaborator in person have a heightened importance:
There were problems with the accent in some cases, especially with the technical
guys. Also with the language, some of them had difficulty explaining computer
terms in English. I met them in India and in the US. Meeting them helped. Now, I
can just call them up. There’s less frustration about communication difficulties.
You can explain the same thing three times and it’s ok.
Respondents who collaborate with both the United States and East Asia traveled more frequently to meet their Asian than their American counterparts. Physical presence helped them to overcome many language issues and made it easier to pick up the subtle cues that let them ask the right questions to resolve uncertainties.
Creating shared understandings
Shared understanding was an issue at many different levels. The most elementary level concerned communication styles. Did the same sentence mean the same thing in both locations? The next level concerned general work and management practices, e.g., what is the role of hierarchy in an organization? Related to this, the interviews revealed that the practice of corporate research was conceived differently by different groups, leading to misunderstandings, frustration and prejudices. Group identities were an important factor in alleviating the difficulties that respondents faced as they tried to create shared understandings across distance.
Respondents were asked about the differences and similarities between Indian researchers and their foreign collaborators. Most responses concerned the level of directness in communications and respect for hierarchy: “In the West in general, people are more direct and blunt. In India, people are quieter. It takes more experience for managers to know what is really happening.”
In general, the relationship to superiors is a noticeable difference between India and other (Western) countries. A strong top-down, hierarchical organization is still dominant in both the education system and many domestic firms: “The mindset of freshers[v] is: My superior will tell me what to do. So bosses have to tell them that people need to take ownership of their projects, encourage them and lead by example.” However, differences between India and the United States appear far less significant when placed in the context of world-wide cultural differences. Respondents who worked with colleagues in multiple countries rarely related questions about cultural differences to the United States. Their initial response in these cases focused on East Asia or Europe. This suggests that differences in the work culture between labs in Bangalore and the United States – while present – recede into the background when compared to the differences between Bangalore and other locations around the world.
Organizational and professional identity
Research managers were acutely aware of the problems that could arise from differences in national culture. They consciously tried to reduce differences between their lab and affiliated labs world-wide. The main way of achieving this was to place an emphasis on a shared corporate culture over and above national differences.
The initial struggle of Bangalore-based labs to be perceived as qualitatively equal to Western labs plays a significant role in replicating corporate culture. In trying to prove their worth, employees in these labs orient themselves by the norms and practices of their international peers and their company’s headquarters. Building this reputation leads to an emphasis on similarities between labs rather than differences:
It becomes the job of internationally experienced people to teach others how their work and their ideas will be perceived by collaborators outside of India: “There has to be a kind of apprenticeship for people hired in India. For that you need people who understand the culture on both sides. It’s important to put people in touch with our other labs.”
As suggested by the idea of an apprenticeship, this happens more often through conversations and providing an example rather than by fiat or through formal training (although the latter is also available in some companies). Such socialization happens ‘naturally’ as long as there are enough people with the relevant experience in the organization.
In addition, travel is generally seen as an important means of replicating a global corporate culture in the Bangalore research unit. About half of the labs in this study have a formal policy of sending new hires abroad when they start or soon thereafter. Researchers are encouraged to visit headquarters or other corporate offices if they are traveling nearby.
Managers also use corporate policies as a way to reinforce the feeling that different labs worldwide work on equal terms. In the case of Google, “Bangalore is a peer to all global labs. The culture, the hiring criteria, everything is similar to the US.” In many cases, architecture and interior design were used to highlight the similarities between a Bangalore lab and international labs. Technology campuses were built to replicate the feel of a Silicon Valley technology campus, complete with gyms, cafeterias, and other amenities. Many campuses are gated – as much for security as to create a border between the dust and chaos of Indian roads and the clean, ordered atmosphere on the campus.[vi] Smaller labs used simpler touches, e.g., corporate colors or self-serve coffee machines.[vii] In addition, newsletters, webcasts of talks and intranets are used to keep people in different labs aware of each other’s work – in the hope of encouraging more spontaneous connections.
Even as they pointed out cultural differences, respondents did not mention insurmountable obstacles. They treated difficulties as tactical rather than as fundamental problems. In part, their confidence stemmed from a feeling of equality and connectedness with their international colleagues. As one respondent noted, “Within the company it feels like it’s all one set of people.” Another suggested “People hired here have a lot of hi-tech exposure. They feel they have the same culture despite being in different places.” Corporate efforts are supported by a sense that technology workers already share a similar ethos, regardless of the country where they work.
Shared understanding in the research process
In order to collaborate effectively, researchers need to understand the activity of performing research in fundamentally similar ways. While some respondents thought that research was understood the same way in all cultures they had experience with, others pointed out nuanced differences. Respondents who thought that there was a difference in how research was understood in India and elsewhere often related this to differences in PhD education. There was a sense that (outside a handful of elite universities) doctoral research was not independent enough, too hierarchical.
A PhD in India or China is not ideal training for research. There is a lot of
development talent here. But for research it is more difficult. The PhD here is
more hierarchical. It is more about finding a solution – more like a Master’s.
PhDs from the US know how to do something new, not implementation. In a
PhD, you should figure out what the problem is, what are the questions. In
pharma, space and physics it might be different. But in computer science we
don’t get that.
This approach encourages the development of problem-solving skills over problem discovery or problem definition. As a result, internationally experienced researchers found some of their Indian- trained colleagues to be ill-prepared for the fuzzy goals and unstructured work environment of multinational companies’ R&D labs.
Overall, the significance of differences in research experience and education is currently fairly limited since many researchers at the R&D labs in this study received some or all of their tertiary education in the United States. Estimates of the number of people with a foreign degree and/or several years of foreign research experience ranged from 30 percent to 90 percent of a given research group.
In-person communication is most important in the initial stages of the research process: discovering problems, understanding and contextualizing them, and defining goals and priorities. In MNCs’ formal processes, defining research goals and setting priorities was highly associated with in-person communication. While some companies have well-defined, IT-supported processes for deciding on a research strategy, the process itself is dependent on physical interaction. Senior managers have meetings with clients to better understand new market needs. (Even those labs most focused on fundamental research placed a high value on input from business units and customers.)
Aside from the in-person meetings required to gather inputs, the management team itself needed to meet face-to-face to evaluate the inputs. It took several days of “sitting together” to sort out this wide array of inputs and to reach a consensus about the meaning and relative importance of different pieces of information.
The research process itself is considered to be highly dependent on shared space if tacit information and its interpretation are to permeate the research group:
There is no fixed, standard process despite all the tools. It requires inspiration,
and inspiration isn’t the outcome of a specific process. Knowing the necessities
doesn’t lead to the solution. You need to sit with the information. It’s important to
make people interact so that osmosis can happen.
Characteristically, face-to-face meetings were used in most labs to kick off a project. After this initial phase, research teams experienced recurrent periods of uncertainty. R&D is iterative: once basic problems are solved, new problems emerge. Therefore, it comes as no surprise that project milestones were also often cited as instances where researchers traveled so that the team could meet in person. The less well-defined a task or goal is, the more important it seems that people can sit around a table or blackboard to figure it out: “This was a difficult problem. We didn’t know how to go about it. We had four- hour meetings to figure out the approach. Once the approach is decided, you can use phone and e-mail.” For this kind of discussion, immediate – or even simultaneous – feedback is necessary and rarely achievable through mediated communications.
The difference between technology-mediated and in-person communication is not that some tasks are impossible to carry out across distance, but rather that they are carried out differently: timelines are different and task definitions are revised to accommodate distance and time zones.
While research collaboration without in-person contact is possible, collaborating on R&D across large distances is far from ideal. Embodied and product-specific knowledge tend to be very important. Researchers do not feel they can assume that a collaborator means well – especially in view of political debates about outsourcing, offshoring and downsizing. There are important differences in communication behavior and comfort levels between collaborators who have met and those who have not. For the researchers in this study, in-person meetings led to a higher level of comfort in collaboration. They trusted their counterparts more and moved the relationship from a purely professional level to a more personal level; they made fewer negative attributions in situations of uncertainty. An increase in spontaneity and a decrease in formality helped overcome some difficulties in distance communication.[viii]
The results also highlight the superiority of in-person communication for researchers while they are trying to develop a common understanding of the research problem and possible approaches to solving it. Coordination of tasks, by contrast, appears to be hindered more by differing time zones than by specific attributes of the communication technology.
In the specific case of high technology firms in Bangalore, a history of business ties with the United States and the high presence of internationally experienced employees have fostered a mutual understanding and some degree of convergence of business practices with the United States. Given that the R&D labs in this study are geared towards international markets and many are owned by multinational companies, they must comply with Western standards of quality and professionalism. The efforts of researchers in Bangalore to prove their competence in a global context reinforce Western norms in local labs since the measures of competence are set by the existing labs in the parent or client company. A large part of the research community also shares a common educational background: doctoral or post-doctoral training at US research universities. Difficulties are more pronounced when collaboration involves communities that do not share this common history.
As we have also seen, research managers put a lot of effort into locally replicating a global corporate culture. The efficiency and quality of collaboration increases when people in different locations feel they share a common culture and identity. In some cases researchers felt a connection through a hi-tech identity, in others they drew on membership in a firm. Focusing on company culture can help bridge differences in national culture – perhaps even to the exclusion of local culture. More than one respondent suggested that anyone who had worked for one of the multinational or the larger outsourcing companies in Bangalore would refuse to go back to a traditional domestic firm.
Lester and Piore (2004) posit that innovative work can be divided into analysis (rational problem- solving) and interpretation (figuring things out through conversations). Analysis can be highly process driven and is – to a large extent – a question of coordination. Interpretation, on the other hand, evolves out of conversations and shared experiences. It describes the way a group develops a shared understanding of the environment and how its own work fits into this environment. According to Lester and Piore, interpretation requires shared time and space to develop. As experienced by some of the researchers in this study, one person cannot bring back all of the information gathered on a trip because some of it only makes sense in relation to the people who provided the input.
The interviews support Lester and Piore’s model. Respondents used phrases like “thrash things out” and “figure out an approach” to describe what they do in face-to-face communication. These are interpretive processes – a way of discovering what the problem is, rather than analytically solving a given problem with a predetermined method. Deciding on goals, priorities and methods requires judgment, negotiation between collaborators and repeated feedback loops – ideally instantaneous feedback. This work cannot be parceled out, nor can the process by which it is achieved be completely mapped out in advance.
Respondents’ descriptions of their use of communication technology provide some additional explanations why mediated communication is ill-suited to interpretation. When using e-mail, phone calls or teleconferencing, respondents felt they had to be more prepared, express things that might otherwise be taken for granted and anticipate difficulties earlier. They also said that processes took longer. There was less immediate feedback and spontaneous, simultaneous communication was more difficult. Mediated communication forces more structured exchanges (which can be beneficial in some circumstances). However, this is the exact opposite of what is needed for interpretive work.
Further research is needed to better understand the sources of trust and shared understanding in collaborative research. Despite the evidence above, it is not clear how much they depend on physical presence. As communication technologies evolve, they may provide more social and physical context, richer non-verbal feedback and an environment that allows deep, shared experiences. In fact, some new technologies have already improved on these specific factors. Alternatively, collaborators may develop new social forms online to replicate physical ones. Nevertheless, it seems that a handshake, a shared meal or brainstorming together in front of a blackboard will remain important for fuzzy, high-risk tasks, no matter the advances in technology.
Agrawal, A., Kapur, D. et al. (2006). Birds of a feather – better together? How co-ethnicity and co-location influence knowledge flow patterns. SSRN.
Bos, N., Olson, J., Gergle, D., Olson, G., & Wright, Z. (2002). Effects of four computer-mediated communications channels on trust development. SIGCHI conference on Human factors in computing systems: Changing our world, changing ourselves, Minneapolis, MN, ACM Press.
Cairncross, F. (1997). The death of distance: how the communications revolution will change our lives. Boston, MA: Harvard Business School Press.
Child, J. (2001). Trust–the fundamental bond in global collaboration. Organizational Dynamics, 29(4): 274-288.
Daft, R. L., & Lengel, R. H. (1986). Organizational information requirements, media richness and structural design. Management Science, 32(5): 554-571.
Davenport, S., Davies, J., & Grimes, C. (1998). Collaborative research programmes: building trust from difference. Technovation, 19(1): 31-40.
Dickey, M. H., Wasko, M. M., Chudoba, K. M., & Thatcher, J. B. (2006). Do you know what I know? A shared understandings perspective on text-based communication. Journal of Computer-Mediated Communication, 12(1), article 4. Retrieved 4 August 2008, http://jcmc.indiana.edu/vol12/issue1/dickey.html
Dodgson, M. M. D. (1993). Learning, Trust, and Technological Collaboration. Human Relations, 46(1): 77-95.
Freeman, C. (1991). Networks of innovators: A synthesis of research issues. Research Policy, 20(5): 499-514.
Gassmann, O., & Zedtwitz, M. von (1998). Organization of industrial R&D on a global scale. R&D Management 28(3): 147-161.
Hinds, P. J., & Weisband, S. (2003). Knowledge sharing and shared understanding in virtual teams. In C. Gibson & S. Cohen, (Eds.), Virtual teams that work: Creating conditions for effective virtual team effectiveness, pp. 21-36. San Francisco, CA: Jossey-Bass.
Jarvenpaa, S. L., & Leidner, D. E. (1998). Communication and trust in global virtual teams. Journal of Computer-Mediated Communication, 3(4). Retrieved 4 August 2008 from: http://jcmc.indiana.edu/vol3/issue4/jarvenpaa.html
Kaufmann, A., Lehner, P., & Todtling, F. (2003). Effects of the Internet on the spatial structure of innovation networks. Information Economics and Policy, 15(3): 402-424.
Kiesler, S., & Cummings, J. N. (2002). What do we know about proximity and distance in work groups? A legacy of research. In P. J. Hinds & S. Kiesler (Eds.), Distributed Work, pp. 57-80. Cambridge, MA: MIT Press.
Kim, K. J., & Bonk, C. J. (2002). Cross-cultural comparisons of online collaboration. Journal of Computer-Mediated Communication, 8(1). Retrieved 4 August 2008 from: http://jcmc.indiana.edu/vol8/issue1/kimandbonk.html
Lester, R. K., & Piore, M. J. (2004). Innovation–the missing dimension. Cambridge, MA: Harvard University Press.
Polzer, J. T., Crisp, C. B., Jarvenpaa, S. L., & Kim, J. W. (2006). Extending the faultline model to geographically dispersed teams: How colocated subgroups can impair group functioning. Academy of Management Journal, 49(4): 679-692.
Salter, A., & Gann, D. (2003). Sources of ideas for innovation in engineering design. Research Policy, 32(8): 1309-1324.
Sarker, S. (2005). “Knowledge transfer and collaboration in distributed U.S.-Thai teams.” Journal of
Computer-Mediated Communication, 10(4), article 15. Retrieved 4 August 2008 from: http://jcmc.indiana.edu/vol10/issue4/sarker.html
Saxenian, A. (1994). Regional advantage: culture and competition in Silicon Valley and Route 128. Cambridge, MA: Harvard University Press.
Storper, M. (1997). The regional world: territorial development in a global economy. New York: Guilford Press.
[i] Polanyi (1961:30), who is credited with coining the term “tacit knowledge”, provides the following illustration: A doctor, after witnessing an epileptic seizure with his students, says, “Gentlemen, you have seen a true epileptic seizure. I cannot tell you how to recognize it; you will learn this by more extensive experience.”
[ii] The choice of format was validated by the course that many interviews took. Direct questions usually elicited short and comparatively uninformative answers. An invitation to speak freely, such as, “I’m researching long-distance collaboration. Can you tell me about your experiences?” was far more likely to lead to animated discussions and detailed descriptions of work experiences. Once the conversation started to flow, interviewer guidelines were used to steer the exchanges.
[iii] In-person communication and face-to-face communication are used interchangeably here. Although the term ‘face-to-face’ is sometimes also used for video communication, this usage is not adopted here.
[iv] Several respondents had experimented with on-line “whiteboard” tools, and some used videoconferencing simply so that they could use a blackboard for figurative communication. However, neither method for these persons replaced the interactivity of people simultaneously talking and drawing on the same board.
[v] A “fresher” is a recent graduate with a Bachelor’s degree and no work experience.
[vi] On some campuses, entering the gates feels a little like crossing an international border. Not only are the physical environment and social norms inside and outside the campus different; security measures at the gates (including presentation of ID and checks of bags and electronic equipment) are reminiscent of border crossings.
[vii] Self-serve coffee machines may not seem relevant to a Western reader, but they stand out in the Indian context where most companies, including MNCs, employ peons to deliver tea and coffee to offices. The idea behind providing self-serve machines is to de-emphasize hierarchies and to provide a location for chance meetings, in other words, to create a water-cooler office environment.
[viii] This is not to say that face-to-face relationships must perforce result in trust. They can also lead to negative outcomes, e.g., when collaborators discover irreconcilable differences that remained hidden earlier. However, the experience of most respondents was one of increased comfort and trust, and that remains the focus of this analysis.