This paper is based on an ethnographic study of a company with widely distributed holdings within a volatile industry. The need for this distributed organization to respond as a unified whole to a rapidly changing environ«ment required it to confront the uncertainties and contingencies introduced by invisibility, distance, and fragmentation of information. In order to transcend the boundaries of time and space, the company coupled new technologies with new ways of organizing work. Groupware and remote meeting technologies provided a common virtual workscape. In this paper, we focus on issues generated by the integration of remote meeting technologies into local work practices. We compare face-to-face meetings with virtual meetings to understand how these new technologies impact the work of a distributed organization, and to uncover the new order that emerges around the use of these technologies.
Contents
Introduction
Managing Complexity: A Case Study
Methods
Mapping Complexity
Conclusions: Remote Work in Context
Introduction
Distributed organizations are not a new phenomenon. Historically, they include far-flung empires, religious institutions, and educational networks. Each of these organizations has had to develop a set of technologies and practices to coordinate activities. The growth of commercial firms around the turn of the century, for instance, precipitated a search for new theories and methods of management to achieve efficient coordination of large, multifunctional firms (Yates, 1989). Technologies such as the telephone and telegraph, combined with new genres of communication, facilitated the coordination of complex processes distributed across multiple locations.
Today's global corporations face an even greater set of complexities placing them on the "chaotic fringe," where workplace and technologies are undergoing dramatic transformations, with change in one often outstripping changes in the other. The increased pace of business means that processes are not only distributed, but that teamwork around these processes must happen synchronously or at least with increasingly rapid turnaround. Managers and executives find their presence required in multiple locations, and find that each location and task requires instant access to a different and unpredictable source of information. Inability to give input or call up data when needed stalls not only one meeting or one task, but also disrupts a complex and far-reaching web of contingencies. The problem, however, is also the solution: a variety of communication and information technologies that have recently come of age allow the members of distributed organizations to approximate a pan-corporate presence.
Electronic mail already serves global corporations as a broadcast medium and a mechanism for one-on-one communication across hemispheres and time zones. In addition, recent years have seen the emergence of "virtual spaces" conveying increasingly rich amounts of information, including gestures, tone of voice, and spatial conceptualizations of information. Groupware technologies serve as "organizational memories", transforming patterns of group work and hopefully galvanizing information sharing. Remote meeting technologies provide the means for distributed groups to work together around shared problems. These technologies are malleable enough to be adapted to a variety of group settings, while providing an integrated shared virtual environment to support information sharing across multiple tasks.
Developers of these technologies anticipate that they will profoundly alter the ways in which geographically distributed team members can work together as a single, purposeful unit. Their implementation, however, brings with it a new layer of complexity as people struggle to effectively integrate the right technologies into established work practices, and to modify those practices to take advantage of new technical opportunities.
Managing Complexity: A Case Study
Our paper describes a company trying to exploit that "chaotic fringe." We report on an ethnographic study of a holding company that is trying to couple advanced communication technologies with new ways of organizing itself. This company had diversified in the 1970s and 80s but early in the 90's decided to concentrate on its core competencies and reorganize its various subsidiaries. It created The Holding Company (THC) to manage this process (all names and identifying features have been changed). The holdings included nine geographically distributed business units in the insurance and financial services industries. A tenth business unit provided technical support to the business units and to THC headquarters. Senior staff saw the consolidation as an opportunity to create an organization that would eliminate traditional barriers of function and hierarchy in the interest of achieving better business results. They started to support this organization with state-of-the-art technologies.
At the time of our study, THC was organized around cross-functional projects and project teams. A variety of computing and communication technologies supported collaboration across geographically distributed groups. Work was centered around projects spanning multiple business units, and required heavy travel and careful coordination of over 200 corporate-wide projects. The tools that made this possible were familiar to most Associates (as permanent headquarters employees were called), such as standard word processing and spreadsheet applications, presentation tools, and graphics packages. However, they also included advanced data sharing and visual communication technologies, which were unfamiliar to most Associates prior to coming to THC, and which require not only new technical skills but also the development of new work practices.
Groupware at THC
Lotus Notes® was used both as a communication technology and as a shared repository or organizational memory. People regularly used e-mail for formal and informal communication, to exchange documents, and to keep people in the loop on projects or events peripheral to their own work. Shared databases formed repositories for project updates, financial information, legal decisions, etc. They also served as conduits between the business units and THC headquarters through an online quarterly reporting structure.
The various databases encompassed both routine and non-routine work, and formed the glue that held different groups together. They also became focal points for group discussion and formal meetings. For instance, THC-wide project reviews, which sometimes included remote members, worked from a THC-wide project database that listed status reports and links in related documents. Weekly IS staff meetings centered on short-term problem and long-term request databases, and quarterly reviews of the business units were based on online financial and planning databases. These common databases exemplified THC's policy of open access to information and embodied the CEO's desire to create a "virtual" organization.
Remote Meeting Support at THC
The primary remote meeting technologies at THC were LiveBoard®, an electronic whiteboard that allowed for shared viewing and manipulation of files between multiple sites, and PictureTel®, which projected video images between sites. At headquarters, the LiveBoard and PictureTel set-up were located in a large conference room, with an additional LiveBoard in a medium-sized conference room. Most of the business units had these technologies in place or were in the process of implementing them.
Group work activities supported by remote communication included status meetings, reviews of various sorts, and prototype demonstrations. They extended participation by enabling people to attend meetings which they would not have attended had extensive travel been required though their content related to their own work. Quarterly operations reviews serves as one example of extending participation at THC. The executive team traveled to the business units for the formal presentation of the quarterly report, which was broadcast live back to THC via PictureTel and LiveBoard. Associates could attend and participate in those parts of the review that touched on their own work with members of the business unit. Remote attendance of online quarterly reports and reviews substantially extended Associates' understanding of corporate operations.
Methods
We carried out ethnographic fieldwork for five months, collecting data through unstructured interviews, participant observation at THC headquarters and several business units, and review of documents and online materials. All 32 headquarter Associates, most of the seven temporary staff, and some business unit members were interviewed over the course of the four months. In addition, we observed a variety of meetings. Photographs and paper documents supplement and illustrate our fieldnotes. In addition, a total of 19 video tapes document a variety of geographically distributed meetings supported by remote technologies. Details can be found in Ruhleder Jordan, and Elmes (1996).
We analyzed these tapes using Video-Based Interaction Analysis in conjunction with ethnographic background information, according to a framework outlined in Jordan and Henderson (1995). This allows us to identify the ways in which these meetings differ from face-to-face encounters, and gain a better understanding of how remote workgroups are adapting to and exploiting these new technologies (Ruhleder and Jordan, 1997; Ruhleder and Jordan, 1999; Ruhleder and Jordan, 2001; Ruhleder, 2000).
Mapping Complexity
People know how to behave in face-to-face meetings. They know what kind of work they have to do beforehand, how punctual they have to be, what materials they have to bring with them to pass around. They know what kind of behavior is expected of them and can modify that behavior to suit the circumstances (if, say, a manager unexpectedly walks into a project meeting). They have a feel for the rhythm and pace of a meeting, and know what goals are realistic. When people hold remote meetings, many of their comfortable assumptions and work patterns are disrupted, and they have to re-think how they can best conduct the work of the meeting. Below, we will draw on our observations at THC to suggest some of the differences and similarities that we find between face-to-face and remote meetings across audio-only and audio-plus-video links.
Meeting Flow: Participating Across the Link
Events of any duration, including meetings, are organized in some way that makes sense to the participants. Some aspects of this organization are formally determined by protocols, agendas, or established frameworks. Others are seamless transitions from one mode into another, and may be indicated by movement of the participants, the introduction or manipulation of new objects, and changes in tone of voice or the nature of the topic up for discussion. Moving to a remote, distributed format changes the way the meeting flow is organized.
Distributed Beginnings and Endings
In face-to-face meetings, people know how to get meetings started, and how to close them off gracefully. They know the importance of the friendly greeting, casual banter, and office gossip that happens before the meeting formally begins - what we might call the "meeting dawn." These activities get people into the right frame of mind to work together, and blend social, personal, and work-related topics. Similarly, the "dusk" of a meeting consists of closing off the official part of a meeting, walking back toward the office together or picking up a related thread for one-on-one discussion. Again, these activities help people solidify personal and professional relationships, while also affording them the opportunity to test out and align their perspectives on the meeting.
We observed that such beginnings and endings, both formal and informal, became vastly more complicated when remote meeting technologies were introduced. Remote meetings rely on technologies that must be set up ahead of time. Files must be properly loaded, often well in advance, by support staff external to the group. Once video, audio and/or software links are established, people engage in a variety of real-time activities to test out the link and get settled. This involves adjusting volumes, moving furniture, and allocating control over pieces of the technical set-up.
Under these circumstances, easy cross-site bantering is not easily accomplished, and attempts at casual chitchat often turn into a major production, liberally laced with misunderstandings and non-sequiturs. At THC, participants on each side were usually busy with their own set-up, with local greetings and note shuffling, and often did not even realize right away that they were being addressed by their counterparts across the link.
The formal ending of the meeting was usually managed in conjunction with closing the technology link. While establishing the connection took some time as people fiddled with the configuration or needed to trouble-shoot problems, formal endings came literally at the touch of a button. At each site, informal meeting activity continued of course, with people analyzing what had been transacted, the demeanor of the other side, and their own behaviors. Yet while both sets of local participants continued in this way, there was no cross-site informal post-meeting activity - breaking the technology link breaks the social link as well. Yet work in other venues, such as doctor-patient interactions, strongly suggests that important business gets transacted at the end of a meeting. Going out the door, the patient's question - "oh, by the way" - may be the most important one asked (Fisher and Todd, 1983). What happens when such questions and comments no longer make it across the link?
Dawn and dusk, as liminal transition spaces, represent critical interaction arenas: it is here that locally shared perspectives are created and sustained. And it is here that the new technologies have already profoundly influenced the nature of group interactions. At THC, meetings became considerably more complex through the technical activity prerequisite to distributed meetings. Meeting dusk, on the other hand, came abruptly, without the possibility of amendments or renegotiation.
Maintaining Visibility Across the Link
In face-to-face meetings, who is present and who is not is apparent to all attendees. Both the audio-plus-data (same file or image viewed on a computer screen at each location) and audio-plus-video remote interactions we observed involved strategies for identifying or "making visible" the participants. In the audio-plus-video meetings this included making sure everyone was within camera range and, if necessary, mentioning who was present but outside of camera range. When this doesn't happen and people are caught off guard by who is present on the side of the link, they feel their trust is broken. In audio-plus-data meetings, this first required people to make sure that everyone had the same on-screen view of the data they were discussing. It also included listing the participants on either side, especially when the composition of the workgroup had changed, or when only a subset was attending a meeting. People also had to maintain that visibility, especially when one side spoke at length. People needed to signal when group members came and went, and periodically had to re-establish evidence of their continued attentiveness and availability. In audio-plus-data links, this verbal signaling sometimes became disruptive and burdensome.
Turn-Taking in Remote Meetings
In face-to-face meetings, people use a wide range of subtle (and not so subtle) cues, relying on gaze and gesture to signal that they are ready to speak, or that they aren't done speaking yet. Others can choose to grant them openings or retain control. In audio-only and video-plus-audio meetings at THC, we repeatedly observed both an unusual amount of speaker overlap and awkward pauses: one side often misjudged whether someone was just taking a breath, looking for a figure in their notes, or expecting a response. With video, the time delay in image projection introduced additional pauses as people tried to determine whether or not the other side was waiting for them to speak. Subtle cues were missed, remote participants sometimes failed to pay attention to the screen, and getting the attention of the other side often required breaking abruptly into an on-going side conversation (Ruhleder and Jordan, 1999; Ruhleder and Jordan, 2001). In other cases, small group meetings that could have been run informally in face-to-face situations required elaborate agreements when they were distributed over multiple sites. As we observed at THC, turn-taking protocols, meeting chairs, and explicit agendas facilitate these kinds of meetings on one level, but add another layer of complexity as people try to negotiate the technology, the meeting protocol, and the problem space.
Finally, our data contain numerous examples of literal breakdowns such as dropped lines or files that wouldn't load. Each of these caused a disruption to group work. When breakdowns occur in face-to-face settings - the overhead projector bulb burns out, or someone has forgotten a memo at their desk - all participants have equal knowledge about the breakdown and can negotiate over its repair (do we need the projector? the memo?). Breaks in the remote link often require elaborate actions to determine the nature of the problem, and outside intervention to fix it, with the remote side often unclear about what is happening and how long the repair will take.
Separate Dialogues: Side Conversations
In face-to-face interactions, side conversations are obvious, often disruptive, and actively discouraged in most settings. Even if participants step outside or go to a corner of the room, everyone knows who is involved and may be able to get some sense of what they are talking about. Local conversation also happen across remote links: one side can hear and/or see that the other side is engaged in conversation, while volume of voice or body orientation signal the other side that they may listen in, but not participate.
However, when only an audio-link is present, or when only some sites can be seen at any given time in multi-point meetings, we observed abundant opportunities for participants to engage in side conversations that are partially or completely hidden to meeting participants on the other side(s) of the link. For instance, people attending operations reviews via PictureTel muted the audio and engaged in running commentaries on the action. While participants on the other side can often tell when such side conversations happen, poor picture quality obscures details, such as facial expressions and subtle asides. In audio-only settings, we observed that one side would sometimes hold private discussions inaudible to the other side, even as that other side continued to address them.
In fact, when working in audio-only settings, group members in one physical location routinely engaged simultaneously in two levels of dialogue: verbal dialogue with the people on the other side, and non-verbal dialogue with their work mates who were physically co-present through gesture, sketching, or note-passing. This opens up the possibility that groups situated in one location may think they are establishing strong ties with a remote workgroup, when in actuality the cooperative or affirmative messages coming across the link hide the true feelings of participants behind this feature of the technology. Thus, while these technologies can provide powerful ways of connecting geographically distributed groups, they can also provide the resources for building and affirming shared views of local reality, including a shared view that insulates local participants from the perspectives of remote group members. In our observations, most of the time both processes are at work.
"Hot" and "Cool" Participation
In some meetings we observed, equal participation was expected of all group members, while others were characterized by "hot" and "cool" modes of participation. Participants on what we call the "hot" side, set the agenda, start and stop the meeting, drive the meeting and set the pace, while the "cool" side takes on a more passive role by responding or observing. At THC, structured, formal meetings, such as quarterly reviews of the business units, were characterized by this "hot"/"cool" split. The "hot" site was that of the business unit: THC's executive team would travel to each business unit for the formal presentation of the quarterly report. Other Associates "attended" the meeting on the "cool" side via PictureTel and LiveBoard, following along with the action.
The "cool side of the house" is by no means unimportant or passive. As a matter of fact, these events were extremely important to junior Associates, who perceived themselves as participating in an important organizational event. In fact, they were participating in two events: the broader event of the quarterly review and a local event, which included discussion of the review before, after, and during the event itself while the audio channel was muted. For the "cool" side participants (especially junior Associates), senior local commentators' remarks offered an opportunity to develop a shared perspective on good and bad moves, effective strategies, what counts as adequate preparation, etc. Again, however, these parallel dialogues - global and local - may work against the development of a shared global perspective, while also requiring individual participants to mentally manage two events at once.
Conclusions: Remote Work in Context
THC is one of many corporations trying to cope with multiple channels, overload, and fragmented expertise. Within its own industry, THC was already at the forefront in its application of information and communication technologies to support distributed group work. It had begun the move to a virtual corporation by adapting new technologies to existing organizational structures and incorporating them into established work practices. This adaptation raises serious issues around different forms of participation and productive joint work across the link. At the same time, it has already begun to reshape and influence how work is done and who is involved. Out of that adaptation will come novel applications and innovative practices, the first steps of which are outlined in Ruhleder et al. (1996).
A hundred years ago, new communication technologies fundamentally transformed the way business was conducted and reshaped corporate organizational structures. Today, the management of complex, distributed workplaces is facing a similar era of transformation. New positions are being created to manage the growing complexity of the technical infrastructure, and new organizational forms are being implemented to exploit these arrangements. Out of the moving edge of this chaos will emerge new order that will guide and shape work practices in the coming century. 
About the Authors
Karen Ruhleder is Assistant Research Professor in the Graduate School of Library and Information Science at the University of Illinois at Urbana-Champaign.
E-mail: ruhleder@uiuc.edu
Web: http://www.lis.uiuc.edu/~ruhleder/
Brigitte Jordan is a consulting corporate anthropologist at Data Design and the Xerox Palo Alto Research Center.
E-mail: jordan@akamail.com
Acknowledgments
Our research was sponsored by the Xerox Palo Alto Research Center, the Institute for Research on Learning, THC, and NSF grant number 9712421 and by the Graduate School of Library and Information Science of the University of Illinois, Urbana-Champaign.
References
Sue Fisher and Alexandra Todd, 1983. The Social Organzation of Doctor-Patient Communication. Washington, D.C.: Center for Applied Linguistics.
Brigitte Jordan and Austin Henderson, 1995. "Interaction Analysis: Foundations and Practice," Journal of the Learning Sciences, volume 4, number 1, pp. 39-103.
Karen Ruhleder, 2000. "The Virtual Ethnographer: Fieldwork in Distributed Electronic Environments," Field Methods, volume 12, number 1, pp. 3-17, and at
http://alexia.lis.uiuc.edu/~ruhleder/e-fieldwork.pdf
Karen Ruhleder and Brigitte Jordan, 2001. "Co-Constructing Non-Mutual Realities: Delay-Generated Trouble in Distributed Interaction," International Journal of Computer Supported Cooperative Work, volume 10, number 1, pp. 113-138, and at
http://alexia.lis.uiuc.edu/~ruhleder/jcscw_delay.pdf
Karen Ruhleder and Brigitte Jordan, 1999. "Meaning-Making Across Remote Sites: How Delays in Transmission Affect Interaction," In: Susanne Bodker, Morten Kyng, and Kjeld Schmidt (editors). Proceedings of the Sixth European Conference on Computer-Supported Cooperative Work, 12-16 September 1999, Copenhagen, Denmark. Boston: Kluwer, pp. 411-427, and at
http://alexia.lis.uiuc.edu/~ruhleder/ecscw_delay.pdf
Karen Ruhleder and Brigitte Jordan. 1997. "Capturing Complex, Distributed Activities: Video-Based Interaction Analysis as a Component of Workplace Ethnography," In: Allen S. Lee, Jonathan Liebenau, and
Janice I. De Gross. Information Systems and Qualitative Research. London: Chapman and Hall, and at
http://alexia.lis.uiuc.edu/%7Eruhleder/publications/97.IFIPWG82.html
Karen Ruhleder, Brigitte Jordan, and Michael B. Elmes, 1996. "Wiring 'the New Organization': Integrating Collaborative Technologies and Team-Based Work," Presented at the 1996 Academy of Management, and at
http://alexia.lis.uiuc.edu/~ruhleder/publications/96.academy.html
JoAnne Yates, 1989. Control through Communication. Baltimore: Johns Hopkins University Press.
Editorial history
Paper received 11 April 2001; accepted 20 April 2001.
Copyright ©2001, First Monday
Managing Complex, Distributed Environments: Remote Meeting Technologies at the Chaotic Fringe by Karen Ruhleder and Brigitte Jordan
First Monday, volume 6, number 5, May 2001
URL: http://firstmonday.org/issues/issue6_5/ruhleder/index.html
