The Evolution of Artifacts in Cooperative Work: Constructing Meaning Through Activity Marlin M. Cluts First National Bank of Rochelle 340 May Mart Drive, Rochelle, Illinois 61068, USA

[email protected] ABSTRACT A two year case study of cooperative work was inspired by the installation of CSCW software in a community bank. The framework for the research was developed by combining activity theory and the principles of communities of practice. This framework provided a useful model and insight into the evolution of artifact meaning, sharing, and credibility. In essence, users needed to experience results within the activity system and social structure of the community to establish meaning and credibility. The study applied the theory and language of activity theory and communities of practice to make sense of case phenomena and provide a richer context for understanding traditional principles.

Categories and Subject Descriptors

Keywords: CSCW applications, activity theory, communities of practice, work ethnography, artifacts, emergence, artifact meaning, cooperative work. In the design of computer-supported cooperative work (CSCW) applications an understanding of how cooperative work transpires is critical. This study was an attempt to further understand how cooperative work is accomplished through the evolution of artifacts. The primary research questions were: How are artifacts shared?

3.

How are artifacts given meaning?

4.

How is the perceived use or credibility of artifacts determined?

How do workers appropriate technology artifacts over time?

7.

How are promises and mutual commitments made?

In keeping with the activity theory model, the activity checklist as developed by Kaptelinin, Nardi, and Macaulay [22] served as the primary source of interview and observation dimensions. Though the framework of activity theory served as the underlying model, concepts derived from communities of practice research, primarily Wenger [39], were added as further defining elements within the activity theory model.

1. INTRODUCTION

2.

6.

Two perspectives make the study unique. First, the research was performed by a participant observer as a complete member of the organization. Second, the study was done using the framework of activity theory with the concepts of communities of practice integrated into the model. This combined framework yielded theory and language to describe results in a richer context and provided new meaning to standard development principles.

General Terms: Design, human factors, management, theory.

How do artifacts evolve? How are they created?

How are the relationships among workers affected by the use of technology?

The impetus for the study was the installation of a major CSCW software package at our community bank. This CSCW software, called Transcend, was implemented to facilitate sharing of information and customer contacts among the customer service team. The software included central information file and contact management, sales management, and profitability analysis capabilities, all of which were new functions for the organization. Management believed the installation of software with these capabilities would have a profound effect on the organization, its culture, and the way business was conducted. The project involved a two-year case study on the uptake of the new technology.

H.4.1 [Information Systems Applications]: Office Automation – groupware; J.4 [Computer Applications]: Social and Behavioral Sciences – psychology, sociology; K.4.3 [Computers and Society]: Organizational Impacts – computer supported collaborative work; K.6.2 [Management of Computing and Information Systems]: Installation management – pricing and resource allocation.

1.

5.

Additional comments should be made regarding the research and the researcher. According to Angrosino and Mays de Perez [3] a complete member participant observer would have the best chance of experiencing the types of tacit knowledge sought by an ethnography. However, with a complete member, issues of bias and research ethics become more prominent. In this case I was the ultimate complete member, a senior manager with over 25 years with the organization. I was able to subjectively live the experience, but I also had to go to great lengths to assure the quality of the research and mitigate ethical concerns. Integral parts of the study included audit trails, consent forms, institutional review board requirements, use of pseudonyms, availability of third party contacts for participants, and painstaking attention to avoiding potential issues regarding privacy and discomfort for participants. In addition, in keeping with the postpositivist or naturalistic research paradigm

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. GROUP’03, November 9-12, 2003, Sanibel Island, Florida, USA. Copyright 2003 ACM 1-58113-693-5/03/0011…$5.00.

144

Activity theory accommodates many of the shortcomings as well as the contributions noted in the literature above. Representing a cultural anthropological approach based on Soviet cultural-historical psychology, activity theory is an overarching framework that nicely encompasses the aspects of dynamics, emergence, and communities of practice, and it provides a theoretical foundation within which to include such factors. It provides a bounded context that the social interaction studies desire. It is grounded in culture to satisfy the concerns of the organizational studies. It accommodates artifacts such as boundary objects and emphasizes the transformation that activity systems undergo, thereby incorporating aspects of the anthropologically oriented studies. In essence, activity theory can be seen as the central piece of the puzzle, linking other pieces together by providing a solution to shortcomings and a common ground for others.

[34], an attempt was made to involve participants in the analysis, shaping, and production of the research as empowered participants [13]. Accordingly, a group of participants was involved in an attempt at Heron and Reason’s [19] cooperative inquiry model. (This group became known as the Transcend group and also eventually became responsible for facilitation of the use of the software.) In summary, the combination of research methods and safeguards produced a study that passed the scrutiny of a dissertation committee and institution review board. However, as in any qualitative research, you, the reader must assess the validity and reliability of the findings. Additional information regarding methodology is available on request. The project will be described in the following sequence. First, a brief summary of prior research will be provided leading to the choice of activity theory. Second, capsule summaries of the principles of activity theory and communities of practice will be advanced. Third, the activity theory framework will be applied to a brief description of the CSCW implementation in the bank. An analysis will then be presented using activity theory and communities of practice concepts to explain the evolution in artifacts observed. Finally, implications for developers and managers will be explored in light of the underlying theories and in the context of the language of activity theory and communities of practice.

Activity theory based studies [5, 7, 12, 15, 25, 33] make use of a bounded unit of study defined as an activity or activity system rather than dealing in social systems that can be too large to be useful for context. They also welcome contradictions as a necessary and natural part of the system, thereby enabling change and development [12], and they provide a consistent framework within which to compare to other activity theory based studies. Since there are a number of excellent sources on activity theory [15, 22, 31], a complete explanation will not be attempted in this paper. The reader knowledgeable in the principles of activity theory and communities of practice may wish to skip to section 5. For those less familiar, the basic principles most applicable to this study are discussed below.

2. LITERATURE SUMMARY A broad spectrum of research is applicable to the understanding of cooperative work. Rather than listing relevant research in detail, a summary of the types of research applicable and the major categories of contributions will be presented along with a number of example articles.

3. ACTIVITY THEORY Activity theory was derived from the work of Lev Vygotsky and the cultural-historical school of Soviet psychology, which maintained that cultures are comprised of shared social meanings brought into existence by activity. The purpose of activity theory, then, is to offer a set of conceptual tools useful in analysis and to “understand the coupling of consciousness with activity” [35, p. 51]. An activity, itself, is the “minimal meaningful context for understanding individual actions” [24, p. 28]. The classic example is provided by Leontev [27]. Consider primitive hunters in search of game. They separate into two groups, one to frighten the game toward the others by beating the bushes and the second group to catch the game for food and clothing. The actions of the bush beaters themselves are not understandable unless considered part of the activity of hunting.

Studies in social interaction yielded useful information regarding individual interaction with other individuals [9, 14, 21, 30]. However, these studies often referred to a need to understand context, which could be an entire social system, including the structure of an organization. Studies conducted from an organizational perspective considered the impact of organization on software implementation and vice versa [6, 10, 11, 17, 23, 32, 37, 38]. A few of these studies noted successful implementations while several others noted unsuccessful implementations. Many considered organizations as structures imposed workers, while studies such as Suchman’s [36] demonstrated clearly that the actual accomplishment of work is rarely a reflection of the structure imposed but rather a matter of negotiated meanings and the everyday practice of the people involved. Many of the organizational studies also cited a need to understand broader context such as culture.

Activities have a basic structure as depicted in the chart below:

Other more anthropologically oriented studies emphasized symbols and culture. Examples include research on artifacts and the concept of distributed cognition [1, 8, 20, 29]. These studies discussed such topics as boundary objects carrying a history of the relationships involved. Distributed cognition emphasized the transformations that external structures undergo when operated on by individuals and artifacts. Figure 1. Structure of an activity.

Additional literature on dynamics and emergence investigated deeply the interaction of people and systems as emergent and requiring negotiating meaning among participants [2, 16, 18, 26, 28, 36, 39]. Some of these studies also used activity theory as a basis.

145

tools and software to assist in mediating the relationships. Bannon and Boedker [4] describe this concept as praxis. An example involves the carpenter who uses a saw and hammer to produce a house out of wood. However, the carpenter is not alone in the world. He works with other carpenters who divide work between them. The ways and methods of doing work are grounded in tradition and shared by the group of carpenters. This shared collective ensemble of ways, methods, and traditions is called praxis. When an apprentice is being trained as a carpenter, he or she is allowed to share this praxis. At the same time, each person who shares a praxis may change it as well by coming up with new ways of doing things. It is this praxis that allows us to talk about more than just individual skills and knowledge rather than just a generic human being. It also allows us to talk about the appropriateness of a certain tool for a certain praxis and to account for the development of new tools over time.

The basic definition of an activity is a form of doing directed at an object. One activity is differentiated from another activity by its object, and transforming the object to the outcome is the motive for the existence of an activity. Objects can be material things or something intangible, such as a plan or an idea, as long as it can be shared for manipulation [24]. Mediation between the subject and object is carried out by introducing a third term that carries with it the history of the relationship. Therefore, the relationship between the subject and object is mediated by a tool embodying the historical development of the relationship between subject and object. Other relationships are mediated by other types of artifacts. The relationship between subject and community is mediated by rules and the relationship between community and object is mediated by division of labor. These mediators, likewise, carry the history of these relationships. Note that the chart implies that a tool between subject and object may not necessarily be available to the rest of the community. Therefore, the history of that relationship may be invisible to others in the community. Each person in the community may have a different view of the object that may be apparent in the tools each is using [24].

4. COMMUNITIES OF PRACTICE Similar in concept to the idea of praxis is the idea of practice. Wenger [39] notes that the concept of practice connotes doing or activity, but it isn’t just this action in itself. It is doing or activity in a social and historical context that gives meaning to what we do.

Activity theory also includes a hierarchy concept as depicted in Figure 2.

Wenger, though not citing activity theory, also addresses the idea of artifacts. Based on his study of insurance claims processors, Wenger [39] proposes, “The communicative ability of an artifact depends on how the work of negotiating meaning is distributed between reification and participation” (p. 64). This takes some additional explanation. First, communicative ability has a great deal to do with meaning and Wenger [39] says that meaning “exists neither in us, nor in the world, but in the dynamic relation of living in the world” (p. 54). “Participation” is defined as “to take part in or share in some process or activity.” Reification means “to treat [an abstraction] as substantially existing or as a concrete material object…the process of giving form to our experience by producing objects that congeal this experience into ‘thingness’” (p. 58). Examples would be writing down a law, composing a memo with new rules or procedures, or developing a tool. In essence, the community of practice engages in the negotiation of meaning through the processes of participation and reification. The community of practice is the environment in which these processes can occur.

Figure 2. Dynamic structure of an activity. The highest level is the activity itself. An activity is composed of various actions that are related to each other by belonging to the same activity. Actions are then composed of various operations which are somewhat automatic tasks. Actions take place due to a conscious decision while operations are automatic. Activities are influenced by motives while actions are influenced by goals and operations by conditions. This hierarchy becomes useful in analyzing behavior. People experience various levels of frustration depending on whether conditions, goals, or motives change.

Note how nicely Wenger’s theory fits with activity theory. The concept of a community of practice is quite similar to the activity theory concept of praxis. The fact that meaning for Wenger is in the “dynamic of living in the world” is entirely consistent with the activity theory concept of shared social meaning brought into existence by activity. The Wenger concept of doing or activity in a social and historical context is also quite consistent with activity theory’s cultural historical roots. What Wenger and other community of practice scholars provide could be an expansion on the concept of community in activity theory that yields additional insight into the workings and importance of community. By defining characteristics of a community of practice and constellations of practice additional dimensions may be explored. Therefore, the concepts of communities of practice were included under the activity theory framework as a part of the community analysis in this project.

The dimensions of possible frustration noted in the hierarchy are not necessarily seen as undesirable. Activity theory allows for dynamic development, and contradictions are seen as important to the process. Contradictions can fuel the frustration in the form of problems, ruptures, breakdowns, incompatibility, or misfits. Contradictions, though agonizing, are an important source of development. The emphasis on community is an important part of activity theory as noted in Figure 1. The subject and community have a mediated relationship as well as community and object. These relationships are critical to the context of the activity and to the development of

146

5. THE CASE In this section the software will be described, a synopsis of the adoption will be presented, and samples of activity systems will be discussed.

5.1 The Software Bank management had been struggling with the concept of developing a sales culture. Various systems had been tested, including manual referral systems, meetings, and other manual tracking systems, most of which were dismissed as being cumbersome, extra work, and not worth the time. The executive committee decided to explore electronic solutions and ultimately selected software called Transcend. The software was designed to support cooperation among different departments in sharing customer data, and as such, it was a data aggregation system to pull information from different areas together. It was also a customer contact management system. The contact management system allowed ticklers to be set for calls to customers, provided data fields to keep notes of comments or commitments made, gathered pertinent information about the customer, and implemented a system to make referrals to other departments of the bank for services needed. It was a sales and referral tracking system for management as well as a marketing central information file (MCIF) that would allow specific marketing campaigns, such as sending a letter to all homeowners soliciting a home equity line of credit. It was truly a major attempt at integrating a number of data sources in a way that could be useful to support better cooperation among bank personnel to provide improved customer service.

Figure 3. Loan Activity System The object is the customer and the desired outcome is a satisfied customer. The loan officer has a number of tools available to mediate the relationship with the customer, including his own records of conversations, loan files and applications, voice mail from the customer, common language, his own planner or Palm Pilot, and many others. It is fairly easy to envision these tools as containing a history of the relationship between the customer and the subject or officer. Figure 3 also shows the community mediating the relationship between the officer and the customer. In this case, the community is the loan department and other loan officers who will see the client in the primary officer’s absence.

5.2 Software Adoption Synopsis

Another activity system within the bank was that of the personal bankers, who handled deposit relationships for customers. The activity system would be quite similar to that of the loan activity system with a personal banker as subject, the customer as the object, and the personal banker department as the community. It differed by the availability of different tools and different rules mediating subject and community.

After a simple half day training session, the software was launched in August of 2000 with usage quickly building to 124 transactions per month in October. Then in December of 2000 the bank started work on a large acquisition project and usage dropped dramatically to 33 transactions, presumably due to lack of management attention. The software languished at around 20 transactions per month until after the acquisition was closed in May of 2001.

Closely related to the number of activity systems in the bank were the number of communities. It was evident that bank employees perceived they were part of various communities within the bank, although they certainly didn’t call them communities. Most times they were aligned along departmental lines. The numerous pleas coaxing everyone to use Transcend were usually in the context of everyone in a department. Priscilla, a personal banker, in lamenting others not using Transcend, refers to others in her department. Likewise, the loan officers, when reporting the value they discovered by finding customer notes in Transcend, refer to notes by other loan officers or themselves. Although information was available to all departments, users typically concerned themselves only with information relative to their own department or group.

In an attempt to resurrect the project, the Transcend group was formed in August of 2001 to foster activity on the software. Over the next nine months the activity steadily ramped up to over 250 transactions per month. Then the unthinkable happened. A new release was installed that caused intermittent but frequent network outages. Usage dropped back to 137 transactions per month accompanied by complaints that the new software, which had finally been accepted, was no longer reliable. However, by October of 2002 the problem was fixed and in a tribute to human flexibility and resilience, activity climbed back up to over 300 transactions per month in spring of 2003.

6. ANALYSIS

But the story was not in the numbers. It was the interactions among people and artifacts that were of interest in this study. In the following sections those interactions will be described.

In this section the research questions will be revisited in the light of the observations in the study.

5.3 Bank Activity Systems

In analyzing artifact evolution, the goal is insight into the issues of artifact creation, evolution, and appropriation by workers over time.

6.1 Artifact Evolution

As contemplated by activity theory there were a number of activity systems within the bank. An example follows.

Activity theory, by its emphasis on context, insists that we study the whole activity system, not just the artifact. This provides a

147

(object). The developers of Transcend had to take into account the other tools available in the design of their software tool. They integrated the data from the bank core systems, added scheduling and tickler functions to replace the personal planner function, and provided places for file notes to replace the loan file. They even attempted to integrate e-mail, although this function was not operative for the bank. This integration is often an evolution impetus, as it forces us to give up former tools for better ones that give us access to more functionality. If we don’t perceive the new tools as better, there is a built-in conflict that should fuel more development.

framework to discuss the artifacts and yields clues as to how they evolve. Activity theory describes three kinds of artifacts. Tools mediate the relationship between subject and object while rules mediate the relationship between subject and community. Division of labor mediates the relationship between community and object. In essence, the role of artifacts is to mediate these relationships, and as such, they carry a history of that relationship embedded in them. In fact, they are created from the need to mediate and preserve the system. An example from this research was the creation of a memo to the personal bankers defining the types of information that should be entered in Transcend. This was an explicit rule that was documented. It was created out of a need to communicate with all of the personal bankers and assure standardization in the system. It was a rule mediating the relationship of each subject, or personal banker, and the community of personal bankers so that all would have an understanding of their obligations and expectations regarding Transcend.

A special case of conflict arose from the system’s lack of accommodation for a particular data item. Our example was the lack of a place to enter officer organizational event attendance on Transcend. A kluge was developed whereby a dummy customer was created, called “Org”, and officers posted these events as meetings under this dummy customer. This in itself was an artifact, developed to get around the system’s shortcomings. It is of special note because these kluges tend to get overlooked by developers as they become a congealed part of the system. By definition, they were created due to the lack of system capability. Users may create these workarounds without ever telling developers, thereby creating a use hidden from the developer.

Another example was of a more tacit nature. It was the idea of, “Check Transcend first.” One of the personal bankers was upset when she was called at home about information for a customer. She had entered the data in Transcend and lamented that the caller could have avoided disturbing her at home if the caller had only checked Transcend. “Check Transcend first” was a rule mediating the relationship between subject and community. This one arose from the need within the community for mutual respect for each other’s time. It enabled the continued function of the activity system.

Another special kind of artifact is a boundary object. Mambrey and Robinson [29] define “boundary object” as an object that needs to satisfy the information requirements of intersecting social worlds or activity systems or, in our case, departments or communities. Boundary objects need to be elastic enough to satisfy the needs and constraints of both parties employing them and yet maintain a common identity across sites. This concept identifies the difficulty in installing an artifact such as Transcend, which was intended to cross departmental boundaries. The differing needs of the communities using boundary objects are likely to trigger numerous conflicts. In our case the different needs of the communities triggered substantial debate at the Transcend group over standards.

Activity theory, in its attention to development, also lends additional insight to how artifacts evolve. Conflicts and contradictions are considered desirable for fueling development. The activity theory hierarchy was presented as a framework for these conflicts. At the highest level, activity is influenced by motives. Activities are accomplished by a series of actions that are influenced by goals, whereas actions are composed of a series of relatively automatic tasks called operations. Operations are influenced by conditions.

There were a number of artifacts that evolved during the life of this research. The customer notes on Transcend posted by officers and personal bankers were examples. The nature of the data to post evolved. The whole attitude of sharing customer data changed. Transcend itself was a very large artifact, intended to displace others. Numerous language terms, notes, memos, shared ideas, lists, tools, etc., evolved as artifacts embodying a history of the relationship between subject and object or subject and community.

The level at which a conflict occurs has some bearing on the result. For example, the relatively automatic operation of entering customer data into Transcend could be disrupted by the condition of a network outage. The most likely response is for the subject to simply wait until later. However, the lack of a software tool to support a goal, such as recording credit life sales, is likely to trigger a conscious action, which could be the creation of another artifact. In our case, it was a report from another system. The frustration of a motive would lead to a much more drastic consequence. For example, if the bank decided it no longer wanted to deal with customers and it would make its profits simply by buying and selling securities, the whole motive in the activity system (taking better care of customers to increase sales) no longer exists. The entire activity system would be discarded, thereby destroying the need for the software artifact.

So how did artifacts evolve and how were they appropriated over time? Using the activity theory framework, artifacts were created to mediate relationships between subject, object, and community in the form of tools, rules, and division of labor. They often evolved out of conflict in the activity system where a disruption forced the actors to create a new artifact. In this study, integration led to adoption of a new, more powerful tool (Transcend). Conflict in the form of incompatibility of the software tool for some actions also drove actors to create other artifacts. In essence, the evolution and adoption was a natural function of the activity system.

Another dimension in evolution should be considered. Activity theory forces us to study the whole activity and not just the artifact. This means we also need to study other artifacts that are used in mediating the relationships. We cannot simply look at the proposed software tool. Consider the loan department activity system. Loan officers use the bank core data system, their own personal planner, the loan file, voice mail, and e-mail, to name just a few tools mediating the relationship between themselves and the customer

6.2 Community In this section, the principles of artifact sharing are addressed, including the relationships among workers and mutual commitments. The activity theory concept of praxis is applicable as well as those of communities of practice.

148

methods of engaging that personal bankers had adopted prior to Transcend. It was understood that this information needed to be shared because the personal bankers served each other’s customers. A new method of sharing was available through Transcend. Although it did not become an automatic operation early on, the new artifact rule of “check Transcend first” will hopefully soon be adopted by the community. This example demonstrates elements of sharing, relationships, and commitment. The information must be shared. The relationship among personal bankers is at stake and their mutual commitment to “check Transcend first” is an evolving commitment to each other.

In this research, a constellation of practices was studied, defined as the whole bank. A few of the practices within the bank were the loan department, personal bankers, and the investment department. The separation into different communities was roughly along department lines, but there were hints of broader communities of practice, such as the officers as a group and the Transcend group itself. Wenger [39] provides a list of guidelines or indicators that a community of practice exists. A few of those indicators as observed in this study were as follows: 1. sustained mutual relationships – harmonious or conflictual

Likewise, the community of the loan officers made a mutual commitment to enter their loan commitments and other loan customer notes into Transcend so each could handle the customer in the other’s absence. This was for the common purpose of serving the customer and enhancing each other’s competence in the customer’s eyes. Again, the idea of sharing was demonstrated through the sharing of information. Relationships were entailed through the desire to preserve each other’s image of competence. Mutual commitment was in the promise to enter the information and allow each other to perform at a perceived higher competence level from the customer’s perspective.

2. absence of introductory preambles, as if conversations and interactions were merely the continuation of an ongoing process 3. very quick setup of a problem to be discussed 4. knowing what others know, what they can do, and how they can contribute to an enterprise 5. specific tools, representations, and other artifacts 6. local lore, shared stories, inside jokes, knowing laughter, jargon and shortcuts to communication as well as the ease of producing new ones. (p. 125)

On the other hand, there were few examples of meaningful sharing across department or community lines. Although the information was visible and the tool of Transcend was available to the constellation, examples of meaningful sharing across communities were infrequent.

Wenger also provides guidelines for a constellation of practices: 1. sharing historical roots 2. having related enterprises 3. serving a cause or belonging to an institution

So how were artifacts shared? How were relationships among workers affected? How were mutual commitment made? Sharing of artifacts was more likely to occur across a community of practice. Within the community, relationships were tighter and communications were easier. Mutual commitments were much more likely within a community than across the constellation. In essence, the definition of the community as opposed to the constellation had a great deal to do with the ease in sharing, relationships, and mutual commitments. In analyzing sharing of artifacts, relationships, and commitments, one must understand the definition of the various communities within a constellation.

4. facing similar conditions 5. having members in common 6. sharing artifacts 7. having geographical relations of proximity or interaction 8. having overlapping styles or discourses 9. competing for the same resources. ( p. 127) Given these guidelines, it was easy to conclude that the bank was a constellation of practices. The loan department was a good example of a community within this constellation with its specific tools in common, such as loan files, shared language regarding loans, and shared ways of engaging. All the attributes were evident, such as a quick setup for discussing a problem.

6.3 Meaning and Credibility In looking for meaning in artifacts, Wenger’s [39] proposal was presented: “The communicative ability of an artifact depends on how the work of negotiating meaning is distributed between reification and participation” (p. 64). Participation involves the experience of the participant, usually within the community of practice. Reification involves the development of abstract ideas or experiences into “thingness.” Wenger implies both are needed. Negotiation of meaning is distributed between both. Note the emphasis on negotiation, which was also considered by Suchman [36] and Gerson and Star [18] as they determined that accomplishment of office work was more the process of local negotiation than written procedures.

So what bearing does this difference between the communities and the constellation of communities have on the questions of sharing, relationships, and mutual commitments? First, a community is much more likely to have specific, shared artifacts to support their activity system. As an example, the loan department had the loan files that were not available to other departments. Second, the ease of engagement in a community will make the likelihood of sharing greater. By definition, the relationships are more “sustained and mutual.” Mutual commitments are more likely to be made in a community of practice due to the mutual relationships. An example from the study should help to clarify.

Our bank had numerous examples of negotiation of meaning. The Transcend group itself often negotiated a kluge or treatment of data to the satisfaction of those involved. The personal bankers negotiated among themselves a better place to enter their “phonecall ticklers”. According to Wenger [39], this negotiation necessarily involved participation and resulted in some reification. The fact that the personal bankers participated and shared reasons for designating another place to record their “phone ticklers” gave meaning to their

The personal bankers constituted a readily observable community of practice. Priscilla Hunter advocated the tacit rule of “check Transcend first.” She was lamenting that, despite the fact that she put information in Transcend for the benefits of her fellow personal bankers, one of them called her at home to ask for the information. The fact that she was called at home demonstrated one of the

149

solution. The idea of recording “phone ticklers” as “note ticklers” was the reification. It was also reified in a memo. The Transcend group’s negotiations involved participation by various stakeholders. Participation in the negotiations led to an eventual agreed solution that had more personal meaning as opposed to simply receiving a memo to put “phone ticklers” in the system as “note ticklers.” Negotiating the meaning of artifacts involved the processes of participation and reification.

effect on credibility. Note a comment regarding a model user: “I think, from my perspective, when Jack Gorman accepted the product as a valuable product…if he’s actually doing this, well, then maybe I don’t have an excuse left.” (Connie Carter, August 29, 2002)

However, even the idea of just participating or using the software was extremely powerful. The battle cry from the Transcend group was, “We need to get people to use it at all costs.” And, in fact, in reviewing the comments of various users, most said once they used it, they could see its value and meaning. The very basis of activity theory is that meaning is in the participation within an activity, and this principle was paramount in this project.

In summary, artifacts gain meaning and credibility from several dimensions. First, meaning can be negotiated through participation and reification. Second, the very act of participation or using the software is crucial. In this study, once people started using it and experiencing the results, significantly more meaning and credibility were attached. Little credibility was derived from the realization that other banks were using the software successfully. However, the community itself attached significant value to the adoption by others within the community or constellation, especially key people.

The community may have informal leaders or those who have commanded special respect of others. When members of the community see these leaders adopting the artifact, more credibility is attached.

The idea that meaning is in doing is not new. It is prevalent in education in many forms, some as expansive as constructivism, some as simple as the idea of on-the-job training. Learning is in the doing, and likewise, meaning is in the doing. In this project the software was so complex that it was very difficult to give a two-hour training session and have people understand the capabilities. However, when users logged on and experimented and experienced the results within the social structure of the community of practice, they then experienced meaning. It was simple enough for Jack Gorman to put the information in Transcend as trained in the initial implementation, but the data artifact did not take on significant meaning until Dorine told him she took care of his customer and found his notes and was able to do just what he had contemplated. The data now had meaning for both Dorine and Jack. Dorine found useful information allowing her to take care of the customer expeditiously, and Jack had assurances that his initial work for the customer was useful and implemented.

The analysis above has situated some of the phenomena encountered in this study within the key elements of activity theory and communities of practice. This information and the language of activity theory and communities of practice can then be used to describe implications for developers and implementation managers. The implications themselves are not new, but the language of activity theory and communities of practice provide a richer context for understanding and making sense of development principles.

7. IMPLICATIONS The implications are generic to all software developers and not specifically to Transcend, Inc., and, to their credit, most of the implications have already been addressed by Transcend, Inc. The company has been extremely cooperative in providing time from their key employees for interviews. Their interest in this topic is evident by their comparatively well-designed software and the history of resolving conflicts with future releases.

Meaning is also related to credibility. Obviously, there is little meaning derived from an artifact with questionable credibility. Understandably, the perceived credibility of artifacts in this study seemed to depend upon usage more than anything else. When asked if knowing other banks were using certain features had anything to do with his perception of value, Jack Gorman flatly said, “No.” His thoughts were that only when he uses a feature and sees it work does he attach credibility and meaning.

Many of the implications below involve standard practice for developers, but the language of activity theory and communities of practice may provide additional meaning or basis for consideration. 1. Software design should consider the context in which the software will be used. Activity theory provides a framework designed to force us to study the whole context, but it still provides boundaries by defining the activity or activity system.

The major focus of the Transcend group was to encourage people to use the software and, hopefully, discover meaning. People kept claiming that when everyone else uses it, then they will feel like they should. But they couldn’t readily ascertain if everyone was using Transcend unless they spent valuable time navigating numerous options to create a report, or they performed an inquiry on a customer for whom data was previously entered. In fact, Ackerman and Starr [2] developed this issue in their study. They noted that people desire to know what other people around them are doing. They further posited that people use other people’s visible activity to frame their own goals. Just knowing others are active can have a dramatic effect. For example, if “Joe” can easily determine that no one is working on the “acquisition project,” he is less likely to work on it himself. However, if the computer system were to somehow reveal that everyone is active on the project, “Joe” would be much more likely to apply himself to the project. This led Ackerman and Starr to suggest social activity indicators for CSCW software.

2. All artifacts within the activity system must be considered for integration. Activity theory provides a framework under which these artifacts can be given meaning. Transcend, Inc., has done this quite well with the level of integration provided. However, even further integration would be desirable, such as the inclusion of voice mail, internal e-mail (rather than relying on external servers), personal planning, and task management. 3. Software should be written to allow the user to engage at the operations (as defined by activity theory) or automatic level as soon as possible while providing flexible alternatives for those cases where actions must be changed to satisfy new goals. For example, it is helpful to have available user-defined fields for those actions that were not contemplated in the original design. Transcend has done this. 4. Conflicts and contradictions are sources for the continuing development of not only the software but also the activity system that uses it. Some contradictions are built into any activity system,

The concept of who uses the artifact within the community of practice or, for that matter, the constellation, seemed to have an

150

the receipt of a referral similar to the common provision in e-mail. This suggestion could be coupled with the screen recommended above showing who is online and active.

such as the time required away from customers to input information versus the need to get to customers as soon as possible. These contradictions spur development, as do the specific shortcomings of the software artifact in mediating the elements of the activity system. Developers should look at these contradictions and conflicts positively and manage them carefully as a primary source of future development. Contradictions and conflicts are a natural part of the activity system and should not be considered in the context of just another user training problem. They should be seen as a valuable source of information and development.

14. This last point is critical. Implementation is a very long process. The activity system must be allowed to evolve and the community must have time to negotiate meaning. Training for two hours on day one and then turning people loose does not work. Implementation needs to consist of a series of meetings within and across communities to negotiate meaning. Time must be allowed for interactions and participation. A reasonable time frame for a project the size of Transcend would be a year-long series of a combination of meetings, exercises, training, and games. The natural forces of participation and reification must be allowed to work and an environment created to foster their development.

5. The concept of boundary objects reinforces the perceived complexity in building artifacts intended to be shared across communities. Each community has its own requirements and predisposition to meaning. The artifact needs to be carefully designed to accommodate the various communities of practice.

In summary, activity theory provides a helpful overriding theory and framework within which to base software design. The principles above, though not new, have been described using the language provided by activity theory and communities of practice. The principles may be standard in design practice, but activity theory provides a unifying theory and embeds additional meaning in those guidelines.

6. In the design of software systems, communities of practice need to be defined and understood within the context of the activity system. Multiple communities will be the norm in an organization and, thus, require the perspective of a constellation of practices. Sharing and commitment will be much more likely within a community as opposed to across communities. 9. Usage is most important. Meaning is found in the activity and in the engagement of the participants within the community in negotiating meaning. Transcend, Inc., recommends many different tools to encourage and engage users, including games and exercises. We found “the tip of the week” to be effective. All users were given a very brief homework assignment involving the use of a different feature and were required to send back the results. The bottom line is do whatever it takes to get people to use the software. It won’t have meaning without usage.

8. ACKNOWLEDGMENTS Many thanks to Dr. Jim Lockard for his advice and review during the study. Also much gratitude to Dr. Rhonda Robinson for guidance through the methodology issues. And most of all, my sincere appreciation to the people at the bank who participated and helped shape this study. It has been and continues to be a pleasure working with you every day.

9. REFERENCES

10. Users also need to participate with each other to establish meaning. Group exercises or group meetings would be helpful. Communities need a forum to discuss, engage, and establish meaning. In keeping with the principles of participation and reification, the results of these meetings and discussions should be documented and shared with all participants.

[1] Ackerman, M.S. and C. Halverson. Considering an organization's memory. Proceedings of CSCW 98, Seattle, Washington, November, 1998, 39-48. [2] Ackerman, M.S. and B. Starr. Social activity indicators: Interface components for CSCW systems. Proceedings of UIST 95, Pittsburg, PA, November, 1995, 159-168.

11. Users will look to see if other users are active and frame their own immediate acceptance and plan for usage based on those observations [2]. Likewise, they will assess credibility of the information based on who entered it [16]. In the absence of such indicators, users may assume no one is active. It would be helpful for makers of multi-user or cooperative software to provide a screen showing who is logged on to the system and the number of transactions each user has completed during the day. In addition, a history should be provided of who accessed the information and who entered the data. This would provide a better picture of community activity to the individual user and may spur usage. It also would assist in establishing credibility of the information.

[3] Angrosino, M.V. and K.A. Mays de Perez. Rethinking observation: From method to context. In Handbook of Qualitative Research, N.K. Denzin and Y.S. Lincoln, Editors. Sage Publications, Thousand Oaks, CA, 2000, 673697. [4] Bannon, L. and S. Boedker. Beyond the interface: Encountering artifacts in use. In Designing Interactions: Psychology at the Human-Computer Interface, J. Carroll, Editor. Cambridge University Press, New York, 1991, Chapter 12.

12. It is important to encourage key people to be active users of the system. The rest of the community will be watching and establishing credibility based on their activity.

[5] Bardram, J. Designing for the dynamics of cooperative work activities. Proceedings of CSCW 98, Seattle, Washington, November, 1998, 89-98.

13. Immediacy is an issue in cooperative work. In one of our activity system conflicts the officers felt uncomfortable making a referral only on Transcend. They were concerned that the receiver wouldn’t see the referral immediately, so they would also call or leave voice mail. A provision in the software to handle immediacy issues is important to avoid duplication and communication channel conflict. A suggestion is to provide a bell ring and pop-up screen triggered by

[6] Barley, S. The alignment of technology and structure through roles and networks. Administrative Science Quarterly, 35, 1990, 61-103. [7] Bertelsen, O. and C. Nielsen. Dynamics in wastewater treatment: A framework for understanding formal constructs in complex technical settings. Proceedings of the Sixth European Conference on Conference on Computer

151

[24] Kuutti, K. Activity theory as a potential framework for human-computer interaction research. In Context and Consciousness, B. Nardi, Editor. The MIT Press, Cambridge, MA, 1997, 17-44.

Supported Cooperative Work (ECSCW'99), Copenhagen, Denmark, 2000, Kluwer Academic Publishers, Dordrecht, 277-290. [8] Bertelsen, O.W. Organizational learning is crystallised into artifacts. SIGOIS Bulletin, 17(3), 1996, 37-39. [9] Bradley, J. Human-computer interaction and the growing role of context. American Society for Information Science Bulletin, 24(3), 1998, 18-19.

[25] Kuutti, K. and T. Arvonen. Identifying potential CSCW applications by means of activity theory concepts: A case example. Proceedings of CSCW'92, Toronto, Canada, November, 1992, 233-240.

[10] Burkhardt, M. and D. Brass. Changing patterns or patterns of change: The effects of a change in technology on social network structure and power. Administrative Science Quarterly, 35, 1990, 104-127.

[26] Kuutti, K. and H. Karasti. Supporting shared interpretations of a space of representations: A case of radiological conferences. Available: http://webtools.cityu.edu.hk/ct1995/kuutti.htm, 2000.

[11] Carasik, R. and C. Grantham. A case study of CSCW in a dispersed organization. Proceedings of the Conference on Human Factors in Computing Systems (CHI '88), Washington, D.C., May, 1988, 61-66.

[27] Leontev, A.N. Activity, Consciousness and Personality. Prentice-Hall, Englewood Cliffs, NJ, 1978. [28] Luff, P. and C. Heath. The collaborative production of computer commands in command and control. International Journal of Human-Computer Studies, 52(4), 2000, 669-699.

[12] Collins, P., S. Shukla, and D. Redmiles. Activity theory and system design: A view from the trenches. Computer Supported Cooperative Work, 11(1-2), 2002, 55-80.

[29] Mambrey, P. and M. Robinson. Understanding the role of documents in a hierarchical flow of work. Proceedings of Group 97, Phoenix, Arizona, November, 1997, 119-127.

[13] Ebbs, C.A. Qualitative research inquiry: Issues of power and ethics. Education, 117(2), 1996, 217-222.

[30] Markus, M.L. Finding a happy medium: Explaining the negative effects of electronic communications on social life at work. ACM Transactions on Information Systems, 12(2), 1994, 119-149.

[14] Ehrlich, S.F. Strategies for encouraging successful adoption of office communication systems. ACM Transactions on Office Information Systems, 5(4), 1987, 340-357.

[31] Nardi, B. Activity theory and human-computer interaction. In Context and Consciousness, B. Nardi, Editor. The MIT Press, Cambridge, MA, 1997, 7-16.

[15] Engestrom, Y. Activity theory as a framework for analyzing and redesigning work. Ergonomics, 43(7), 2000, 960-974. [16] Erickson, T. and W.A. Kellogg, Social translucence: An approach to designing systems that support social processes. ACM Transactions on Computer-Human Interaction, 7(1), 2000, 59-83.

[32] Orlikowski, W. Evolving with notes: organizational change around groupware technology. Available: http://ccs.mit.edu/papers/CCSWP186.html, 2000. [33] Sachs, P. Transforming work: Collaboration, learning, and design. Communications of the ACM, 38(9), 1995, 36-44.

[17] Flores, F., Graves, M., Hartfield, B., & Winograd, T. Computer systems and the design of organizational interaction. ACM Transactions on Office Information Systems, 6(2), 1988, 153-172.

[34] Savenye, W.C. and R.S. Robinson. Qualitative research issues and methods: An introduction for educational technologists. In Handbook of Research for Educational Communications and Technology, D.H. Jonassen, Editor. Simon & Schuster Macmillan, New York, 1996, 1171-1195.

[18] Gerson, E., & Star, L. Analysing due process in the workplace. ACM Transactions on Office Information Systems, 4(3), 1986, 257-270.

[35] Shukla, S. Context and consciousness activity theory and human-computer interaction: Two book reviews. SIGCHI Bulletin, 29(2), 1997, 51-55.

[19] Heron, J. and P. Reason. The practice of co-operative inquiry: Research 'with' rather than 'on' people. In Handbook of Action Research, P. Reason and H. Bradbury, Eds. Sage Publications: Thousand Oaks, CA, 2001, 179-188. [20] Hutchins, E. Cognition in the wild. The MIT Press, Cambridge, Massachusetts, 1995.

[36] Suchman, L. Office procedure as practical action: Models of work and system design. ACM Transactions on Office Information Systems, 1(4), 1983, 320-328.

[21] Jettmar, E.M. and M.W. Rapp. Computer mediated communication: A relational perspective. Paper presented at Annual Convention of the Western States Communication Association, Pasadena, CA, 1996.

[37] Tyre, M. and W. Orlikowski. Windows of opportunity: Temporal patterns of technological adaptation in organizations. Organization Science: A Journal of the Institute of Management Sciences, 5(1), 1994, 98-119.

[22] Kaptelinin, V., B. Nardi, and C. Macaulay. The activity checklist: A tool for representing the "space" of context. Interactions, July & August, 1999, 27-39.

[38] Vandenbosch, B. and M.J. Ginzberg. Lotus notes and collaboration: Plus ca change... Journal of Management Information Systems, 13(3), 1996/1997, 65-81.

[23] Karsten, H. and M. Jones. The long and winding road: Collaborative IT and organisational change. Proceedings of CSCW 98, Seattle, Washington, November, 1998, 29-38.

[39] Wenger, E. Communities of Practice: Learning, Meaning, and Identity. Cambridge University Press, Cambridge, 1998.

152

Proceedings Template - WORD

Management believed the installation of software with these capabilities would .... certain praxis and to account for the development of new tools over time. 4.

242KB Sizes 0 Downloads 332 Views

Recommend Documents

Proceedings Template - WORD
This paper presents a System for Early Analysis of SoCs (SEAS) .... converted to a SystemC program which has constructor calls for ... cores contain more critical connections, such as high-speed IOs, ... At this early stage, the typical way to.

Proceedings Template - WORD - PDFKUL.COM
multimedia authoring system dedicated to end-users aims at facilitating multimedia documents creation. ... LimSee3 [7] is a generic tool (or platform) for editing multimedia documents and as such it provides several .... produced with an XSLT transfo

Proceedings Template - WORD
Through the use of crowdsourcing services like. Amazon's Mechanical ...... improving data quality and data mining using multiple, noisy labelers. In KDD 2008.

Proceedings Template - WORD
software such as Adobe Flash Creative Suite 3, SwiSH, ... after a course, to create a fully synchronized multimedia ... of on-line viewable course presentations.

Proceedings Template - WORD
We propose to address the problem of encouraging ... Topic: A friend of yours insists that you must only buy and .... Information Seeking Behavior on the Web.

Proceedings Template - WORD
10, 11]. Dialogic instruction involves fewer teacher questions and ... achievment [1, 3, 10]. ..... system) 2.0: A Windows laptop computer system for the in-.

Proceedings Template - WORD
Universal Hash Function has over other classes of Hash function. ..... O PG. O nPG. O MG. M. +. +. +. = +. 4. CONCLUSIONS. As stated by the results in the ... 1023–1030,. [4] Mitchell, M. An Introduction to Genetic Algorithms. MIT. Press, 2005.

Proceedings Template - WORD
As any heuristic implicitly sequences the input when it reads data, the presentation captures ... Pushing this idea further, a heuristic h is a mapping from one.

Proceedings Template - WORD
Experimental results on the datasets of TREC web track, OSHUMED, and a commercial web search ..... TREC data, since OHSUMED is a text document collection without hyperlink. ..... Knowledge Discovery and Data Mining (KDD), ACM.

Proceedings Template - WORD
685 Education Sciences. Madison WI, 53706-1475 [email protected] ... student engagement [11] and improve student achievement [24]. However, the quality of implementation of dialogic ..... for Knowledge Analysis (WEKA) [9] an open source data min

Proceedings Template - WORD
presented an image of a historical document and are asked to transcribe selected fields thereof. FSI has over 100,000 volunteer annotators and a large associated infrastructure of personnel and hardware for managing the crowd sourcing. FSI annotators

Proceedings Template - WORD
has existed for over a century and is routinely used in business and academia .... Administration ..... specifics of the data sources are outline in Appendix A. This.

Proceedings Template - WORD
the technical system, the users, their tasks and organizational con- ..... HTML editor employee. HTML file. Figure 2: Simple example of the SeeMe notation. 352 ...

Proceedings Template - WORD
Dept. of Computer Science. University of Vermont. Burlington, VT 05405. 802-656-9116 [email protected]. Margaret J. Eppstein. Dept. of Computer Science. University of Vermont. Burlington, VT 05405. 802-656-1918. [email protected]. ABSTRACT. T

Proceedings Template - WORD
Mar 25, 2011 - RFID. 10 IDOC with cryptic names & XSDs with long names. CRM. 8. IDOC & XSDs with long ... partners to the Joint Automotive Industry standard. The correct .... Informationsintegration in Service-Architekturen. [16] Rahm, E.

Proceedings Template - WORD
Jun 18, 2012 - such as social networks, micro-blogs, protein-protein interactions, and the .... the level-synchronized BFS are explained in [2][3]. Algorithm I: ...

Proceedings Template - WORD
information beyond their own contacts such as business services. We propose tagging contacts and sharing the tags with one's social network as a solution to ...

Proceedings Template - WORD
accounting for the gap. There was no ... source computer vision software library, was used to isolate the red balloon from the ..... D'Mello, S. et al. 2016. Attending to Attention: Detecting and Combating Mind Wandering during Computerized.

Proceedings Template - WORD
fitness function based on the ReliefF data mining algorithm. Preliminary results from ... the approach to larger data sets and to lower heritabilities. Categories and ...

Proceedings Template - WORD
non-Linux user with Opera non-Linux user with FireFox. Linux user ... The click chain model is introduced by F. Guo et al.[15]. It differs from the original cascade ...

Proceedings Template - WORD
temporal resolution between satellite sensor data, the need to establish ... Algorithms, Design. Keywords ..... cyclone events to analyze and visualize. On the ...

Proceedings Template - WORD
Many software projects use dezvelopment support systems such as bug tracking ... hosting service such as sourceforge.net that can be used at no fee. In case of ...

Proceedings Template - WORD
access speed(for the time being), small screen, and personal holding. ... that implement the WAP specification, like mobile phones. It is simpler and more widely ...

Proceedings Template - WORD
effectiveness of the VSE compare to Google is evaluated. The VSE ... provider. Hence, the VSE is a visualized layer built on top of Google as a search interface with which the user interacts .... Lexical Operators to Improve Internet Searches.