Business & Economics

Social Considerations in the Development, Deployment and Adoption of Web-Based Organizational Memories

Description
Social Considerations in the Development, Deployment and Adoption of Web-Based Organizational Memories Andrew Gorman Center for LifeLong Learning and Design Department of Computer Science and Institute
Published
of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
Social Considerations in the Development, Deployment and Adoption of Web-Based Organizational Memories Andrew Gorman Center for LifeLong Learning and Design Department of Computer Science and Institute of Cognitive Science University of Colorado Boulder, CO Abstract This paper begins with the premise that people want to participate in designing their future. While an individual unaided human mind is powerful, real power is derived from humans working in conjunction with tools and other humans. Developing computational support for collaboration among groups of individuals is a difficult task, but its potential benefit is tremendous. This paper will describe three broad models of collaborative systems along with examples of systems that typify each model. The sociotechnical aspects of collaboration and participation will be discussed as they relate to the development, deployment, and adoption of an organizational memory system intended to support the collaborative design of a new building. Introduction The Discovery Learning Center (DLC) is a new building being constructed on the campus of the University of Colorado at Boulder. This process gathers groups of people who have varying backgrounds and interests. The goal that is common to all is the construction of a new building. However, each has his or her own unique agenda. In order to support such a process, it is desirable to create an information space that can be useful to all stakeholders as they participate in the design and construction of the new building. The DLC information space began as a static repository of design documents and background information. It has since grown into a more dynamic system for disseminating design alternatives and gathering feedback from stakeholders. The adoption of such a system by stakeholders who have little prior exposure to one another can be a very complicated process, which is affected by both technical and social factors. Information Access is Necessary but not Sufficient In today's workplace, people need to know how to access information. However, simply knowing how to access information falls short of what is truly needed in today s workplace. For example, in the President s Information Technology Advisory Committee report (PITAC, 1999), there is a call for ubiquitous information access. This call for accessibility needs to be extended to include the understanding that the key to the future lies not only in greater access to information, but in greater support for knowledge construction (Scardamalia & Bereiter, 1994). Although there is value in such an access model, its focus is incomplete. It is based on an impoverished view that relevant knowledge already exists, waiting to be accessed. In order to truly gain the benefits of information technology, what is needed is not simply greater access to information, but a greater ability for average people to construct and distribute new knowledge (Arias, 1999 (in press)). People Want to Participate How can more than 261 million individual Americans define and reconcile their needs and aspirations with community values and the needs of the future? Our most important finding is the potential power of and growing desire for decision processes that promote direct and meaningful interaction involving people in decisions that affect them. Americans want to take control of their lives (PCSD, 1996, p.7). This finding of the President's Council on Sustainable Development (PCSD) supports the claim that something more is needed than access alone. Therefore, an important challenge for future information technology is to enable stakeholders of problems to become involved informed participants (Brown, Duguid, & Haviland, 1994). To make informed participation a reality, we need support for new forms of knowledge creation, integration, and dissemination. People seldom explore large repositories of information in the abstract (Fischer, Lemke, McCall, & Morch, 1996; Moran & Carroll, 1996). Instead, information is typically sought in response to breakdowns encountered during meaningful, real-life activities (Fischer, 1994; Popper, 1965). By overcoming such breakdowns, new knowledge is created, which then must be integrated with any knowledge that may have been generated during prior breakdowns. This cycle leads to the creation and evolution of rich information spaces that can empower interested community members as they take control of their lives. 2 M1-M3 Models of Participation A typical model of information sharing (e.g., (Ackerman & Malone, 1990; Ackerman & McDonald, 1996)) focuses on experts sharing information with non-experts. This is through a single person or small committee. This is typical of information generated in an open source model of software development (Raymond, 1999). Here there are many contributors, but only a few (or one) that integrates feedback from the community back into a coherent structure. There has been much success using this model in open source movements (Fielding, 1999; Torvalds, 1999) and while this is an advantage over the M1 model, in terms of collaborative construction, it can lead to problems of scalability. Furthermore, there needs to be a highly dedicated person that has the full-time responsibility of analyzing and structuring all of the feedback. Figure 1 - The M1 Model of Collaboration represented by the M1 model (seen in figure 1) in which a class of experts controls the production of information and individuals act as consumers whose only need is that of access. In the M2 model (figure 2), all information is funneled Figure 3 - The M3 Model of Collaboration The M3 model (figure 3) can theoretically support distributed collaboration in a more direct way. In this model, there is no gatekeeper. All contributors have the ability to add content directly. Developing systems that support this type of collaboration can be extremely difficult. One way of providing structure to collaboratively constructed information is by codifying the knowledge and expertise of the gatekeeper described in the M2 model. Another approach is to distribute this responsibility among the community members by establishing policies (Edwards, 1996) to govern the construction, organization and use of information. Human Cognition and Tasks The memory of an individual can be roughly divided into two categories: short-term memory (STM) (also described as 3 working memory) and long-term memory (LTM). 1 STM is relatively small, typically thought to have a capacity of 7 ± 2 chunks of information (Miller, 1956). In contrast, LTM is virtually unlimited (Matlin, 1998). In an information processing theory of human cognition (Pinker, 1997), data needed for a given task is typically activated and retrieved from LTM and then held in STM while it is actively used. In this process, information is constantly being swapped in and out of STM as new information is constructed and encoded into LTM. This model of human memory is analogous to register and disk storage used in modern-day computers. Expert behavior is often based on a well-developed technique, or pneumonic, for encoding and retrieving information. Intelligent behavior, therefore, is often attributed in large part to being able to effectively transfer information into LTM so that it can later be activated for future use in STM. Because of the limitations of STM and human attention, cognition can be viewed as a limited scarce resource that needs to be allocated as properly during task performance. This view is analogous to operating systems that allocate computer resources during the execution of a process or sub-process. Given that human cognition is limited, it makes sense that the nature of a task (i.e., the cognitive resources it demands) affects our ability to perform the task. Over the course of human history, cultures have invented tasks that not only push, but also transcend the limits of human cognitive capabilities. Because of this, humans have needed to develop cognitive artifacts (Norman, 1991) and systems to aid in their artificial tasks. For example, long division is an artificial system that produces such a cognitive load that it is difficult to solve even a moderately advanced problem without the aid of cognitive artifacts such as memory aids like paper and pencil. With the advent of computers, some of these artificial tasks have been codified so that they may now be performed with computational devices such as handheld calculators. Distributed Cognition Distributed cognition (Brown, et al., 1993; Fischer, 1995; Hewitt & Scardamalia, 1996; Hutchins, 1993; Norman, 1993; Salomon, 1993) emphasizes that the heart of intelligent human performance is not the individual human mind in isolation but the interaction of the mind with tools and artifacts as well as groups of minds in interaction with each other. It is important to understand the fundamental difference between these two forms of distributed cognition. When distributed cognition is at work between the individual human mind and cognitive artifacts, it often functions well because the knowledge an individual needs is distributed between her/his head and the world (e.g., calculators, address books, messages, filing cabinets). On the other hand, when cognition is distributed among groups of minds, a group has no head, no single mind to store the information about this distribution of knowledge, which is available to 1 Sensory memory is also described in the cognitive psychology literature, but this is outside the scope of this paper. For more information on sensory memory see (Matlin, 1998) all members of the group. In this case, externalizations are critically more important. Externalizations (1) create a record of our mental efforts, one that is outside us rather than vaguely in memory and (2) represent artifacts that can talk back to us (Schön, 1992) and form the basis for critique and negotiation. These can be thought of as cognitive artifacts for groups. Although creative individuals are often thought of as working in isolation, the role of interaction and collaboration with other individuals is critical (Engelbart, 1995). Creative activity grows out of the relationship between an individual and the world of his or her work, and out of the ties between an individual and other human beings. The predominant activity in complex problem solving is that participants teach and instruct each other (Greenbaum & Kyng, 1991). Because complex problems require more knowledge than any single person possesses, it is necessary that all involved stakeholders participate, communicate, and collaborate with each other. For example, during the design and construction of a new building on a college campus, there are building architects, tenants, collage administrators, and government legislators, all of whom have different agendas and background knowledge. Furthermore, communication breakdowns are often experienced because stakeholders belonging to different cultures use different norms, symbols, and representations (Snow, 1993). Such a setting is governed by a symmetry of ignorance (Rittel, 1984) in which all stakeholders are aware that even though they each possess relevant knowledge, none of them has all the relevant knowledge. Each participant must act as a reflective practitioner rather than as an all-knowing expert (Schön, 1983). Organizational Memories for Supporting Long and Short Term Collaboration The term organizational memory has no clear or agreed upon definition within the computer science literature (Ackerman & Halverson, 1998). In this paper, the following operational definition will be used: Organizational memory provides a shared information space that supports a group of people (an organization) to do work. The information space should be living in the sense that it is an evolving product of the work done by the members of the organization as opposed to simply being a static storage of information (Fischer, 1998a) In this definition, an organizational memory can be viewed as a cognitive artifact that provides an externalization for groups of minds interacting with each other. One could argue that an organization or group s memory exists independent of any explicit externalization. Every group, organization, or culture has tacit knowledge that helps govern or define acceptable behavior, standard operating procedures, and social norms. It can be very beneficial to identify and make explicit a groups tacit knowledge, but this not the concern of this paper. Instead, this paper will consider the use of organizational memory systems supporting the collaboration of a heterogeneous group of stakeholders during the design of an artifact. 4 Media and Systems Supporting Organizational Memory and Collaboration A challenge for supporting collaboration is in providing a mechanism that allows various participants to integrate their perspectives (Stahl, 1993) in a meaningful way. Supporting informed participation requires processes that integrate the individual and the group knowledge through collaborative constructions. Information spaces need to be constructed collaboratively (Scardamalia & Bereiter, 1994) and integrated into the work and social practices of the community (Lave, 1988). These collaborative constructions result in work products that are enriched by the multiple perspectives emerging through community discourse. Ackerman (Ackerman, 1998) describes the development of Answer Garden, a system for capturing and managing an organizations knowledge and expertise. One of the primary motivating forces behind such a system is an organization wishing to share expertise and avoid reinventing the proverbial wheel. The most common scenario for using a system like Answer Garden is when a worker is faced with a problem to which they have no answer. Answer Garden supports this dilemma in two possible ways: it makes a previously recorded solution available for retrieval or, in the absence of such a solution, it provides access to the person who is likely to have the appropriate expertise. This approach focuses on recording and structuring knowledge so that it may be available for later use. However, this model makes a strict distinction between those who possess knowledge and those who do not. There are two separate roles in this scenario and there is no intersection seen between these two groups. This resembles the M1 model of collaboration where a class of experts provides access to information for a class of non-experts. While systems such as Answer Garden may be useful in many situations, it falls short of supporting groups of collaborators where each person is simultaneously an expert and a non-expert (e.g., an expert building architect, but a non-expert campus planner). What is needed in this case is a system that is open to all users. The DynaSites system (Ostwald, 1997) is another type of system for the creation of dynamic and user-extensible webbased information spaces. This system is designed to support M3 types of collaboration where all users can directly contribute to the construction of new knowledge. One of the fundamental problems of sustaining a useful open information space is that of maintaining and organizing the information. One approach to this problem is the use of a gatekeeper (Raymond, 1999), but as mentioned earlier, this has problems of scalability. Another approach is to view an information space as a seed (Fischer, 1998b). In this approach, an information space begins as a seed which evolves over time through use. At some point, the information space grows to be so unwieldy that it is no longer useful (i.e., it is too difficult to find relevant information). At such a time, the information needs to be restructured, or re-seeded. However, in practice, this is difficult to accomplish. Questions of what information is pruned from the space can be difficult to answer. Usage data can help answer questions like these, but still, any major restructuring of critical information source can potentially be very disruptive. Participation Revisited All of the discussion thus far has been based on the premise that people want to participate. After all, according to the President's Council on Sustainable Development (PCSD, 1996, p.7), Americans want to take control of their lives. However, just because people want to take control, does that mean that they actually do? What evidence exists to support this conclusion? According to the Institute for Democracy and Electoral Assistance, voter turnout in US presidential elections shows a downward trend in participation (IDEA, 1999). In fact, 1996 has had the lowest percentage of voter turnout (47.2%) in over 50 years 2. However, is voter turnout in US presidential elections an indicator of people s desire and willingness to participate in designing their future at a local, more intimate level? Not necessarily, but it does raise some questions. One factor affecting the use of group information spaces is that of perceived utility. The utility of a system can be viewed as a function of perceived benefit and required effort. Two fundamental question that must be asked (and answered) are, Who does the work? and Who receives the benefit? (Grudin, 1994). A classic example is found in the practice of software documentation. The software developer who is required to document his or her code may perceive no direct benefit in doing so. For this reason, software documentation often goes undone. One way of increasing utility is to decrease the required effort. Another is to increase the perceived benefit. Traveen, et al. (Terveen, Hill, Amento, McDonald, & Creter, 1997) have sought to do both. They have developed PHOAKS (People Helping One Another Know Stuff), which is a recommended system, or social filter. This system was able to automatically search Usenet groups and find positive references, or recommendations of Web sources. In this case, user effort is negligible: no effort was required to input or organize information. What about it s perceived benefit, though? As it turns out, users perceived considerable benefit, but the programmatic identification and categorization of sources was imperfect. At times, PHOAKS recommendations were seen to be misleading or the sources that were recommended became outdated (dead links). The mechanism for correcting these problems was a manual process. In an M2-like fashion, users would their feedback to the PHOAKS administrators, who would then evaluate and respond as they saw fit. While this method was somewhat successful, the issue of scalability and sustainability became obvious: as the use of the system grew, could they keep up with user demand? By analyzing usage patterns and types of requests, the PHOAKS developers were able to create mechanisms that allowed users to directly modify the content of their system (Hill & Terveen, 1997). One interesting point that was made by these researchers was that there was a marked distinction in consumer / producer roles for users of their system. Specifically, they found that only a small minority of users expended the effort of contributing modification. 2 Statistics were only available dating back to 1948. 5 The Design and Construction of the Discovery Learning Center The Discovery Learning Center (DLC) is a building being constructed on the campus of the University of Colorado, Boulder. It is being built to support a broad-reaching vision for the College of Engineering and Applied Science known as the Discovery Learning Initiative. (DLI) In an effort to embody the spirit of the initiative, a web-based information space was created (Gorman, 1999) in May of 1999 to support the informed participation of all stakeholders involved in the project. Stakeholders in the project include the Dean and associate Deans of the college, perspective tenants, the university s facility management group, state legislators, local industry and private donors, and the bu
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks