Teledemocracy by Ytterstad, Akselsen, Svendsen, and Watson

The field trial	Salangen municipality, in the county of Troms, was chosen as the project partner and site for the field trial. Salangen is in many aspects a typical Norwegian rural community with approximately 2,500 inhabitants. It is also quite limited in terms of geography, which means that benefits from the use of telecommunications due to distance between participants in the project could be greater in larger municipalities. Finally, Salangen was chosen because representatives of the political scene responded with enthusiasm to the idea of evaluating new technology to support their everyday work.
Approach	The project was divided into major phases (see Figure 11). The lower elements of Figure 11 are activities involving the users. The figure illustrates the involvement of the users in systems design, an iterative activity.
Figure 11. Basic project activities [click on the rectangle in the image map to read about the respective activity].
	After developing the first version of the system, the technology was deployed in two steps. First, project board members piloted POT intensively for three weeks. Their feedback led to several improvements in the system and also made them capable of taking on a supporting role in the next phase, when the rest of the users were included in the trial. Prior to the distribution of the technology, a one-day hands-on training course was held in November 1993. The period following the training session was used by the politicians to become acquainted with the equipment, electronic word processing, and other basic PC operations. About 30 percent of the users expressed a need for another training course, which was held in March 1994. During the first phases of the project, the different political parties took the initiative for defining internal routines based on POT (e.g., the distribution of meeting agendas and the co-authoring of political documents). Since September 1994, the municipality's administration has electronically distributed material to politicians. Considerable effort was spent following up with the users in order to assist whenever they had technical problems, to receive feedback for further adjustment of the POT software, and to compile data for a continuous evaluation of the project. Software was implemented to capture data on the systems' usage. During the initial phase of the project, we experienced some technical problems with POT, mostly tele-technical issues due to an old and unreliable network infrastructure. This led to several upgrades of the software in order to make POT more stable in a demanding environment. POT was designed to allow for remote upgrading of software. However, this functionality requires reliable network connections. The updates, as a result, had to be done on site, could not be carried out by novice users, and required considerable resources. On the other hand, this gave the team a good opportunity to follow-up with the users, which in turn contributed to preserving the politicians' faith in the project.
Observations	Use of the system (see Figure 12) during the first parts of the field trial period can to some extent be explained, with hindsight, by events that occurred: the mayor's death in January 1994, the second training session in late March 1994, and the start of the administration's electronic distribution of information in September 1994.
Figure 12. Frequency of use per month. All explicit actions taken by the politicians in order to send or receive electronic mail are included.
	Championship and mediation From the early phases, the mayor was the project's champion. He actively promoted it and exhibited transformational leadership. He used the three resources traditionally used by champions (Kanter, 1983): information, material resources, and political support. He had access to the necessary information and resources from his partnership with Telenor, and he also had the full support of the local council for piloting new technology. As the head of the local council, he had a rather unique and strong position in terms of political support. In January 1994, a few months after the project started, the mayor died. This was of course a great loss for his family and the Salangen community. In retrospect, it also turned out to be a greater loss to the project than we first anticipated. The lack of a driving force for the project in the community increased the need for support of the project from the research institute, located in Tromsø, three hours away. Among the Salangen politicians, we observed the presence of technology-use mediators in some of the political parties. This was reflected in more extensive use of the system among a party's members and also in more innovative uses of the system (e.g., in the co-authoring of position papers for the budget discussion). Through the part-time employment of one of the most experienced users (for doing support), we recognized and rewarded this mediation role. This was not the case with another skilled computer user in the administration who obviously had the potential to become a mediator. As a result, he was not very involved in the project, and the project lost a valuable agent for adoption. Recognizing and rewarding intervention roles may be a crucial step in the successful adoption of new information technologies. Adoption of POT in the user group The Salangen politicians and their administration constitute a group of users with great diversity, both in background and computer experience. Some of them had never seen a keyboard before and spent considerable time just finding the spacebar! Further, training time was limited because it was difficult to assemble a group of more than 30 persons, all of whom had tight schedules. These factors and the need for understanding the file handling aspects of the system were expected to be major obstacles for establishing a critical mass of users. Surprisingly, these potential problems were quite easily handled. Users with extremely limited prior computer experience were able to perform the basic operations needed for communicating via POT within a short time. The users identified the system's intuitive interface and low user threshold as reasons why they were able to learn easily how to use the system. The internal POT log-files, as well as the spoken and written feedback from the Board and the rest of the user group, indicate little relationship between previous computer experience and use of the system. Individual motivation tended to be the crucial factor for attaining broad use of the system. The less experienced users were motivated to use the system in order to become more acquainted with computer and communication technology, whereas some of the more experienced users awaited the introduction of POT-based routines mandatory for their political work. Effects of design decisions on inexperienced users By using a graphical directory, the system presents symbols, pictures, or other graphical information readily recognizable to users. This interface design was successful not only because of the system's ease of use but also due to its ability to create an illusion of proximity, because of the comforting sight of familiar symbols. The use of party logos as pointers to directory information seems to have been a crucial design decision. In the first version of the POT software, members of a specific party group were made accessible by double-clicking the party's icon. However, by observing some of the inexperienced users, especially older politicians, it became obvious that a double-click was a difficult skill to perform as many users were performing the operation using full body movements -- with a tremendous punch all the way from their tiptoes to their fingertips. Therefore, where possible, a double-click was replaced by a single-click. Initially, some of the politicians expressed apprehension about the introduction of computer technology because they feared it may give more power to experienced computer users and thus limit the possibility to participate on the political scene on equal terms. We observed little relation between prior experience and use of the system, and an unequal distribution of power due to computer proficiency did not emerge. However, it seems crucial that the user-threshold is sufficiently low so that the inexperienced user is able to quickly and easily learn how to use the system. Benefits and workload During early 1994, the political parties defined their own internal routines based on POT. The participation as well as the nature and frequency of use varied among the different parties. The most common uses within the political party groups included: distribution of meeting agendas and protocols; co-authoring political documents; meeting coordination, both convening and holding. E-mail has the potential to open discussions to all politicians, not just those within a particular party. Nonetheless, politicians have made minimal use of this option probably because they are uncertain whether such use is approved behavior. Also, some have expressed a fear for having non-agreed-upon positions discussed too early outside the party group. As with most new media, this is really a question of discovering what the stakeholders define as acceptable standards of behavior. POT introduces a disparity between those who benefit from the application -- the politicians -- and those who perform additional work to support it -- the administration. This tension has been considered as one of the classical challenges for design of CSCW systems (Grudin, 1989, p. 248): The application fails because it requires that some people do additional work, while those people are not the ones who perceive a direct benefit from the use of the application. In addition, the application may fail if it does not allow for the wide range of exception handling and improvisation that characterizes group activities (Suchman, 1983), or if, through ignorance, it violates social taboos, threatens existing political structures, or otherwise de-motivates some users (Dyson, 1988). The design process fails because our intuitions are poor for multi-user applications. Decision-makers see the potential benefits for people similar to themselves, but don't see the implications of the fact that extra work will be required for others. For a transitional period, the Salangen administration had to distribute protocols, meeting agendas, and meeting books in the traditional way, in addition to the electronic distribution of these documents. Although this placed an extra workload on the administration, the prospects of being able to copy and distribute all material electronically motivated administrative personnel. Electronic copying and distribution can eliminate considerable routine work, including photocopying, putting materials into envelopes, and taking them to the nearest post office. The potential value of electronic distribution became obviously apparent to the administration one day when its photocopier was out of service for an extended period. The fact that the politicians receive some issue information earlier (as much as 14 days) is highly appreciated as it increases their opportunities to shape the political process. In addition, because of ease and low cost, some material is now distributed to a larger number of politicians. Also, e-mail is used by some politicians to contact the administration in order to get hold of information on previous issues. Communication patterns and the social dimension A fear expressed prior to initiation of the project was that communication through the system eventually would substitute for face-to-face contact between politicians. This concern was expressed by the mayor of a nearby municipality who criticized parts of the project believing they had the potential to transform politicians into mere readers of electronically distributed issue information. As a result, he postulated that contacts between the politicians and the citizens, as well as among politicians, would be significantly reduced. Research on computer-mediated communication, however, suggests that the use of e-mail creates a higher degree of contact and contributes to considerably less communication isolation (Eveland and Bikson, 1988). Electronic messages sent by POT contain informal communication, social chat, as well as strictly political matters. Throughout the field trial, we observed that communication through more traditional media was to a very little extent replaced by electronic mail. On the contrary, the most extensive users of the system claim that contact between them improved. This finding is in contrast to Finholt, Sproull and Kiesler (1990) who found that increased e-mail use in ad hoc groups reduced communication (face-to-face meetings, phone conversations, memos, etc.). This disparity may be due to the nature of the groups -- politicians conducting their everyday business versus students participating in an experiment intended to clarify organizational communication processes. The use of students ( e.g., Finholt, Sproull and Kiesler, 1990; Siegel et al, 1986) and others, has been highly debated: "Not only are college students a very particular kind of population with respect to workers in public or private organizations, ...; it is also worth noting that laboratory experiments are designed precisely to exclude all the social cues that make organizations 'real' in everyday situations" (Mantovani, 1994, p. 54). Some users point out that POT has introduced a common denominator in the political community, cutting across the traditional party lines. This has strengthened the ties between the different party groups because they face a common challenge. The POT project board argues (May 1994): "This challenge seems to have united the politicians, brought them closer together, and thereby added a social dimension to the project". Possible Hawthorne effects The Hawthorne effect, first identified by (Mayo, 1933), states that when one introduces a new method of performing a task and participants know that it is part of an effort to improve performance, there is a temporary gain, even if the new method is no better (or even worse) than the old way of doing things. There is a tendency for increased effort and higher motivation resulting in better short-term performance. Thus, the Hawthorne effect can seriously affect the validity of evaluation results, particularly when assessing an intervention such as POT, where: The publicity within the community of being the field trial site might have made the politicians want the project to be successful; The individual attention received by politicians might have influenced their use of the system, motivation, self-confidence, and perception of the system's usefulness. Thus our observations and statements made by the politicians should be considered in this context.