User Evaluation of a Community-centric Music Voting Service

Transcription

1 User Evaluation of a Community-centric Music Voting Service Timo Koskela*, Janne Julkunen, Otso Kassinen, Mika Ylianttila Department of the Electrical and Information Engineering, University of Oulu, Oulu, Finland firstname.lastname@ee.oulu.fi Abstract In this paper, we present the user evaluation results of a Community-centric Music Voting service that relies on a DHT-based P2P network. For each user community, an independent DHT-overlay is established. These separate DHT-overlays are then connected to specific commercial premises or larger areas. A user evaluation was conducted in a real usage environment, where the user communities were comprised of the clientele of a bar. In the user evaluation, we examined the effect of the attitudinal, social and perceived behavioural control constructs on the usage intentions towards the Community-centric Music Voting service. The research data were collected with both quantitative questionnaires and short interviews. The results indicate that the suitability to the users behavioural models and the earlier experiences with mobile services are the two most important factors affecting the usage intentions towards the Community-centric Music Voting service. Keywords Peer-to-peer, theory of planned behaviour, mobile music 1. Introduction The placing of new mobile services successfully to the markets has been seen as a very challenging task. This is mainly due to the fact that the most of the mobile services are completely new to the users, and thus, the users perception of the added value is hard to predict [1]. Therefore, new knowledge about quickly evolving user preferences is needed. The music services entered the mobile devices at the turn of the millennium, and nowadays, the mobile device as a music consumption medium has become a common phenomenon. In general, the term mobile music has been defined as any musical activity using portable devices that are not tethered to a specific stationary locale [2]. Although mobile music has received a plenty of attention in the research community (e.g. [2][3][4][5]), the feasibility of the community-centric mobile music services has been examined only in a few studies [6][7]. In this paper, we present the evaluation results of a Community-centric Music Voting (CCMV) service. Using the CCMV service, the clientele of a bar can affect the music played in commercial premises by voting for songs with their mobile device. In addition, the clients of the bar can interact with each other using the messaging functionality. Thus, the CCMV service provides modern citizens means for affecting their surroundings in both virtual and physical spaces [8]. The goal of the user evaluation was to examine the feasibility of the CCMV service, as well as its business potential. More specifically, we focused on identifying and tackling the obstacles for a large-scale adoption of the CCMV service. Furthermore, based on users perception on the pricing of the CCMV service, we also examined the potential revenue models. The user evaluation was conducted in a real usage environment, where the user communities were formed by the customers of a bar. The research data were collected both with quantitative questionnaires and short interviews. For the analysis of the collected research data, two approaches were utilized: the first one was based on behavioural models, and the second one on defined special interest groups whose views on the service were compared in order to detect differences in their perceptions. This paper continues the work done in [9], where a small scale user evaluation was conducted in a laboratory environment with the earlier version of the CCMV service. The rest of the paper is organized as follows: Section 2 introduces the CCMV service in brief, Section 3 presents the two evaluation approaches, Section 4 describes the details of the user evaluation event, Section 5 presents the results of the user evaluation, and finally, Section 6 concludes the paper. 2. The Community-centric Music Voting Service The CCMV service relies on the Converged P2P and Web service architecture [10] and on a commercial virtual DJ system called DJOnline by Kutalab Ltd. The Converged P2P and Web service architecture provides

2 functionalities for establishing and managing user communities; in the architecture, each user community is created as a separate DHT-overlay. The established user communities are then connected to specific commercial premises or larger areas by using shortrange communication technologies for binding a community s information to a specific location. The access to these user communities is provided either with Bluetooth or RFID technologies that are used to retrieve the correct DHT-overlay joining parameters from a Bluetooth beacon or an RFID tag installed in the proximity of the specific commercial premises or a larger area. The DJOnline contains a huge database of music that is streamed to the client front-ends residing in the commercial premises. Each piece of music is enriched with metadata that has been added by music professionals. The metadata is used for automatic creation of playlists according to the music profile of the commercial premises. In CCMV, the DJOnline also offers functionality for the staff of the commercial premises to create voting events. A voting event contains two or more alternative songs that can either be selected manually by the staff or automatically by the DJOnline. The members of the user community can vote which of the songs they would like to be played. to music, the users can examine some pieces of context information (such as the mood or taste for music) of the other community members (if permitted), and exchange messages. The current version of the CCMV service has been implemented on the (mobile) Maemo Linux platform. 3. The Evaluation Approaches The feasibility of the CCMV service was evaluated with two different approaches. The first one is based on the Theory of Planned Behaviour (TPB) [11][12] and the second one on the defined special interest groups. 3.1 The Decomposed Theory of Planned Behaviour The research model of the TPB describes the Behavioural Intention (BI) of a user with three constructs that are Attitude (ATT), Subjective Norm (SN) and Perceived Behavioural Control (PBC) [11]. The TPB has been extended from the Theory of Reasoned Action (TRA) that has its roots in social psychology [12]. Nevertheless, both theories have also been successfully applied in studies related to the acceptance of new technology. Figure 1. The user interface of the CCMV service. The main view of the CCMV service displays the active playlist in the user s current community. By selecting a song from the playlist, a user can view more information about the song and the artist. The voting events available are also shown in the same playlist. By selecting a voting event, the user is shown the voting alternatives, as illustrated in Fig. 1. Every member of the community has a single vote per song. In addition to voting, the members of the community can request specific songs that they would like to hear. However, it is up to the barkeeper to decide whether the requested songs fit the musical style of the bar and are added to the playlist. Besides the activities related Figure 2. The Decomposed Theory of Planned Behaviour. In this paper, the constructs in the TPB are further decomposed into an underlying belief structure that

3 illustrates the factors that we presuppose to have an effect on the constructs [11]. This more elaborate version of the TPB is usually called the Decomposed Theory of Planned Behaviour (DTPB). The defined belief structure in the DTPB is illustrated in Fig. 2. The detailed descriptions of the constructs and the defined hypotheses follow our previous work [9], but are also presented here to improve the clarity of this paper. After each hypothesis, the corresponding statements from the questionnaire are presented. Perceived playfulness (PPF) is the term for an individual s subjective experience concerning the human-computer interaction [13]. In cases, where an entertainment-oriented service is used, PPF is an important factor for both encouraging the usage of the service and developing an attitude towards the system [14]. In addition, the PPF has been found to affect significantly the usage of the WWW for entertainment purposes [15]. Thus, we assume that the following hypothesis will apply: H1: A user s perception of playfulness is positively associated with attitude. PPF1: It was fun to vote for music and to follow the progress of the voting results. PPF2: Voting for music has a positive impact on my enjoyment of the cafeteria. PPF3: While voting for music, I did not realize the elapse of time. For Perceived Usefulness (PUF), there are different definitions in the literature. According to the Technology Acceptance Model (TAM) [16], PUF stands for the degree to which the system is seen to improve the job performance. The definition mentioned seems to be rather limited and would exclude the use of PUF from our case, because our service is obviously more entertainment- than professionally-oriented. However, we selected to take a broader view, where PUF is defined as some value that the product or the service provides to an individual user [17]. Based on the latter definition, we assume that the following hypothesis will apply: H2: A user s perception of usefulness is positively associated with attitude. PUF1: The Music Voting service is useful to me. PUF2: It is important to me that I can affect the music played in a cafeteria. PUF3: The Music Voting service would be a useful auxiliary service to the offerings of a cafeteria. PUF4: The availability of the Music Voting service would positively influence my choice of a bar. Image (IM) is the degree to which the usage of an innovative product is seen to contribute to an individual s image or status in his or her social system [18]. The manufacturers of mobile devices have noticed the significance of image and have accordingly identified different user groups that value different aspects of their mobile devices [11]. The same trend supposedly applies to mobile services as well, and thus, we assume that the following hypothesis will apply: H3: A user s perception of image is positively associated with attitude. IM1: Usage of the Music Voting service would be considered positively in general. IM2: The Music Voting service is trendy. IM3: A possibility to vote for music would polish opinion about the bar. The price paid for the service is, in essence, a sacrifice that is given in exchange for being able to take advantage of a specific service. Clearly, the price is a subjective concept and is often described using qualitative measures instead of plain numbers. [19] Thus, the users individual perceptions on the relation of the Perceived Price Level (PPL) and the predicted benefits from a service affect their attitude towards the service significantly [14]. H4: A user s perceptions of perceived price level are negatively associated with attitude. PPL1: I would not want to pay for the usage of Music Voting service. PPL2: The expenses for the Music Voting service would be a barrier to me. PPL3: The benefits gained from the Music Voting service are not of monetary value. PPL4: The costs of the Music Voting service should be included in the costs of other offerings of a bar. When individual users adopt a new innovation or service, their beliefs are strongly influenced by their reference groups [20]. In the case of the adoption of Internet-based technologies, these reference groups can be, for instance, colleagues, friends, supervisors or top management [11]. However, taking into account the fact that our system is a mainly leisure-time oriented entertainment service, only Friend Influence (FI) was seen relevant to this context. Thus, the following hypothesis is assumed to apply: H5: Friend influence is positively associated with subjective norm.

4 FI1: My friends would assume that I used this kind of Music Voting service. FI2: I believe that the most of my friends and my family members would use the Music Voting service. FI3: My friends would think positively of me when I use the Music Voting service. Cultural Influence (CI), in a sense, takes FI to the next level, pertaining more to the values and norms of the entire society. Cultural values do not only determine the behaviour that is acceptable in particular situations, but also influence how we interact with other members of our society [21]. Culture has also been seen to have a strong influence on people s motivations, lifestyles, and even product choices [22]. Consequently, we assume that the following hypothesis will apply: H6: Cultural influence is positively associated with subjective norm. CI1: The Music Voting service would fit well into bars. CI2: Using the Music Voting service in bars would not be generally disturbing. CI3: I think that people would spend their time with the Music Voting service in bars. When people interact with a computer system for the first time, they tend to feel embarrassed and stressed. After becoming more familiar with the system, interaction is likely to become easier and more spontaneous [23]. In addition, it has been recognized that people take a more positive stand towards systems that are compatible with the ones they have adopted and used earlier [24]. Since people are less reluctant to adopt those new systems and services that utilize familiar technologies, we assume that their earlier Internet Experience (IE) will play an important role: H7: Internet experience is positively associated with perceived behavioural control. IE1: I think that I am an experienced user of Internet. IE2: I use Internet services such as Messenger, Skype, and IRC to keep in touch with my friends and family. A theory similar to the one presented in the context of IE applies also to Mobile Experience (ME), but it should be noted that familiarity with technology in the fixed-line realm does not guarantee competence in the mobile realm. Thus, Internet and mobile experience need to be examined separately. The two realms are distinguished by different screen sizes, input mechanisms and processing capabilities [25]. Thus, the usage of mobile devices and services may appear challenging for beginners. Thus, we assume that the following hypothesis will apply: H8: Mobile experience is positively associated with perceived behavioural control. ME1: I think that I am an experienced user of mobile phones. ME2: I think that I am an experienced user of mobile services. ME3: I think that I am an experienced user of small mobile devices such as PDAs, Communicator, etc. Finally, the Behavioural Intention (BI) to adopt the Community-centric Music Voting service is defined by using the combination of three factors: Attitude (ATT), Subjective Norm (SUN), and Perceived Behavioural Control (PBC). ATT describes a person s feeling or perception towards the usage of the CCMV service [12]. SUN describes the social influence, experienced by the user, that may have an effect on his or her intentions to take the CCMV service into use [11]. In turn, PBC describes a person s perception of how easy or difficult it is to use the CCMV service [12], and it also reflects whether the individual possesses the resources and skills needed to use the service. Thus, we assume that the following hypotheses will apply: H9: Attitude is positively associated with behavioural intention. H10: Subjective norm is positively associated with behavioural intention. H11: Perceived behavioural control is positively associated with behavioural intention. ATT1: The Music Voting service was, in every way, a positive experience for me. ATT2: I followed the success of the song I had voted for with interest. ATT3: As a whole, the Music Voting service was a good idea. SUN1: The usage of the Music Voting service would increase the affinity of the bar s clientele. SUN2: I would like to use the Music Voting service together with my friends. PBC1: It was not hard for me to use the Music Voting service with the provided mobile device. PBC2: I completely master the use of the Music Voting service. PBC3: Usage of the Music Voting services was effortless to me. BI1: I intend to use the Music Voting service if it becomes available. BI2: I intend to use the Music Voting service frequently if it becomes available.

5 3.2 The Special Interests Groups In addition to the research model based on the DTPB, we have defined three special interest groups that presumably have different kinds of expectations and attitudes towards the CCMV service. The three groups consists of test users that (1) are enthusiastic about new technology, (2) are enthusiastic about listening to music, and (3) use their mobile phones for listening to music on a daily basis. Not every user was a member in a special interest group; only those users who fulfilled the specified criteria, described below, were considered as part of a special interest group. It should also be noted that the groups were partly overlapping. People who are very enthusiastic about new technology. Test users that had either regularly used the Internet with their mobile phone or had installed several applications to their mobile phone were included in this special interest group. These test users also claimed that among their friends they are usually the first ones to try new mobile services. Moreover, their technology-related usage habits are usually not influenced by other people. Therefore, we presuppose that these test users are more excited to try, but also more skilful in using, the CCMV service. Often, this kind of people are called early adopters [11]. People who are very enthusiastic about listening to music. Test users that had daily spent more than four hours for listening to music were included in this special interest group. Since the music plays an important role in their lives, we presuppose that these test users are excited to use and have a more positive attitude towards the CCMV service. People who use their mobile phone for listening to music. Test users that had used their mobile phone for listening to music several times a week were included in this special interest group. These test users are typically experienced in using the mobile music services. Thus, we presuppose that they have higher expectations and demands for the CCMV service. 4. The User Evaluation This Section describes in detail how the user evaluation of the CCMV service was conducted. 4.1 Testing Environment In May 2009, we conducted a user evaluation in a real usage environment, where user communities were formed by the customers of a bar that was situated in the city centre of Oulu, Finland. The bar had the DJOnline system that was used for creating appropriate voting events according to the musical style of the bar. Altogether 53 test users participated in the evaluation. The test users were equipped with Nokia N810 Internet Tablets that were connected to the Internet through a WLAN base station (IEEE b/g). Due to the limited capacity of the used WLAN station, the test users were divided into five separate groups. Thus, in each testing event, the user communities were formed approximately by 10 test users. 4.2 Collection of Data The background information about the test users was collected beforehand using an online questionnaire. The online questionnaire mainly contained questions about the test users consumption habits related to mobile services and music in general. Based on this questionnaire, the demographics profile was created and the special interest groups were defined. After the user evaluation, the experiences gained were collected using a quantitative questionnaire with a 5-point Likert-scale, ranging from strongly disagree to strongly agree. In addition, short interviews were conducted focusing on the usage experience and ideas for future improvements. 4.3 Demographics All 53 test users were students from various faculties of the University of Oulu. Slightly unexpectedly, 58.5% of the test users were female and 41.5% male, since usually technology-oriented events raise more interest among men. The average age of the test users was around 24 years. The basic demographic profile of the test users is shown in Table 1. Table 1. The Demographic Profile. Frequency Percentage Gender Male 22 41,5 Female 31 58,5 Total ,0 Age , , , ,8 Total ,0 Occupation Student ,0 Total ,0

6 5. The Results of the User Evaluation This Section presents the results achieved with the two evaluation approaches. 5.1 The Decomposed Theory of Planned Behaviour Before examining the claimed hypotheses, the validity of the research model was tested. The internal consistency of the constructs was evaluated with the Cronbach s Alpha coefficients that are used for examining how well a set of items (i.e. statements) measures a single construct. According to [26], the Cronbach s Alpha coefficients should not be lower than The acceptable level of Cronbach s Alpha coefficients has been argued, however. Some studies (e.g. [27]) indicate that the acceptable level should be as high as In our study, the Cronbach s Alpha coefficients stay above 0.70, which indicates not excellent, but an acceptable level of internal consistency. The construct validity was tested by conducting a factor analysis. The construct validity examines whether the set of items (i.e. statements) really measure what they intend to measure [9]. In this exploratory factor analysis, we used an eigenvalue of more than 1.0 as a determinant criterion. As a rule of thumb, the factor loading should exceed the value 0.60 [14]. In our study, the factor loadings stay mainly above 0.70, which indicates a significant level of construct validity. Because of the poor results in either internal consistency or construct validity testing, we needed to drop some items. The results of the validity testing are shown in Table 2. Table 2. Internal Consistency and Construct Validity Testing. Constructs Loadings Cronbach s Alpha Items eliminated PPF PUF PUF4 IM PPL PPL4 FI CI CI3 IE ME ATT ATT2 SN PBC PBC1 The claimed hypotheses were tested by conducting multiple regression analyses with SPSS In the research model, the overall effect of each determinant of each construct was tested. The values of the standardized coefficient β, t-value, significance, and the adjusted R-square are shown in Table 3. Table 3. Analysis of the Hypotheses. Hypotheses β t-value Sig. H1: PPF to ATT Adjusted R-Square H2: PUF to ATT H3: IM to ATT H4: PPL to ATT H5: FI to SUN H6: CI to SUN H7: IE to PBC H8: ME to PBC H9: ATT to BI H10: SUN to BI H11: PBC to BI The hypotheses H1-H4 examine the constructs affecting ATT. According to the regression analyses, only H1 is significantly supported. The findings suggest a strong positive association between PPF and ATT. This highlights the fact that the PPF greatly influences the attitude towards the entertainmentoriented mobile services, which has also been earlier discovered in the context of WWW [14]. PUF and IM seem to have a positive association with ATT, but not significant enough in the statistical terms. In any case, the test users thought that the CCMV service was quite useful to them and a nice addition to the service category of a bar. Moreover, they thought that it was trendy to use the service. As predicted, PPL seems to be negatively associated with ATT. In general, the test users were really confident that the costs related to the CCMV service should be low and likely included in the costs of the other services of a bar. The hypotheses H5-H6 examine the constructs affecting SUN. In this case, only H6 is significantly supported. The findings suggest a strong positive association between CI and SUN. Thus, it is important that the CCMV service does not collide with the behavioural models and norms of the local society. This notion is contrary to the results gained in our earlier user evaluation [9], where FI was seen as a clearly more influential factor towards SUN than CI was towards SUN. It has to be noted that the earlier user evaluation was conducted in a laboratory environment, and especially with small groups of friends. Therefore, both the fictitious evaluation

7 settings and the strong presence of friends have presumably influenced the results. After all, the CCMV service was seen to be very suitable in the bar environment, which indicates that there should not exist strong cultural barriers to the wider-scale adoption. FI also seems to have a positive association with SUN, but not significant enough in the statistical terms. The hypotheses H7-H8 examine the constructs affecting PBC. Now, only H8 is significantly supported. The findings suggest a strong positive association between ME and PBC, whereas the positive association between IE and PBC is statistically insignificant. As predicted, the young adults have become so familiar with the Internet technology that it is not seen as an obstacle to the service adoption [9]. On the contrary, the test users did not treat themselves as experienced consumers of mobile services, which especially highlights the need for user-friendly and straight-forward mobile user interfaces. Finally, the hypotheses H9-H11 examine the constructs affecting BI. H10 and H11 are significantly supported, but H9 is slightly unexpectedly not supported. The findings suggest a strong positive association between SUN and BI and also between PBC and BI. The usage intentions towards to the CCMV service are greatly influenced by the behavioural models and norms of the local society. As in all community-centric services, their value lies in the number of the members, and thus, it is essential that the CCMV service attracts large populations. In addition, the usage intentions are influenced by the users perceptions whether they are confident and able to use this kind of novel community-centric mobile services. In this study, there does not appear to be any clear association between ATT and BI. According to the results, the test users really enjoyed using the CCMV service, and thus, did not presumably see the content of the service as an obstacle for the adoption. In our earlier user evaluation [9], ATT had a negative effect on BI, which was a result from a strong negative association between PUF and ATT. Now, due to the improved version of the CCMV service and the real usage environment, the perception of usefulness was more positive. The values of the standardized coefficients, embedded in the research model, are shown in Fig. 3. An asterisk (*) denotes that the hypothesis was not significantly supported. Figure 3. The Research Model with the Standardized coefficients. 5.2 The Special Interest Groups Three special interest groups were also formed in order to discover whether (1) enthusiasm on technology, (2) enthusiasm on listening to music, or (3) usage of a mobile phone as a music consumption medium had a significant effect on user perceptions of the CCMV service. Table 4 combines the answers (in a 5-point Likert-scale) to the statements that reflect the differences in the perceptions of the three special interest groups. A little bit surprisingly, the first group did not see the device-independency of mobile services to be as important as the third group. This indicates that the technology-oriented test users tend to be more committed to specific platforms and expect the mobile services to support them. In comparison, the test users that use their mobile phones for listening to music presumably expect that they are able to enjoy the mobile services device-independently, in the same way as they currently enjoy their mobile music. All three groups enjoyed using the service equally much, even though we expected that the musicoriented test users would be the most excited ones. Moreover, all three groups thought that the service was quite useful and would presumably affect their choice of a bar. In both cases, the perceptions of all three groups were more positive than the ones of an average test user. All test users, but especially the first and the second group, were really confident that they did not want to pay for using the service. However, at the same

8 time, these two groups considered the usage expenses a slightly smaller barrier to the service adoption than the others. This indicates that they predict the service to be more valuable to them than to an average user. Table 4. Differences between Perceptions of the three Special Interest Groups. Statement It is important to me that mobile services are deviceindependent. PPF1: It was fun to vote for music and to follow the progress of the voting event. Grp. 1) Grp. 2) Grp. 3) AVG PUF1: The Music Voting service is useful to me PUF4: The availability of the Music Voting service would positively influence my choice of a bar. PPL1: I would not want to pay for the usage of the Music Voting service. PPL2: The expenses for the Music Voting service would be a barrier to me ATT1: The Music Voting service was, in every way, a positive experience for me. ATT3: As a whole, the Music Voting service was a good idea FI1: My friends would assume that I used this kind of Music Voting service. CI1: The Music Voting service would fit into bars well CI2: Using the Music Voting service in bars would not be generally disturbing. ME2: I think that I am an experienced user of mobile services. PBC1: It was not hard for me to use the Music Voting service with the mobile device provided. BI1: I intend to use the Music Voting service if it becomes available. BI2: I intend to use the Music Voting service frequently if it becomes available All three test groups liked the service idea, but as we expected, the third group was a little bit more sceptical about it. Among all three groups, there existed a significant gap between the perceptions of the service idea and the usage experience, which clearly indicates that the service did not completely fulfil their expectations. Thus, the service can be further developed by exploiting the experiences gained in the user evaluation. According to the interviews, the usage experience was mostly hindered by the faults in the operation of the service, the cumbersome mobile user interface and the limitations in the functionality of the service. For instance, the number of songs per voting event was, in this test event, limited to only two. All three groups thought that the service would fit the bars very well and the usage would not disturb the atmosphere. However, all three groups considered the service a little bit more distracting than an average test user. All three groups clearly regarded themselves as more experienced with mobile services than an average test user. Nevertheless, the second group, in particular, was facing more difficulties when using the service than an average test user. This can be explained by the fact that either the used mobile device was more unfamiliar to them or they considered the user interface more cumbersome. 6. Discussion and Conclusion This paper presents the user evaluation results of the CCMV service. The user evaluation was conducted in May 2009 with 53 test users. The user evaluation took place in a real usage environment, where the user communities were formed by the customers of a bar. The research data was collected with both quantitative questionnaires and short interviews. The evaluation results were analyzed with two separate approaches: using the Decomposed Theory of Planned Behaviour and the analysis of answers after categorizing some of the test users into special interest groups. In this paper, the defined research approaches were utilized in the evaluation of a single community-centric mobile service. However, the research approaches could also be applied in studies focusing on the evaluation of other community-centric mobile services. The user evaluation results indicate that the cultural-related habits and norms, as well as the earlier experiences with mobile services, are the two most important factors affecting the usage intentions towards the CCMV service. Thus, before the market entry, a strong focus should be on designing the actual end-user application in such a way that the usage situations would easily merge into the customary behavioural models of the people that enjoy their leisure time in entertainment premises such as bars. According to the interviews, this would require an intuitive and simple end-user application that is compatible with the most of the mobile phone models. In addition, a clear consensus existed among the test users that the service itself should be free-of-charge, but at the same time, the test user did not oppose adding the service expenses to the prices of other

9 services provided by the entertainment premises. Thus, it is essential to define the business logic in a way that the customers do not feel they are paying for the use of the service. The added revenue could come from the polished imago that leads to an increase in the number of customers or, for instance, from advertisement or revenue-sharing business models. In an example of the revenue-sharing business model, the information about a song could be accompanied with a link to a mobile music store and when the user purchases the song, the music store s revenue is shared with the bars. In addition to bars, the test users perceived that the service would be suitable in nightclubs, cafeterias and waiting areas. As a predicted group of early adopters, we identified people with interests in technology and music, as well as people that use their mobile phone as a music consumption medium. Acknowledgement The financial support from the National Technology Agency of Finland (Tekes) is gratefully acknowledged. In addition, this work has been supported with scholarship grants by Foundation of Technology, Nokia Foundation, Foundation of Walter Alhström, Research Foundation of HPY, and Research and educational Foundation of TeliaSonera. Finally, the authors would like to thank the fellow researchers in the ITEA2 Expeshare project consortium. In addition, support from Kutalab Ltd. is gratefully acknowledged. 7. References [1] M. Kesti, A. Ristola, H. Karjaluoto, T. Koivumaki, Tracking consumer intentions to use mobile services: empirical evidence from field trial in Finland, E- Business Review, IV, 2004, pp [2] L. Gaye, L.E. Holmquist, F. Behrendt, A. Tanaka, Mobile music technology: Report on an emerging community, Proc. the conference on New Interfaces for Musical Expression, Jun. 2006, pp [3] E. Nettamo, M. Nirhamo, J. Häkkilä, A Cross-Cultural Study of Mobile Music - Retrieval, Management and Consumption, Proc. the OZCHI conference, Nov. 2006, pp [4] D. Schnitzer, T. Pohle, P. Knees, G. Widmer, One- Touch Access to Music on Mobile Devices, Proc. the conference on Mobile and Ubiquitous Multimedia, Dec. 2006, pp [5] A. Bassoli, J. Brewer, K. Martin, P. Dourish, S. Mainwaring, Underground Aesthetics: Rethinking Urban Computing, Pervasive Computing, IEEE, vol. 6, no. 3, 2007, pp [6] K. O Hara, M. Lipson, M. Jansen, A. Unger, H. Jeffries, P. Macer, Jukola: democratic music choice in a public space, Proc. the 5th Conference on Designing interactive Systems: Processes, Practices, Methods, and Techniques, Aug. 2004, pp [7] H. Kukka, R. Patiño, T. Ojala, UbiRockMachine: a community-driven Music Distribution and Consumption for Shared Urban Spaces, Proc. the conference on Multimedia and Expo, Jun. 2009, pp [8] J. Kjeldskow, J. Paay, Public Pervasive Computing: Making the Invisible Visible, Computer Journal, vol. 39, no. 9, 2006, pp [9] T. Koskela, J. Julkunen, V. Keränen, N. Kostamo, M. Ylianttila, User Experiences on a Community-based Music Voting Service, Proc. 3rd International Workshop on Service Intelligence and Computing (SIC2009), Jul. 2009, pp [10] T. Koskela, J. Julkunen, J. Korhonen, M. Liu, M. Ylianttila, Leveraging Collaboration of Peer-to-Peer and Web Services, Proc. 2nd IEEE conference on Mobile Ubiquitous Computing, Systems, Services and Technologies, Sep.-Oct. 2008, (CD-ROM). [11] T.S.H. Teo, and S.H. Pok, Adoption of WAP-enabled mobile phone among Internet users, Omega, Elsevier, vol. 31, no. 6, 2003, pp [12] M.K. Chang, Predicting Unethical Behavior: A Comparison of the Theory of Reasoned Action and the Theory of Planned Behavior, Journal of Business Ethics, Kluwer Academic Publishers, vol. 17, 1998, pp [13] J. Moon, and Y. Kim, Extending the TAM for World- Wide-Web context, Information and Management, vol. 38, no. 4, 2001, pp [14] J. H. Cheong, and M.-C. Park, Mobile internet acceptance in Korea, Internet Research, vol. 15, no. 2, 2005, pp [15] M.A. Atkinson, and C. Kydd, Individual characteristics associated with world wide web use: an empirical study of playfulness and motivation, Database for Advances in Information Systems, vol. 28, no. 2, pp [16] F. Davis, Perceived usefulness, perceived ease of use, and user acceptance of information technology, MIS Quarterly, vol. 13, no. 3, pp [17] J. Nielsen, Usability Engineering. Academic Press, 1993, 358 p. [18] E.M. Rogers, Diffusion of innovations, 3rd edition. The Free Press, 1983, 453 p. [19] V.A. Zeithaml, Consumer perceptions of price, quality and value: a means-end model and synthesis evidence, Journal of Marketing, vol. 52, 1988, p [20] R.L. Thompson, C.A. Higgins, and J.M. Howell, Influence of experience on personal computer utilization: testing a conceptual model, Journal of MIS, vol. 11, no. 1, 1994, pp [21] N. Singh, and H. Matuso, Measuring Cultural Adaption on the Web: A Content Analytical Study of U.S and Japanese Web Sites, Journal of Business Research, vol. 57, iss. 8, 2004, pp [22] D.K. Tse, R.W. Belk, and N. Zhou, Becoming a consumer society: a longitudinal and cross-cultural content analysis of print ads from Hong Kong, The People's republic of China, and Taiwan, Journal of Consumer Research. vol. 15, iss. 4, Mar. 1989, pp

10 [23] G. Hackbarth, V. Grover, and M.Y. Yi, Computer playfulness and anxiety: positive and negative mediators of the system experience effect on perceived ease of use, Information and Management, vol. 40, no. 3, 2003, pp [24] J. Dearing, and G. Meyer, An exploration tool for predicting adoption decisions, Science Communications, vol. 16, no. 1, 1994, pp Conference on Intelligent User Interfaces, Jan. 2006, pp [26] J.C. Nunnally, Psychometric theory, McGraw-Hill, 1978, 701 p. [27] D. Straub Validating instruments in MIS research, MIS Quarterly, vol. 13, 1989, pp [25] R. Bridle, and E. McCreath, Inducing shortcuts on a mobile phone interface, Proc. the International