ࡱ> '` 0bjbjLULU ..?.?z    8Z nrr:ch GGG  yyyG  |yGyyR  f aY~ \0תD< Fy^GGGGCK3K       Public Dialogs in Social Network Sites: What is their Purpose? Mike Thelwall, David Wilkinson Statistical Cybermetrics Research Group, School of Computing and Information Technology, University of Wolverhampton, Wulfruna Street, Wolverhampton WV1 1SB, UK. E-mail: m.thelwall@wlv.ac.uk, d.wilkinson@wlv.ac.uk Tel: +44 1902 321470 Fax: +44 1902 321478 Social network sites (SNSs) like MySpace and Facebook are important venues for interpersonal communication, especially amongst youth. One way in which members can communicate is to write public messages on each others profile, but how is this unusual means of communication used in practice? An analysis of 2,293 public comment exchanges extracted from large samples of US and UK MySpace members found them to be relatively rapid but rarely used for prolonged exchanges. They seem to fulfil two purposes: making initial contact and keeping in touch occasionally, such as at birthdays and other important dates. Although about half of the dialogs seem to exchange some gossip, the dialogs seem typically too short to play the role of gossip-based social grooming, for typical pairs of Friends, although close Friends may well still communicate extensively in SNSs with other methods. Introduction Social network sites (SNSs) seem to have been adopted as key social device by a broad range of society in many countries (boyd & Ellison, 2007), particularly teenagers (e.g., Lenhart, Madden, Macgill, & Smith, 2007; Pew Research Center for the People & the Press, 2007) and those in higher education (e.g., Tufekci, 2008b). Improved understanding of the new phenomenon of SNS communication is therefore needed for those seeking to help users, including communication skills teachers, people working with socially struggling pupils and students, and psychiatrists and social workers seeking to identify social or psychological problems through online behaviour (as achieved for offline text, e.g., Pennebaker, Mehl, & Niederhoffer, 2003; Pennebaker, Mayne, & Francis, 1997). In addition, improved knowledge of how SNSs work could aid the identification of threatening or inappropriate online activities (a goal of the project funding the current research) and creative industries seeking to accurately portray or exploit contemporary social media use. Moreover, SNS communication seems to be informal and to often (but not always) occur between genuine friends and so may potentially offer valuable evidence about how friendship communication occurs, and what makes it succeed or fail. For all these varied reasons, basic research is needed to gain broad understanding of how various forms of SNS communication work. In this paper the focus is on public interpersonal communication between SNS members in the form of public comments posted on each others profile page, focussing on one of the most used sites, MySpace. These comments seem to be typically informal short messages (Thelwall, 2009b) between Friends (boyd, 2006) and are apparently an essential component of the popularity of SNSs (Kim & Yun, 2007). Nevertheless, there is no quantitative evidence for the importance of public comments since statistical analyses of SNSs based upon questionnaires or interviews tend to emphasise purposes for using SNSs rather than teasing out the specific communication methods used (e.g., Joinson, 2008; Lampe, Ellison, & Steinfield, 2008; Raacke & Bonds-Raacke, 2008). This study examines basic characteristics of dialogs between Friends in MySpace expressed through public comments. This is not only a preliminary step towards future research to fully understand the purposes for which social network comments are used, but is also to assess what they can reveal about the success of SNSs. In particular, since social grooming through gossip has been hypothesised to be an important factor in the success of SNSs (Donath, 2007; Tufekci, 2008b), this study assesses whether gossip-like exchanges are common in public comments. Literature review This section examines research relevant to communication in MySpace, starting with a general theoretical overview of computer-mediated communication (CMC) research. General CMC Theories In terms of inter-personal communication, the Internet and electronic devices seem to be used in varied and complementary ways by individuals to help maintain their personal and professional networks. For example, posting and commenting on personal videos in YouTube can be a strategy to maintain contact with distant relatives (Lange, 2007). Early theories of electronic communication assumed that bandwidth was a key consideration, especially for ambiguous communication, because sensual properties such as media richness (Daft & Lengel, 1984; Kock, 2005) or the ability to create a sense of social presence (Biocca, Harms, & Burgoon, 2003; Short, Williams, & Christie, 1976) were necessary for effective communication. Later research showed that even very simple communication tools could be effective for ambiguous messages. This is clear, for example, from the widespread success of text messaging: exchanging messages can build a feeling of affinity between participants (Hancock, Landrigan, & Silver, 2007; Nardi, 2005; see also Riva, 2002) because user inventiveness (Walther, 1992), such as through emoticons, can partly circumvent the limitations of simple message formats. This is the essence of Walthers (1992) Social Information Processing theory, which therefore emphasises the ability of users to circumvent technical limitations of devices. Hence, even short text-based comments in MySpace should not be regarded as necessarily conveying little information. Nardis (2005) work-based theory of the key dimensions of (online or offline) connection includes: affinity (creating a feeling of affinity between participants), commitment (expressing commitment to the relationship) and attention (engaging the attention of another for an important communication). Of these, affinity seems directly relevant to social networks. Humans use various methods to create affinity, including touch (hugging, kissing, patting, shaking hands), and informal conversation (Nardi, 2005). Nardis analysis of instant messaging (IM) showed that touch was often simulated through descriptions (e.g., hugs, X for a kiss, O for a hug) (see also Derks, Bos, & von Grumbkow, 2008). Informal communication could often be relatively content-free, such as exchanges of simple greetings or discussion of the weather, known as phatic communion (Malinowski, 1923). From this, it seems likely that MySpace comments would frequently convey emotion or simple greetings for the purpose of maintaining contact. Most CMC research seems to have focused on communication in a non-recreational purposive context, such as computer-supported cooperative work or electronic learning environments, or on communication in relatively self-contained online environments, such as internet discussion forums or chatrooms (Walther & Parks, 2002; Whittaker, 2003), although Walther and Parks (2002) do take into account offline communication. SNSs do not easily fit these descriptions and so existing CMC theories should be applied to them cautiously. In contrast, the useful notion of networked individualism (Wellman et al., 2003), is derived from contrasting pre-Internet offline communities particularly geographically-oriented groups with communication patterns generated by various Internet technologies. In particular, the social affordances of the new technology help to support a move away from traditional styles of community towards a more interconnected, less cohesive society. The networked individual may be part of geographically-organised communities but is also likely to use a variety of electronic means to maintain connections with a much wider group of people that would not form a community in the traditional sense. The structures formed by these connections are likely to be more flexible and to incorporate much looser ties than appropriate for the traditional concept of community. The motivation for networked individualism may lie not only in the affordances of the technology but also the importance of individuals networks to their personal status (Larsen, Urry, & Axhausen, 2006). SNSs and networked individualism SNSs are Web sites that allow members to create their own profile containing personal information, to connect to other members of the community as Friends or equivalent, and to display lists of Friends for other users to browse (boyd & Ellison, 2007). Well-known SNSs include MySpace and Facebook but there are many different types and national variations in the most popular (boyd & Ellison, 2007; Thelwall, 2009c). SNSs tend to allow members to communicate with each other in many ways. In particular, users can express or broadcast their identity by the appearance or content of their profile (boyd & Heer, 2006; Liu, 2007), including music, video and images. Members can also normally communicate privately with email or instant messages and can broadcast personal messages to Friends in the form of public comments written on Friends profiles. Hence, the typical SNS (including Facebook and MySpace) is ostensibly a richly-featured multimedia hub for communication between Friends although the term Friend could actually mean close friend, more distant friend, acquaintance or nothing (boyd, 2006; Thelwall, 2008). Do SNSs typically support the formation of specific, relatively self-contained communities, as forums seem to, or are they more commonly one of a range of means used by networked individuals to manage their interpersonal communication? Whilst both are almost certainly widespread uses of SNSs, the latter seems to be more common and is supported by current research. Although SNSs contain the public connections necessary to form a community, most research suggests that the major SNSs integrate into the lives of members, with online Friends normally being known offline (boyd, 2004, 2006; Ellison, Steinfield, & Lampe, 2007; Lampe et al., 2008) and usage patterns reflecting offline activity rhythms (Golder, Wilkinson, & Huberman, 2007). A study of US teenage MySpace members argued that the networks in MySpace did not form groups (or a community in the traditional sense) but should be thought of as ego-centred (boyd, 2006). This is supported by research emphasising the use of Facebook to manage relatively loose networks of friends (weak ties) (Steinfield, Ellison, & Lampe, 2008). The more specialist SNSs that seem to support online community formation, such as BodySpace.com (Ploderer, Howard, & Thomas, 2008) and perhaps Last.FM (Baym, 2008), are much less successful than general SNSs like MySpace and Facebook. In summary, Friend connections should be thought of as primarily meaningful from the perspective of individuals rather than groups. In fact boyd casts MySpace as part of a significantly weaker connecting mechanism. Using the concept of networked publics, meaning transient, loosely connected audiences, she emphasises that MySpace profiles are searchable so can be read by a wide potential audience. An awareness of this can lead members to perceive of their profile, if not their public comments, as partly a performance to an unknown audience rather than specific communication with friends and acquaintances (boyd, 2006; boyd, 2008). This idea is consistent with other research that emphasises the performance aspects of MySpace (Liu, 2007; Tufekci, 2008a). SNSs and social grooming Related to the kinds of connections fostered by SNSs is the issue of why they are so popular. Donath (2007) argues that they support the development of trust, identity projection and risk-taking, which are important in a social context. This explains why they can be successful but not why they are successful. One promising reason is that SNSs are a forum for the fundamental human activity of social grooming maintaining relationships with others through gossip or other minor activities (Donath, 2007; Tufekci, 2008b). Social grooming has been claimed to be an important human activity from an evolutionary perspective because it helps people to build trusting relationships that could be useful in times of crisis (Dunbar, 1998). SNS Friending and communication mechanisms offer a technological way to allow individuals to maintain much larger networks of relationships than the 150 that had previously been normal in everyday life (Donath, 2007; Dunbar, 1998). This is because the technology makes communication at a distance possible and efficient. There is indirect evidence for this in the sense that people who do not join SNSs seem to be those that are less interested in gossip (Tufekci, 2008b) and from one study that found 96% of college students tested used SNS to keep in touch with friends (Raacke & Bonds-Raacke, 2008). The social grooming hypothesis incorporates relationship building through social exchanges but there is an important outcome of the gossip aspect of social grooming that does not necessarily entail building relationships: social information gathering (Baumeister, Zhang, & Vohs, 2004). Gossip is loosely defined as rumor or report of an intimate nature or chatty talk according to the Merriam-Webster online dictionary and therefore implies more information exchange than in phatic communion. Although dictionary definitions have tended to emphasise the role of gossip in spreading negative information, it is now acknowledged that gossip is also frequently used for positive messages (Baumeister et al., 2004). Whilst the exact function of gossip is debated, it seems to be used not only for finding out about the gossip partner, but also for sharing information about others (Wert & Salovey, 2004), and discovering the cultural norms of society (Baumeister et al., 2004). There is clear evidence of this kind of social learning from one large study of Facebook in terms of learning how to actively use the SNS itself (Burke, Marlow, & Lento, 2009) and from another study indicating that information browsed in Facebook Wall postings was used to evaluate those commented on (Walther, Van der Heide, Kim, Westerman, & Tong, 2008). In primitive societies this intelligence may have been gained almost exclusively by gossip or direct observation but can also be achieved in other ways that suggest that the human need for social information gathering is independent of the need to gossip. For example, reading newspaper gossip columns and gossip magazines is a popular impersonal form of social information gathering. SNSs can support social information gathering either via gossip exchanging public or private messages or by unobtrusively browsing Friends profiles and public comments. If the latter is more common, then the social grooming hypothesis for the popularity of SNSs overstates the importance of relationship-building and underestimates the importance of information gathering. Evidence against the social grooming hypothesis and in favour of the pure social information gathering alternative can be found in the widespread popularity of finding out about others in Facebook (Lampe, Ellison, & Steinfield, 2006), later described as social searching and surveillance (Joinson, 2008), although the prevalence of positive emotion in MySpace comments suggests a supportive rather than an information dissemination role (Thelwall, Wilkinson, & Uppal, in press). In addition, the viewing of others personal pictures (Raacke & Bonds-Raacke, 2008) could also be part of social information gathering, although picture commenting is also common (Haddon & Kim, 2007) and this could be a form of gossip. Despite the hypothesised importance of gossip, the role that direct communication in SNSs plays in social information gathering is unknown. For example, is gossip-type information primarily exchanged directly or is it primarily exchanged indirectly, by profile browsing? It seems likely that closer Friends would exchange more direct gossip, perhaps using private messages, since one indication of friendship is the sharing of private information (Tardy & Dindia, 2006), particularly for females (Aukett, Ritchie, & Mill, 1988). It also seems possible that distant Friends would use profile-browsing instead of direct gossip, or would perhaps not indulge in information gathering. Close Friends may also use indirect methods instead of just private messages, however, since closer ties can generate multiple communication channels (Haythornthwaite & Wellman, 1998). Moreover, a majority of registered MySpace members (62%) in one study had not received any comments on their profile (Parks, 2008) and MySpace, comments seem to be predominantly trivial and short (Thelwall, 2009b), suggesting that little social information is exchanged overall. Research questions The research questions are descriptive, as is appropriate for early exploratory basic research into a new online phenomenon (Thelwall, Wouters, & Fry, 2008), but are also relevant to the role of public comments in social grooming in MySpace. For the latter, the questions support a discussion of whether the typical public comment dialog between Friends is used for gossip in the form of extended message exchanges. This is based upon the assumption that gossip requires several messages to be exchanged because initial communication between people is typically phatic, which a qualitative analysis of the MySpace data supports. Hence extended communication seems to be needed for genuine exchanges of social information. What is the typical length of public MySpace comment dialogs? The predominance of Short dialogs would indicate that they are not normally used for extended chatting or exchanges of gossip. How rapid are MySpace public comment exchanges? If they are almost instant like chat or slower, like email, then this would suggest the type of communication present in them. What are the common temporal patterns in comment dialogs? Knowledge of typical patterns would suggest possible uses, such as whether they were likely to be used for gossip. What proportion of public dialogs contains gossip in the sense of exchanging information about the participants or others? Methods In order to investigate dialogs in MySpace, three different samples of members were used. First, a systematic sample of long term members was taken: those who had joined up to 3 July 2007. This sample was taken on the basis of member IDs, taking about every 5,000th member for a total of 40,000. This sample contains both old and newer members and there may be differences in patterns of behaviour by membership length, so two additional samples were taken: members joining on 3 July 2006 (30,000 selected by member ID in steps of 15) and members joining on 18 June 2007 (30,000 selected by member ID in steps of 15). From each of these three sets two subsets were selected: UK-based members with public profiles and US members with public profiles. The restriction to a single country ensures a relatively homogenous subset and the UK and US have the most MySpace members (Thelwall, 2008). Only members with public profiles were taken as the dialogs of other members are not visible to non-Friends. For each member, a Friend was chosen at random using a random number generator to select from all Friends that (a) commented on the members profile and (b) had a public profile. All comments between such pairs of Friends were recorded along with their creation time and date. Dialogs were only used if both partners participated throughout, which was operationalised as: (a) at least three comments were exchanged, and (b) both partners contributed at least one of the first four and one of the last four comments. The resulting dialogs formed the raw data, which was collected between November 2008 and December 2008. See Table 1 for the sample sizes. Here is an example of a dialog of five comments between members A and B extracted as above (comments modified for anonymity). 9 Jun 2007 15:19 A->B: safee pudu fuk knows ow 2 use dis ting lol 9 Jun 2007 15:22 B->A: hey big boy 6 Mar 2008 14:22 A->B: HAPPY B-DAY. Hope u n boyz r all well n gud. 7 Mar 2008 2:36 B->A: [Image Message] 9 Dec 2008 9:48 B->A: Mezzy xmas [Embedded Image] geeza To address the first two research questions, simple programs were written to count comments per dialog and the gaps between comments within dialogs. A more complex procedure was developed to address the third research question. Based upon a manual exploration of the data, dialogs seemed to frequently consist of periodic short exchanges. To capture this, dialogs were conceptualised as comprised of a series of exchanges punctuated by gaps. A gap was defined as a period of at least 7 days without comments. Exchanges were the groups of comments between gaps (or at the start or end of the whole dialog). These were classified into three types, as follows, with examples given below (comments modified for anonymity). A long exchange has at least 10 comments, with at least two comments from each participant. A medium exchange has as at least 4 comments, with at least one from each participant. A short exchange or monolog covers everything else: either all comments from one participant or up to 3 comments. The long exchange below took 10 minutes for all except the last comment, which was posted 14 hours later. There were no other public comments between these two and the exchange also illustrates possible communication channel switching at the end. C->D: hi wats happnin? who this? D->C: hi im sara i thought u are really hott , sry if i freeked u out, Sara xox C->D: hey noo u diddnt...ur pretty hott urself.. so watt up? D->C: jus at a party .. n u? C->D: do u hav a gf? D->C: whyy r u on here if u r at a paarty? C->D: im not i wuz at a party b4 .. jus got home D->C: ooohh...coolio... C->D: so do you? D->C: yah i do.... C->D: you got a bf? D->C: awwww that sucks whos the luckyy girl? C->D: da 1st on my top 12 D->C: wow i want 2b her C->D: so... you have sn? D->C: yah.... turkeyweazel3493...you? The medium exchange below took 8 hours and also illustrates organising offline activities online. E->F: Madonna F->E: You going? E->F: I want to see Madonna but don't know yet F->E: I'm saving my money until Tom is back. The short exchange or monolog below took three days. Comment 3 seems to refer to communication not in the form of a MySpace comment. Many of this type of exchange were birthday greetings, with or without a reply. G->H: Whattup Jenny how u been? H->G: hey sunshine, hope you are good. at school again (blahh blahh blahh) Have a good day ttyl8r :) H->G: good good, and urself?? where are you guys now?? anything new? ohh did you get a pressie for your bro? talk to ya soon k have a good day In an attempt to differentiate between different apparent uses for MySpace comment dialogs, three different types of dialog were distinguished on the basis of the gaps and types of exchanges present. No basis is claimed for these definitions in terms of prior theory: they are merely labels for a convenient differentiation between apparently different types of dialog based upon a preliminary qualitative analysis of the data. Long exchange(s) dialog: A dialog with at least one long exchange (with or without gaps or additional exchanges). Multiple exchanges dialog: A dialog with gaps but without long exchanges. Single exchange dialog: A dialog with neither gaps nor long exchanges. See the Appendix for examples of these dialog types. All dialogs were automatically classified using the above scheme. This labelling approach is useful to distinguish and describe patterns in an otherwise huge mass of undifferentiated data and is motivated by a similar labelling approach in a different online communication context (Gruhl, Guha, Liben-Nowell, & Tomkins, 2004). The five comment example involving A and B above can be used to illustrate the terminology. It has two gaps of over seven days and is therefore classed as three consecutive exchanges: two comments, two comments and one comment. All three of these sets have less than 4 comments so are classed as a short exchange or monolog. The example is therefore of the type multiple exchanges dialog because it has gaps but no long exchanges. To further illustrate the definitions, if all comments had been made on the same day then the dialog would have had no gaps, would be classed as a medium exchange and as a single exchange dialog. If the exchange of five comments had been repeated twice on the same day, then it would be a single long exchange and the dialog would therefore be classed as a long exchange(s) dialog. For the final research question, a random sample of 100 dialogs was selected and each dialog was checked to see whether it contained information about the participants or others. In addition, the dialogs were checked for two other purposes: coordination of offline activities and the exchange of other types of information. Results Table 1 reports basic statistics for the six data sets. The relatively low total number of dialogs, given that the total of the three initial samples is 100,000, reflects several factors that combine to eliminate the vast majority of candidates: a significant minority of members are not from the US or UK; two thirds of members do not have public profiles; one third of members do not actively use MySpace; many members do not use the comment feature extensively with Friends; data collection errors ruled out a proportion of the results, and some members were eliminated because dialogs were initially selected for them that did not have three comments with contributions from both members (i.e., the minimum criteria for use here). Table 1. Basic statistics for the six data sets (data gathered November-December 2008). Data setDetailsEligible membersDialogsMin. com-mentsMax. com-mentsTotal com-mentsCom-ments per dialogUK allMembers joining before 3 July 2007614753896288.4UK July 06Members joining 3 July 200676415332021,92412.6UK June 07Members joining 18 June 2007355603273896.5US allMembers joining before 3 July 20076,37488435579,54410.8US July 06Members joining 3 July 20068,59388631639,06210.2US June 07Members joining 18 June 20072,83723531152,42310.3Total/overall19,5372,293355723,97010.5 The median percentage of Friends that commented on the public profiles downloaded was 33%: for a typical member, a third of their Friends left public comments on their profile. Moreover 78% of members received at least one comment. This is higher than previously found, probably because the prior research had included all members rather than just members with some indicator of activity (Parks, 2008). For 29% of members a dialog could be found involving public comments by themselves and a Friend. Of the dialogs extracted, 63% were classed as valid for the analysis and are reported in Table 1. Given that for some of the 37% of dialogs rejected, there may have been other valid dialogs for the same Friend, it seems that 19%-29% of MySpace members with public profiles and matching the selection criteria here have at least one valid public dialog. Distribution of public MySpace comments This section examines the gaps between successive comments and the total number of comments per dialog, focussing on the main US all data set. This nationally homogeneous set was chosen to illustrate the distribution patterns present: the other data sets give patterns consistent with these. Gaps - US all data set The median gap between public comments in the US all data set is just under 24 hours: half of all public comments that receive a public reply receive it within a day. Nevertheless, many public comments receive quick replies and the modal length of time to wait for a reply is one minute (MySpace logs comments by minute rather than by second). Figures 1 to 4 illustrate the distribution of gaps between successive public comments. Figure 1 includes the most rapid replies, issued within one hour. Presumably many replies arrive quickly because the Friends know that each other are online, access MySpace frequently or access it at the same time of day. Figure 2 shows a similar pattern for the first two days following a public comment, illustrating that the chance of receiving a reply drops significantly after just over a day. Presumably people who check MySpace daily would tend to reply within about 24 hours, but otherwise the wait for a reply might be much longer. Figure 3 confirms that long gaps between a public comment and its public reply (or unrelated new public comment) are rare and the logarithmic graph presentation of the Figure 3 data in Figure 4 shows that gaps between public comments generate the characteristic power law broomstick graph (Adamic & Huberman, 2000; Pennock, Flake, Lawrence, Glover, & Giles, 2002). The vast majority of gaps (96.8%) last for less than one year. An additional 3.0% of gaps last for two years, and 0.2% last for longer.  Figure 1. The distribution of gaps between public comments of up to 2 hours for the US all data set.  Figure 2. The distribution of gaps between public comments of up to 2 days for the US all data set.  Figure 3. The distribution of gaps between public comments for the US all data set.  Figure 4. The distribution of gaps between public comments for the US all data set (logarithmic axes, gaps of 0 days not shown). Dialog lengths- US all data set It is clear that although there are some long public dialogs, the majority of public dialogs are very short (Figure 5). The median public dialog length is 5 comments (Table 2), i.e., the typical number of public comments exchanged by US MySpace Friends is 5 for those who exchange at least 3. Because of the power law relationship for public dialog lengths, the mean public dialog length is higher at 10.8 comments, which seems barely enough for a single brief chat. Just under 1% of pairs of members matching the criteria here are apparently able to communicate extensively through hundreds of public comments.  Figure 5. The distribution of public dialog lengths, as measured by the total number of comments for the US all data set (logarithmic axes). Table 2. Dialog lengths, as measured by the total number of comments for the US all data set. CommentsFrequency (pairs of Friends)Cumulative %320523%415741%59552%67260%74565%83569%94774%102677%112280%121681%131783%141585%15786%161187%17788%18989%19890%-32195%-81199%-5571100% Distribution of public MySpace dialog types Table 3 illustrates that there are very few long exchanges, with the vast majority of public dialogs containing none. There are even few medium exchanges, and these are absent from the majority of public dialogs. The most common type is the short exchange or monolog, of which two are present in the typical dialog, although an additional one is present in the typical dialog of the main US all data set perhaps reflecting longer average membership durations for US all members and hence more time in which to communicate. Table 3. Median (mean) features in dialogs in each of the six data sets. Data setGaps (7+ days) between sets Long exchange (10+ comments, 2+ from each)Medium exchange (4+ comments, 1+ from each)Short exchange or monologUK all1 (1.5)0 (0.2)0 (0.5)2 (1.8)UK July 20061 (2.1)0 (0.2)0 (0.6)2 (2.3)UK June 20071 (1.1)0 (0.1)1 (0.7)1 (1.4)US all2 (3.0)0 (0.1)0 (0.5)3 (3.4)US July 20062 (2.8)0 (0.1)0 (0.5)2 (3.2)US June 20072 (2.3)0 (0.1)0 (0.4)2 (2.7) Table 4 shows that MySpace public comment dialogs (containing at least three comments and at least one from both Friends) most commonly fall within the multiple exchanges dialog category, meaning that they do not include long exchanges (10 comments, at least two from each participants, without gaps of longer than a week) but have at least one gap of over a week between comments. For example, a public dialog consisting of a yearly message or exchanges of birthday or festivity greetings would match this profile. Roughly 20%-30% of the public dialogs fall within the single exchange dialog category, however, meaning that there are no public comments after the initial exchange, and that the initial change does not fall into the long exchange(s) dialog category. In terms of national differences, it seems that UK members seem to use MySpace public comments less frequently in the modal multiple exchanges dialog form and more often in the single exchange dialog form. This could perhaps reflect less extensive use of MySpace by UK users, since the US was an early adopter, or a lower tendency to use public communication. There is no specific research to confirm the latter but one study of MySpace has shown US users to use public communication more frequently than German users (Banczyk, Krmer, & Senokozlieva, 2008). Table 4. Types of dialog in each of the six data sets. Long exchange(s) dialogMultiple exchanges dialog (excludes long exchanges)Single exchange dialog (excludes long exchanges)UK all15%52%33%UK July 200614%63%23%UK June 20075%53%42%US all8%76%16%US July 20068%74%18%US June 200711%70%19%Gossip in dialogs The content analysis of 100 random dialogs gave the results in Table 5. This showed that half of the dialogs contained some gossip in the sense of social information sharing and that this was more common than coordination of offline activities and exchanging other types of information. The information shared was not only personal information about the participants but also some discussion of others, as the examples show. Table 5. Frequency of three uses of MySpace dialogs from a random sample of 100. FactorPercentage of dialogsExamples (modified for anonymity)Dialog contains any gossip - discussion of own or others' behaviour, attributes or activities53%I moved to Houston, Tx. I come home at the beginning of July well i just diyed my hair nearly black!! i regret not going to UMSX bc MZU is so much harder i sooo messed up :(( i went out with car on friday night for a white guy tim knows a lot of rap song Tina talks about you all the time. she looked so much happier with you Nigel said you were feeling bad How far away does Tony go to school? J keeps askin if your comin to the wedding I heard u might be gettin a bike thats cool. so i heard that you started at the mall and just wondering how it is working out for youDialog contains any other information or facts1%Oprah lives in a mansion because shes rich.Dialog contains any arrangement or coordination of offline activities or other communication18%CALL ME WHEN YOU GET A CHANCE hey text me sometime.. [number] i hope to see you toniiite <3 I'm gonna be in ABD in Jan. for like a week, we gotta hang out Hey I can call you 2day?!!Implications In answer to the first research question, the typical length of public MySpace comment dialogs is five comments for exchanges in which both partners participate and exchange at least three messages. This short length suggests that they are rarely used for gossip in the sense of chatty talk although they may still be used for sporadic exchanges of gossip in the sense of rumor or report of an intimate nature. In answer to the second research question, MySpace public comment exchanges occur with intervals between comments that are sometimes minutes but are more commonly about a day or more. Hence, whilst MySpace public comments can be used for chat sessions they typically have the longer duration of email. For example, an analysis of email exchanges within a university found a median gap between one message and its reply of about 10 hours (Eckmann, Moses, & Sergi, 2004), which is even shorter than the median for MySpace exchanges. In contrast, a study of a chatroom (in which public messages are typically broadcast rather than directed at individuals) with typically 10-20 active users found gaps between messages to be under 5 minutes for 99.8% of the time (Acar, amtepe, Krishnamoorthy, & Yener, 2005). This difference seems to make comments a less natural vehicle for casual chatting, since this is typically rapid during face-to-face communication and in chatrooms. It would therefore be noteworthy, but not impossible, for comment dialogs to perform the function of casual chatting despite their typically longer time span. For the third research question, typical comment dialogs are multiple short exchanges, perhaps to stay in touch without extended public communication. This conclusion is on the basis that prolonged exchanges are rare but that most of the comment dialogs had at least one gap of at least seven days. Finally, about half of the public comment dialogs contained gossip-type sharing of information about the participants or others. Limitations This research has a number of significant limitations. Most importantly, it does not involve interviews with MySpace members, except indirectly via the literature review, and assumes that public comments (after Spam removal) are genuine communications between the sender and receiver. This is most evident in the heuristics designed to differentiate between usage patterns (Tables 3 and 4). The second major limitation is the restriction of the data to public comments. It seems possible, for instance, that members will feel freer to comment on private profiles - which form two thirds of profiles (Thelwall, 2009b) - than public profiles, although this does not seem likely to make a big difference because the commenter is not necessarily aware whether they are commenting on a private or public profile. This is because private profiles are only hidden from non-Friends. More significantly, it seems likely that much MySpace communication would be hidden in MySpace private messages. Presumably this would be the medium of choice for most close friends because friendship is partly defined by sharing information that is withheld from others (Tardy & Dindia, 2006). It also seems unlikely that traditional gossip in the form of negative discussions of third parties would be conducted in public comments. Another issue is that since members under 16 all have private profiles, the youngest members of MySpace are all excluded, unless part of the minority misreporting their ages (Hinduja & Patchin, 2008). This is not a real problem, however, since the sample includes data from people who joined when under 16 but made their profile public when older. The data has some sampling limitations because it is restricted to just U.S. and U.K. members (and reliant upon self-reported geographical location). Patterns of use may well be different for other countries, especially given the high degree of national homophily amongst MySpace members (Thelwall, 2009a). In addition, it may be that the most recent MySpace members use it differently from the older members sampled here. This seems quite likely, especially in the U.S., since MySpace has recently ceased to be the dominant social networking site and so may be attracting a more niche audience. Conclusions Relationship maintenance For the typical MySpace member with a public profile and at least two Friends, a third of their friends post a comment on that profile. In about 19%-29% of cases, members with a public profile also have at least one public dialog with a Friend using comments, and with both participating. Hence public dialogs are not standard in MySpace: they are a minority activity. The results show that whilst MySpace public comment dialogs are typically rapid, they are rarely long. Dialogs seem to be primarily used for keeping in touch occasionally with friends or acquaintances (e.g., about 7.5% of public comments mention birthdays, Thelwall, 2009b), for establishing initial contact or making an initial brief exchange without subsequently staying in touch, unless switching to another communication medium. Such exchanges may serve to maintain relationships over time that might otherwise lapse. Although there are many definitions of relationship maintenance, the most fundamental defines it as the process of keepinga relationship in existence (Dindia & Canary, 1993). Clearly any kind of exchange, such as annual birthday or festivity greetings, keeps a relationship from being completely forgotten (e.g., Hill & Dunbar, 2003)and hence some messages may merely replace the traditional greetings card. MySpace public comments are rarely used for chat-like long exchanges, however. In comparison, Facebook members actively engage with a tiny minority of their SNS network. An average male Facebook user has 120 friends and makes public or semi-public responses to only 7 of them, and a typical female user responds to 10. Moreover, the average male user engages with (private or public) two-way communication with only 4 members and the average female engages with 6 (Kluth, 2009). This suggests an even lower level of activity than for MySpace public comments, despite the Facebook data including all types of profiles and Facebook profiles being typically private in the sense of not open to everyone. Nevertheless, Facebook users typically cite maintaining relationships with people you may not get to see very often as a very important SNS activity (Joinson, 2008) but it may be that it is important only for a small number of Friends, with the rest being effectively forgotten, or it may be that this maintenance is only passive or only occurs in theory. Gossip Given the sparseness of long public dialogs between Friends, it seems that public comments are rarely used for gossip-based social grooming. Although about half of the dialogs exchange what could be described as gossip, the average public dialog seems too short for a systematic exchange of information: for dialogs with at least 3 comments, at least half have less than 6 comments: moreover comments are typically very short (Thelwall, 2009b). This seems to be too little to perform a grooming function because grooming suggests a significant time investment in order to build up trust with others (Dunbar, 1998). Nevertheless, social grooming could still occur via gossip exchanged in private messages or in gossip-type passive information gathering via mutual profile browsing since browsing profiles and virtual people watching are popular SNS activities (Joinson, 2008; Raacke & Bonds-Raacke, 2008). Conversely, it is possible that Friending in MySpace may be typically used for symbolic connections rather than for information exchange between Friends and previous research suggests that such friend connections are important in themselves (Donath & boyd, 2004), particularly for older users (at least in Facebook) (Livingstone, 2008). Would the same results hold for other SNSs, and particularly Facebook, which is currently dominant? Facebook users actively communicate (e.g., exchange messages) with less than half of the people that they passively communicate with (e.g., click on a news feed story, visit a profile) (Facebook, 2009). This suggests that social information gathering is more common than directly maintaining contact through Facebook and is probably much more common than gossip, assuming that most messages exchanged do not contain gossip. Perhaps Facebooks initially controversial News Feed feature (boyd, 2008), which brings news of Friends activities automatically without the need for browsing, performs an automatic social information gathering function that partly avoids the need for direct contact. In contrast, one of the special characteristics of the popular Korean Cyworld SNS is that it suggests random Friends to visit each day (Kim & Yun, 2007), which may encourage keeping in touch with Friends. Thus it would be hard to argue that gossip-type information exchanges were rare in all SNSs and so the results for MySpace should not be generalised to other SNSs without further research. Finally, given the findings from MySpace and the cited Facebook research, it seems that future research into textual public social network communication could focus not on typical pairs of Friends but on pairs of Friends that communicate the most. This would presumably reveal much lengthier communication patterns. Networked individualism In the wider context, MySpace comments are probably one of a number of communication tools used by networked individuals as part of their strategy to establish and maintain contact with an increasingly large set of friends and acquaintances (e.g., Tufekci, 2008a). Whilst the results do not prove that typical MySpace Friends rarely communicate, since they may exchange private messages, unilaterally comment, comment on each others blog or pictures, or use other online or offline means, this is still a plausible interpretation since the data covers only pairs of Friends that at least started to exchange MySpace comments. Nevertheless, closer Friends may tend to use private rather than public messages, and members may switch between private messages and public comments, for example establishing contact with public comments and then switching to private messages. Future research using interviews and questionnaires with users will be needed to identify the exact role of public comments in MySpace members lives. Appendix Single short exchange dialog example 1: 08/03/2008 16:43: I->J: wassup, long time no see. how you been? 08/03/2008 16:46: J->I: im doing very good!!! how bout you?? 08/03/2008 17:10: I->J: I been doing great too... Single short exchange dialog example 2: 02/05/2007 07:07: K->L: ur video above.. is da reason why kids have car crashes!!! 02/05/2007 07:51: L->K: We were stopped by a train, on of them looong trains with 224,538,562,000 carts so we had time we werent driving but did you like the vid? 02/05/2007 15:15: K->L: hahaha ya you are CrAzY!!!! Multiple short exchanges dialog example 1: 29/06/2007 14:50:M-> N: I'm not letting your brother be the only one posting on your myspace. Are you moving back tomorrow? 30/06/2007 03:54: N-> M: Thanks for the message. Yes. Saturday I will be back in DW! Finally... Soon you'll have to come meet my daughter. 23/07/2007 15:43: M -> N: Good luck on Saturday! (No, I'm not psychic - I saw Nigel at Trest) Multiple short exchanges dialog example 2: 29/05/2007 01:48:O-> P: Ur hot 29/05/2007 13:34:P-> O: u love it 29/05/2007 19:51: O -> P: Nope, I h8 it as much as I h8 u. 30/05/2007 19:42:P-> K: awwwww...you made me tear up :_) 13/12/2007 05:27: O -> P: ru alive? 23/01/2008 20:52: O -> P: Cleverest girl in the country you are, which is why I'm gonna take you 2 a bar. 17/03/2008 04:57: O -> P: u r a piirate hooker, I'll be down this weekend, soooo quit ur job. Extended exchange(s) dialog example: 179 comments from 05/12/2006 to 14/10/2008 including the following extract: 15/12/2006 03:40:Q -> R: [Image Message] 18/12/2006 23:56: R -> Q: high there my dear little sis,did I ever tell you that you are my fav lil sis....... 19/12/2006 00:00: Q -> R: Awwww how sweet, and ur my favoritiest big bro... see you Sunday for a little while... OK? Better b there... Love Ya... 19/12/2006 00:04: R -> Q: only if Tom will be there too,he know that I miss him....... 19/12/2006 00:06: Q -> R: You miss Tom more than me.... I'll tell him... heeeheee Ohhh and my daughter - yes shes an angel next to Thompson!Hes "animal" from the muppets like you said...LOL 19/12/2006 00:16: R -> Q: that's right!!!Everybody in the ghetto has nicknames nobody calls you by your name,Thompson is Animal off the mupppet babies,give him drum sticks and watch the helll out! 19/12/2006 22:39: Q -> R: Yepp... or a billy club like the one with his police officer birthday costume... heeeheee... so what should we use for MarySues "Rosedale" nickname LOL... 20/12/2006 14:14: R -> Q: [Long message from a myspace application describing Animal from the Muppets] 21/12/2006 00:31: Q -> R: Yeah... that kinda sounds like Thompo.. has to bang bang bang and doesn't mean to be messy all the time and blame his sister! 21/12/2006 14:14: Q -> R: Good Morning Brother... Have a great day, dont work to hard... heeeheee... love yer.. 23/12/2006 17:20: Q -> R: [Image Message] Acknowledgements Thank you to the anonymous referees for very helpful suggestions. This work was supported by a European Union grant by the 7th Framework Programme, Theme 3: Science of complex systems for socially intelligent ICT. It is part of the CyberEmotions project (contract 231323). References Acar, E., amtepe, S. A., Krishnamoorthy, M. S., & Yener, B. (2005). Modeling and multiway analysis of chatroom tensors. Lecture Notes in Computer Science, 3495, 256-268. Adamic, L. A., & Huberman, B. A. (2000). Power-law distribution of the World Wide Web. Science, 287(24 March), 2115a. Aukett, R., Ritchie, J., & Mill, K. (1988). Gender differences in friendship patterns. Sex Roles, 19(1-2), 57-66. Banczyk, B., Krmer, N. C., & Senokozlieva, M. (2008). "The wurst" meets "fatless" in MySpace. The relationship between self-esteem, personality and self-presentation in an online community. Paper presented at the Congress of the International Communication Association, Montreal, Canada. Baumeister, R. F., Zhang, L., & Vohs, K. D. (2004). Gossip as cultural learning. Review of General Psychology, 8(2), 111-121. Baym, N. (2008). Tunes that bind? Predicting friendship strength in a music-based social network. Retrieved October 21, 2008 from: http://www.onlinefandom.com/wp-content/uploads/2008/2010/tunesthatbind.pdf. Biocca, F., Harms, C., & Burgoon, J. K. (2003). Toward a more robust theory and measure of social presence: review and suggested criteria. Presence: Teleoperators and Virtual Environments archive, 12(5), 456-480. boyd, d. (2004). Friendster and publicly articulated social networks. In Conference on Human Factors and Computing Systems (CHI 2004, Vienna: April 24-29) (pp. Retrieved July 3, 2007 from: http://www.danah.org/papers/CHI2004Friendster.pdf). New York: ACM Press. boyd, d. (2006). Friends, Friendsters, and MySpace Top 8: Writing community into being on social network sites. First Monday, 11(2), Retrieved April 29, 2009 from: http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/1418/1336. boyd, d. (2008). Facebook's privacy trainwreck: Exposure, invasion, and social convergence. Convergence, 14(1), 13-20. boyd, d. (2008). Taken out of context: American teen sociality in networked publics. University of California, Berkeley, Berkeley. boyd, d., & Ellison, N. (2007). Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication, 13(1), Retrieved May 7, 2009 from: http://jcmc.indiana.edu/vol2013/issue2001/boyd.ellison.html. boyd, d., & Heer, J. (2006). Profiles as conversation: Networked identity performance on Friendster. Proceedings of the Hawai'i International Conference on System Sciences (HICSS-39, January 4-7), Retrieved July 3, 2007 from: http://www.danah.org/papers/HICSS2006.pdf. Burke, M., Marlow, C., & Lento, T. (2009). Feed me: motivating newcomer contribution in social network sites. In Proceedings of the 27th international conference on Human factors in computing systems (pp. 945-954). New York, NY: ACM Press. Daft, R. L., & Lengel, R. H. (1984). Information richness: a new approach to managerial behavior and organizational design. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior 6 (pp. 191-233). Homewood, IL: JAI Press. Derks, D., Bos, A. E. R., & von Grumbkow, J. (2008). Emoticons and online message interpretation. Social Science Computer Review, 26(3), 379-388. Dindia, K., & Canary, D. J. (1993). Definitions and theoretical perspectives on maintaining relationships. Journal of Social and Personal Relationships, 10(2), 163-173. Donath, J. (2007). Signals in social supernets. Journal of Computer-Mediated Communication, 13(1), Retrieved June 17, 2008 from: http://jcmc.indiana.edu/vol2013/issue2001/donath.html. Donath, J., & boyd, d. (2004). Public displays of connection. BT Technology Journal, 22(4), 71-82. Dunbar, R. (1998). Grooming, gossip and the evolution of language.London: Faber and Faber Limited. Eckmann, J.-P., Moses, E., & Sergi, D. (2004). Entropy of dialogues creates coherent structures in e-mail traffic. Proceedings of the National Academy of Sciences, 101(40), 14333-14337. Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook "friends:" Social capital and college students' use of online social network sites. Journal of Computer-Mediated Communication, 12(4), 1143-1168. Facebook. (2009). Facebook Data Team: Maintained relationships on Facebook. Retrieved July 15, 2009 from: http://www.facebook.com/note.php?note_id=55257228858&id=58394258414&ref=nf. Golder, S. A., Wilkinson, D., & Huberman, B. A. (2007). Rhythms of social interaction: Messaging within a massive online network, 3rd International Conference on Communities and Technologies (CT2007), East Lansing, MI. Gruhl, D., Guha, R., Liben-Nowell, D., & Tomkins, A. (2004). Information diffusion through Blogspace. Paper presented at the WWW2004, New York, Retrieved July 10, 2006 from: http://www2004.org/proceedings/docs/1p491.pdf. Haddon, L., & Kim, S. D. (2007). Mobile phones and web-based social networking - Emerging practices in Korea with Cyworld. Journal Of The Communications Network, 6, 5-12. Hancock, J. T., Landrigan, C., & Silver, C. (2007). Expressing emotion in text-based communication. In CHI '07: Proceedings of the SIGCHI conference on Human factors in computing systems (pp. 929-932). New York, NY: ACM. Haythornthwaite, C., & Wellman, B. (1998). Work, friendship, and media use for information exchange in a networked organization. Journal of the American Society for Information Science, 49(12), 1101-1114. Hill, R. A., & Dunbar, R. (2003). Social network size in humans. Human Nature, 14(1), 53-72. Hinduja, S., & Patchin, J. W. (2008). Personal information of adolescents on the Internet: A quantitative content analysis of MySpace. Journal of Adolescence, 31(1), 125-146. Joinson, A. N. (2008). Looking at, looking up or keeping up with people? Motives and use of facebook. Proceeding of the twenty-sixth annual SIGCHI conference on Human factors in computing systems, 1027-1036. Kim, K.-H., & Yun, H. (2007). Cying for me, Cying for us: Relational dialectics in a Korean social network site. Journal of Computer-Mediated Communication, 13(1), 298-318. Kluth, A. (2009). Primates on Facebook. The Economist, February 26 2009. Retrieved July 2015, 2007 from: http://www.economist.com/science/displaystory.cfm?story_id=13176775. Kock, N. (2005). Media richness or media naturalness? The evolution of our biological communication apparatus and its influence on our behavior toward e-communication tools. IEEE Transactions on Professional Communication, 48(2), 117-130. Lampe, C., Ellison, N., & Steinfield, C. (2006). A face(book) in the crowd: Social Searching vs. social browsing. Proceedings of the 2006 20th anniversary conference on Computer supported cooperative work, 167-170. Lampe, C., Ellison, N., & Steinfield, C. (2008). Changes in use and perception of facebook. In Proceedings of the ACM 2008 conference on Computer supported cooperative work (pp. 721-730). San Diego, CA: ACM Press. Lange, P. G. (2007). Publicly private and privately public: Social networking on YouTube. Journal of Computer-Mediated Communication, 13(1), Retrieved May 8, 2008 from: http://jcmc.indiana.edu/vol2013/issue2001/lange.html. Larsen, J., Urry, J., & Axhausen, K. (2006). Mobilities, Networks, Geographies. Farnham, UK: Ashgate. Lenhart, A., Madden, M., Macgill, A. R., & Smith, A. (2007). Teens and Social Media: The use of social media gains a greater foothold in teen life as they embrace the conversational nature of interactive online media (12/19/2007). Pew Internet & American Life Project, Retrieved May 1, 2008 from: http://www.pewinternet.org/PPF/r/2230/report_display.asp. Liu, H. (2007). Social network profiles as taste performances. Journal of Computer-Mediated Communication, 13(1), Retrieved June 5, 2008 from: http://jcmc.indiana.edu/vol2013/issue2001/liu.html. Livingstone, S. (2008). Taking risky opportunities in youthful content creation: Teenagers' use of social networking sites for intimacy, privacy and self-expression. New Media & Society, 10(3), 393-411. Malinowski, B. (1923). The problem of meaning in primitive languages. In C. K. Ogden & I. A. Richards (Eds.), The Meaning of Meaning: Routlledge & Kegan Paul (pp. 296-346). Nardi, B. A. (2005). Beyond bandwidth: Dimensions of connection in interpersonal communication. Computer-Supported Cooperative Work, 14(1), 91-130. Parks, M. R. (2008). Characterizing the communicative affordances of MySpace: A place for friends or a friendless place? Paper presented at the International Communication Association, Montreal. Pennebaker, J., Mehl, M., & Niederhoffer, K. (2003). Psychological aspects of natural language use: Our words, our selves. Annual Review of Psychology, 54, 547-577. Pennebaker, J. W., Mayne, T., & Francis, M. E. (1997). Linguistic predictors of adaptive bereavement. Journal of Personality and Social Psychology, 72(4), 863-871. Pennock, D., Flake, G. W., Lawrence, S., Glover, E. J., & Giles, C. L. (2002). Winners don't take all: Characterizing the competition for links on the web. Proceedings of the National Academy of Sciences, 99(8), 5207-5211. Pew Research Center for the People & the Press. (2007). Social networking websites and teens: An overview. Retrieved June 4, 2007, from http://www.pewinternet.org/PPF/r/198/report_display.asp Ploderer, B., Howard, S., & Thomas, P. (2008). Being online, living offline: the influence of social ties over the appropriation of social network sites. In Proceedings of the ACM 2008 conference on Computer supported cooperative work (pp. 333-342). San Diego, CA: ACM Press. Raacke, J., & Bonds-Raacke, J. (2008). MySpace and Facebook: Applying the uses and gratifications theory to exploring friend-networking sites. Cyberpsychology & Behavior, 11(2), 169-174. Riva, G. (2002). The sociocognitive psychology of Computer-Mediated Communication: The present and future of technology-based interactions. CyberPsychology & Behavior, 5(6), 581-598. Short, J., Williams, E., & Christie, B. (1976). The social psychology of telecommunications.London: Wiley. Steinfield, C., Ellison, N. B., & Lampe, C. (2008). Social capital, self-esteem, and use of online social network sites: A longitudinal analysis. Journal of Applied Developmental Psychology, 29(6), 434445. Tardy, C., & Dindia, K. (2006). Self-disclosure: Strategic revelation of personal information in personal and professional relationships. In O. Hargie (Ed.), A Handbook of Communication Skills (3rd Ed.) (pp. 229-266). London: Routledge. Thelwall, M. (2008). Social networks, gender and friending: An analysis of MySpace member profiles. Journal of the American Society for Information Science and Technology, 59(8), 1321-1330. Thelwall, M. (2009a). Homophily in MySpace. Journal of the American Society for Information Science and Technology, 60(2), 219-231. Thelwall, M. (2009b). MySpace comments. Online Information Review, 33(1), 58-76. Thelwall, M. (2009c). Social network sites: Users and uses. In M. Zelkowitz (Ed.), Advances in Computers (pp. 19-73). Amsterdam: Elsevier. Thelwall, M., Wilkinson, D., & Uppal, S. (in press). Data mining emotion in social network communication: Gender differences in MySpace. Journal of the American Society for Information Science and Technology. Thelwall, M., Wouters, P., & Fry, J. (2008). Information-centred research for large-scale analysis of new information sources. Journal of the American Society for Information Science and Technology, 59(9), 1523-1527. Tufekci, Z. (2008a). Can you see me now? Audience and disclosure regulation in online social network sites. Bulletin of Science, Technology and Society, 28(1), 20-36. Tufekci, Z. (2008b). Grooming, gossip, Facebook and MySpace: What can we learn about these sites from those who won't assimilate? Information, Communication & Society, 11(4), 544-564. Walther, J., & Parks, M. (2002). Cues filtered out, cues filtered in: computer-mediated communication and relationships. In M. Knapp, J. Daly & G. Miller (Eds.), The Handbook of Interpersonal Communication (3rd ed.).Thousand Oaks, CA: Sage. Walther, J., Van der Heide, B., Kim, S., Westerman, D., & Tong, S. T. (2008). The role of friends' appearance and behavior on evaluations of individuals on Facebook: Are we known by the company we keep? Human Communication Research, 34, 28-49. Walther, J. B. (1992). Interpersonal effects in computer-mediated interaction: A relational perspective. Communication Research, 19(1), 52-90. Wellman, B., A., Quan-Hasse, J., Boase, W., Chen, K., Hampton, I., de Diaz, I., et al. (2003). The social affordances of the Internet for networked individualism. Journal of Computer-Mediated Communication, 8(3), Retrieved April 15, 2009 from: http://jcmc.indiana.edu/vol2008/issue2003/wellman.html. Wert, S. R., & Salovey, P. (2004). A social comparison account of gossip. Review of General Psychology, 8(2), 122-137. Whittaker, S. (2003). Theories and methods in mediated communication. In A. Graesser (Ed.), The handbook of discourse processes (pp. 253-293). Hillsdale: NJ: Lawrence Erlbaum.  Thelwall, M. & Wilkinson, D. (2010). Public dialogs in social network sites: What is their purpose? Journal of the American Society for Information Science and Technology, 61(2), 392-404. copyright 2009 John Wiley & Sons.     Page  PAGE 15 of  NUMPAGES 19 Last printed  PRINTDATE \@ "dd/MM/yyyy HH:mm:ss" 00/00/0000 00:00:00 6=>? 5 ` { $'/67H<MKL/5@:g##F$]$^$t$v$>%P%~~vhnMYhR] hEzhRhEzhR6h:NhK&;hR6] hAqhR hdrhR hhRh8yh]h-hdsh&)h9h&)5CJ\aJh9hR5CJ\aJhLPhRmHsHjh-0JU hX%hRhR.@_ 5 _ ` k[@U"N(o(,255T: ASEIgd-`gd-gd-gd-gd-P%)***B.C.66@6C666992;N;O;;;;;;;= =EEEFGGIIIIIJPJZJ|J}J9K?K@KAKLKMKKRLSLYLLMMM2M?MMMYMO PQP h9\hRhM hX%hR h%NhR h<hR hhR h3JhR h1_ hR hR6 hB!hRhN*KhR6hEhR6]h$h]h-hR=IILoMNNGOOO8SUFVVVVWPWZ{ZZE[<\]\\\ & F gd- & Fgd-^gd-`gd- & Fgd-gd-gd-QPSPoPuP,Q7Q}QQNTTVVVVVVVV!Z.Z}ZZZZZZZZZC[%\&\]]~]]]]]]]]]]]]]]]]A^g^bbbŽԉh!NhR6 h>hRhRmHsHhShRmHsHh>hRmHsH hYhRhV]hR6h^ihR6 h^ihR hR6 hnDhR hV]hR hghR h9\hRhR hahR6\]]G]x]]]]]]^(^A^g^^^^_D_`9``-abKcc & Fgd- & F gd- & F gd-gd- & F gd-bb ccHcIcKc]cdccccccccccccclllmmmm&m'm7mCmDmHmImRmSmWmXmgmhmnmomqmrmmmmʵʵy(h%h-CJOJQJ^JaJnH tH hMih-CJaJnH tH .hGh-5CJOJQJ\^JaJnH tH (h-5CJOJQJ\^JaJnH tH (hMih-CJOJQJ^JaJnH tH h-h^h!NhR6hR h!NhR hR6/ccXehiillmm'm0m8m@mOm^mnmmmmmm $$Ifa$gdhDFfj $$Ifa$gdhD $$Ifa$gdhDgd-`gd-gd- & Fgd-mmmmmmmmmmmmmmmmmmmmmmmnnn nnnn n"n#n'n(n-n1nAn­­­˜˜˜t_˜˜˜ttt_(hh-CJOJQJ^JaJnH tH (hGh-CJOJQJ^JaJnH tH hMih-CJaJnH tH (hMih-CJOJQJ^JaJnH tH (h h-CJOJQJ^JaJnH tH "h-CJOJQJ^JaJnH tH (h%h-CJOJQJ^JaJnH tH +h%h-CJOJQJ\^JaJnH tH %mmmmmmmmmmmmmmnnn-n1n4n6n9n=nAnBnInFfk Ff $$Ifa$gdhDFf $$Ifa$gdhDAnBnDnHnIn`nanbnfngnknlnmnnnrn}n~nnnnnnnnnnnnnnnnnnnnnnnnnnnnnooooo%o'o(o4o5oʵʵʵܠʠʋʋʋʋʠ(hGh-CJOJQJ^JaJnH tH (hh-CJOJQJ^JaJnH tH (h h-CJOJQJ^JaJnH tH "h-CJOJQJ^JaJnH tH (hMih-CJOJQJ^JaJnH tH hMih-CJaJnH tH 6Inlnrnvnxn|nnnnnnnnnnnnnnnnnoo ooFfFf $$Ifa$gdhD $$Ifa$gdhDoooo&o,o.o2o9o>o?o@orrssyyy!z"z$zzz$a$gd-gd-gd-gd-Ff $$Ifa$gdhD $$Ifa$gdhDFfl5o>o?o@oppr;sAssssss&t,t0y1yyyyyyyzz"z#z$zwz}zzzzzzzzػثؤؠؘس؅}rk}`jh-h-U h-h-jh-h-Uh-hR6jdUh-U h-hRh-jLh-Uh^ hhRhhhR6h%hR6 hOhR h0hRh0hR6 hohRhRhMih-CJaJnH tH (hMih-CJOJQJ^JaJnH tH %zzzzzc{{}}}x~y~~~~ $$Ifa$gdhDlwgd-gd-$a$gd-zz%{+{s{y{d}m}x}}}}}T~Z~w~y~~~~    !*+45>?HIRS]^hist~аwwwwwwwwwwww(h}h-CJOJQJ^JaJnH tH h}h-CJaJnH tH (h-5CJOJQJ\^JaJnH tH .h}h-5CJOJQJ\^JaJnH tH h%h-6h-hRjh-h-U h-h^h-hR6 h-hR h-h-.   [E//$$Ifa$gdhDlw$$Ifa$gdhDlwkdo$$IfTlFvJ& g06    44 lapyt<%T [E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%T !#&*[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkdu$$IfTlFvJ&& &g06    44 lapyt<%T*+-04[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%T457:>[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd{$$IfTlFvJ&& &g06    44 lapyt<%T>?ADH[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%THIKNR[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%TRSVY][E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%T]^adh[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%Thilos[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd $$IfTlFvJ&& &g06    44 lapyt<%Tstwz~[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%T~[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%T[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%T,uv~ǂׂ ººƳʞʞʞʞʞʞʞʞhKEh-CJaJnH tH (hKEh-CJOJQJ^JaJnH tH  hKEh-hhR6hRh-"h-CJOJQJ^JaJnH tH (h}h-CJOJQJ^JaJnH tH h}h-CJaJnH tH 8[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%T[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%T[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%T[E//$$Ifa$gdhDlw$$Ifa$gdhDlwkd$$IfTlFvJ&& &g06    44 lapyt<%T[E/$$Ifa$gdhDlw$$Ifa$gdhDlwkd"$$IfTlFvJ&& &g06    44 lapyt<%TE/$$Ifa$gdhDlkd$$IfTlFvJ&& &g06    44 lapyt<%T$$Ifa$gdhDlwE/$$Ifa$gdhDlkd($$IfTlFvJ&& &g06    44 lapyt<%T$$Ifa$gdhDl/kd$$IfTlFvJ&& &g06    44 lapyt<%T$$Ifa$gdhDl$$Ifa$gdhDl/kd. $$IfTlFvJ&& &g06    44 lapyt<%T$$Ifa$gdhDl$$Ifa$gdhDl/kd $$IfTlFvJ&& &g06    44 lapyt<%T$$Ifa$gdhDl$$Ifa$gdhDl/kd4 $$IfTlFvJ&& &g06    44 lapyt<%T$$Ifa$gdhDl$$Ifa$gdhDl,-vǂׂ $$Ifa$gdhDgd-gd- /# $$Ifa$gdhDkd $$Iflr ''''T'0644 la]p2ythD%-5 $$Ifa$gdhD #$%(+,-03456ABDJKNQRSVYZ[^abcdoprxy|ƒŃƃȃ΃(hr h-CJOJQJ^JaJnH tH hKEh-CJaJnH tH "h-CJOJQJ^JaJnH tH (hKEh-CJOJQJ^JaJnH tH G56C/# $$Ifa$gdhDkd $$Iflr ''''T'0644 la]p2ythDCKS[c $$Ifa$gdhDcdq/# $$Ifa$gdhDkd $$Iflr ''''T'0644 la]p2ythDqy $$Ifa$gdhD/# $$Ifa$gdhDkd$$Iflr ''''T'0644 la]p2ythD $$Ifa$gdhDǃ/# $$Ifa$gdhDkd/$$Iflr ''''T'0644 la]p2ythDǃσ׃߃ $$Ifa$gdhD΃σ҃Ճփ׃ڃ݃ރ߃ R։܉ĵĵj뵘(hYh-CJOJQJ^JaJnH tH 0hPh-CJOJQJ^JaJmH nH sH tH h-CJOJQJ^JaJ hKEh-hRh-hKEh-CJaJnH tH (hr h-CJOJQJ^JaJnH tH "h-CJOJQJ^JaJnH tH (hKEh-CJOJQJ^JaJnH tH )/# $$Ifa$gdhDkdM$$Iflr ''''T'0644 la]p2ythD  $$Ifa$gdhDQR/***gd-kdk$$Iflr ''''T'0644 la]p2ythDR։ $$Ifa$gdhD $IfgdhDgd-G>222 $$Ifa$gdhD $IfgdhDkd$$Ifl\ $ '0 644 la]p(ythD)-56CFNOVYaborz{PQ@DOV[\Íčɍʍݍ'*BGefιιιιιܵ hVh- hX?vh-h- hH!hRhR(hKEh-CJOJQJ^JaJnH tH h-CJOJQJ^JaJhKEh-CJaJnH tH (hgh-CJOJQJ^JaJnH tH =)-15G>222 $$Ifa$gdhD $IfgdhDkd$$Ifl\ $ '0 644 la]p(ythD56CFJNG>222 $$Ifa$gdhD $IfgdhDkd$$Ifl\ $ '0 644 la]p(ythDNOVY]aG>222 $$Ifa$gdhD $IfgdhDkd$$Ifl\ $ '0 644 la]p(ythDaborvzG>222 $$Ifa$gdhD $IfgdhDkd$$Ifl\ $ '0 644 la]p(ythDz{G>222 $$Ifa$gdhD $IfgdhDkd$$Ifl\ $ '0 644 la]p(ythDPQGB===gd-gd-kd}$$Ifl\ $ '0 644 la]p(ythD@DEkd{$$IflF ? t6 t06    44 laythD$IfgdhDl$$Ifa$gdhDlD\ލCfώ'$IfgdhDlΎЎ!%&'JNikďԏُ>B`axܐސ+,fl}~ғ3U[˕ҕݕ A_hRh5h h1_ h 68hh%^h 68h^hn@hR hBnh- hVh-h- hX?vh-LnXEE$IfgdhDl$$Ifa$gdhDlkd$$IflF ? t6 t06    44 laythD>BaސnXEEEEE$IfgdhDl$$Ifa$gdhDlkdq$$IflF ? t6 t06    44 laythDސɘ[VQHHCgd-`gd-gd-gd-kd$$IflF ? t6 t06    44 laythD$IfgdhDl_pqJMh01[\tuTẒ>?89ӦԦEHmn hYp\hR6]hhR6] hLChR h{ 1hR hC\YhRh- hOP[`gd-gd-$a$gd-gd-տֿ,-OP[]i%C`#zt_*h.Gf)ĺĶh-hhRmHsHhhR6h5h5h56] h5h5 hdshds h'Chdshds hGEhR hhRhR hR6]h~khR6];[|\17-' 8SpOhE$0^`0a$gd-)]p?E8`"Sc_I*in?m5 (Oq*Z7c乮h]hhRmH sH hhR6mH sH hh-6h-hhRmH sH hhR6hRh-hRmHsHh-hh-6CEoI8*P\hG$0^`0a$gd-fc <Z[\#m~*D |Oraɿhmk_h<%CJaJmH sH jhUh h9h9h<%jh<%0JUhG hhRmHsHhhRmH sH hhR6mH sH hJ/hRhhR6:sH}$P|gd9gd-$0^`0a$gd-òhG hh<%h<%CJaJmHnHuhmk_h<%CJaJh9CJaJmHnHsH uhmk_h<%CJaJmH sH %jhmk_h<%CJUaJmH sH gd-gd1,1h. A!"#$% h$$Ifl!vh555R55&5-55#v#v#vR#v#v&#v-#v:V l0 !6555R55&5-59alpPyt--kd$$Iflִ  !&R&-0 !6    44 lalpPyt-z$$Ifl!vh555R55&5-55#v#v#vR#v#v&#v-#v:V l0 !6,,555R55&5-599alpPyt--kd$$Iflִ  !'R&&&&-&&0 !6    44 lalpPyt-z$$Ifl!vh555R55&5-55#v#v#vR#v#v&#v-#v:V l0 !6,,555R55&5-599alpPyt--kdD$$Iflִ  !'R&&&&-&&0 !6    44 lalpPyt-z$$Ifl!vh555R55&5-55#v#v#vR#v#v&#v-#v:V l0 !6,,555R55&5-599alpPyt--kd $$Iflִ  !'R&&&&-&&0 !6    44 lalpPyt-z$$Ifl!vh555R55&5-55#v#v#vR#v#v&#v-#v:V l0 !6,,555R55&5-599alpPyt--kd$$Iflִ  !'R&&&&-&&0 !6    44 lalpPyt-z$$Ifl!vh555R55&5-55#v#v#vR#v#v&#v-#v:V l0 !6,,555R55&5-599alpPyt--kdE$$Iflִ  !'R&&&&-&&0 !6    44 lalpPyt-z$$Ifl!vh555R55&5-55#v#v#vR#v#v&#v-#v:V l0 !6,,555R55&5-599alpPyt--kd$$Iflִ  !'R&&&&-&&0 !6    44 lalpPyt-$$Ifl!vh555R55&5-55#v#v#vR#v#v&#v-#v:V l0 !6,,,555R55&5-599alpPyt--kd$$Iflִ  !'R&&&&-''0 !6    44 lalpPyt-8Dd<$jjl  C HA0Figure 1 us mar 08 minsbX7dp,K8|،z347n,7dp,K8|،z3PNG  IHDRu)tPLTE3f3333f333ff3fffff3f3f̙3f3333f3333333333f3333333f3f33ff3f3f3f3333f3333333f3̙333333f333ff3ffffff3f33f3ff3f3f3ffff3fffffffffff3fffffff3fff̙ffff3fffff3f̙3333f33̙3ff3ffff̙f3f̙3f̙̙3f̙3f3333f333ff3fffff̙̙3̙f̙̙̙3f̙3f3f3333f333ff3fffff3f3f̙3fDbPtRNS%bbKGDH cmPPJCmp0712Hs3IDATx^ EosUͱ?O`0uwOyf+ K?@@p lӃ0 1QB lzu!I^^{F:Pl[q{x}N6S lNB/r^SIo'AtB^qc_1>\`Po?{ W`An5 rz0~m/;z@U]tl!=@V f&5?`.^Sna1Fу{r  åz_!=?xF"1Rw@}c1 61l4jzE.8"=PuwJsz 3T~AW'%7Ϩ]R@譎/H' UfTfd\TD6#@pdn!B 0!p )(„ '@B '[@@"L@pdn!B 0!p )(„ '@B '[@@"L@pdn!B 0!p )(„ '@B '[@@"L@pdn!B 0!p )(„ '@B '[@@"L@pdn!B 0!pkr- R?҂<@JA8J \ %yhOО1wpR q>@ KpCB?(]01׃B(&A?-R@Wy}0JjaпQȥ4g]Ah90aVaoe{ J$R3: ]dN@|ZetwVH/{oGGk?`]ȝ0GB7'p1r=p3ǟ_p\H?0A8ue~_3dO(cL-Ѓ?7}OA`l)Շ$$YȕtAvk?1'zb'1*1VV/ {r >@E0mi4=i9oDPȢ5#Z^lhSU͜(1lhF?%%a#?Wd fВ0+2@3fhICA3$ q⊌!ЌI͊!˛ ЌZ8qEFhF - C@"#`4#?h! @\04CKG %a#?Wd f.,XnV$ .lhZdfВ0+2@3fhICA3$ q⊌!ЌZ8qEFhF - C@"#`4#p?`r%al$p?`C"t4#?h! @\04CKGJ+CN,0ق AEhL?x0D11_; eL(%p6;ԃ;H JK pc=JP0ɼ@]H(HpNrl#pXWa 5b 88|^{Op< E7轂I= (f p @#jYS @`#@O1T^fpz?XXqT`-2X~F@8A4GJ bOH?7G ĆhD?0ШH  $k^IC*cVurL0QK[ /wÇeL(%p(5ПcowC<#pFgL2*[F ^N*r+bR@1c`^@M06ڠR=Aqq"N$p@\h)xY0Beέ F?p~3V`MCB?ZB35cН;n B U y$@O+1z?xd#=8Ɨt,n 7 p @_P?Y }3U?((NBtOd̂^( pD`C2_(( NhQ>.I%9e}%O&p؁D7*8]іF,A(O9P,V$IՎvJ?ؙ+=~,deG?>v\vؒ2H+1@;vlIBA; iJx!Ў[R4i%FhGЎ-)C@#^#?hǖ! @Z/,;2)C+: ./@7@rˋ A7EA " tCMQ.'?} 5CN<_5v1nh.Jqr%*ʥKun?ÑT P@79IxUP"I@:?0wx4 Q\U?D P6!- J7BtT'P\gaTPBR dIGdLb`>B{dyf|5/ؓ4&K+!Ў@45M(5hY,6.@yCfݨA%Q z!Ճޮ&"R4#[ 5(jiQj7W)}6`A5$}`Vzڀ P@[? )?P zEGLA +Or{ m%uAXO&zb}m[{ 0~:G%s+OHj: pF>*ae?$AdO&^r 6cP4fI'WD?4`N@0ׅ8?(]x2PGKgx祌j|OX0N w=nz!M1~+B ?8~&Izt!Є@?SV/(>n\ זẐ嶅MY?V5={q ZTٝT}myB&}"|= Ar"w>GZA(&ȉhA/&>n\qEtj% &T_؍ ! @ďQ?`}b%$Z'+v24%-n6iB@?(N@!gB?|:$ 4 M-˚!#?x?(EU +#*xI; d=L;o.Vf߸)WB8bqi'?8^wW3$'Pݞj(I dA0X)`bђ6@jo\)@6 @rPk,XW`{W@?hrgA$ r9nDp$+8H oh.gEҀ@mYHV&Y\I6Tܜ@]q;Fܼ d˿SuZǀCnM iPO\[̂[W729?0wjyY_P1 A`;`>mۓ_bu}" 1&!l'y1c`N@%YO=~R ~?SR\w 1j˥Ɨ4IK;@`;+Qߙ0~G/%hOF5`W).>jWe>xW5ڗ,wvy`]c_bfyԿ9Ђ1W@=?_AbaZ᛾Y)Rpl_k@|X> MZFg> G*#ac}[[,D =Xݧ*w~A}zHe(9@d*J?`AZ[4#xT%П?PGT-c@)kA敭 'A*kA^A@IO\Bp vn . eͿ~Fu8~Ό+ugpz@|B!p@5ꢼ\(k =Lm~ {Jn%{.WC?^t!  t@5|La~5҇# Ӄi8QkG@cGA/;bav!6:E.N ~_0LDr %P(%0ٯNh#Ӳe 8@EyRrr %<@?8"p?$ڲɺ: Plȟ\&p?|HJ( \z@!Ы?`@ g!+A%C\8|Z2 g!+A%C\8@ +w|̹#z%?d Ϝ;BW^K p>̹#z%Я?`±:C\%Я?`񾵎Jk'?89w@ qA|sGJk'?89w@:L8Zizy)m@OhDؗ#7wdO>˅ pnz;=L_%C!p>ypG<@' 7x c*r=`r9W`۵Q{- E\t5 A K?bGKu.ālr8twzGYR z@ZYr;n`? 6\wT#p4A&!$̹s!PŽa-Mʚ #PD f1u-zb ;jb# 0*kҁ@@EO4ǯK0ʐCO^RIAk8Q{8@MO?`zN j 7x; oArv0~d  k]Դd zq{ƸL@?@6+@`î"l$  ˕!žl$   !@aGr 6%N@?PV!l/_rzh9 zq PN@?OA@ 3!`4.M 9<'馵lB@?0K![4oY T/$'x^*q܎0(j ꉀ4U{=r7_[ $?=Tk H}M2+qǯߵ*STB!Aa=$K}p}QUDݎ]T7dR;v_tuA0r ΂@EyRrr pT 86o[s#0[4@}ZAea)*"p#37?=*~x?xFK!`1n>kh 2U@s3AV9@9 XUNk2x,;OK@kA^x`A+yNQY3QS7"ac9UsZ2\kFyɌ7D iom i7鉍l+Z4ݍ3eV-"^"ݔ͟.W5j [VUC$GrNG*EMs*M7yb7o[if1c \Er+sXj?fd15vFۢˮ;i՝ 5-eqXEd9pʊprV` EG߶۽pom@YLXɹϧq3Y›/5}$WDed!CAshg;߬]<4[0jc:KJeo#[TQe }YMM߯4h%g|ɭw\*w8T<Ÿ!#rwyr!\-J-x$O\m(a{~Eo>@?z 7Ӄ&įo>]/VL߳PAG8ӑN j2 y\iG[6r.$w,..pۄLngܤ6yUÅ`S#3/X|8hdɋO䤕?xB\ЃƵ6ߟ ~꿗%3Wr߯Qt'mzdM#}yK70~RB1MN3Oa؈HΞPG)C?oX9"ϭ| Ae7=ZqRv(?W $ߔ/J|`-䷱'ed.Z7j%-LNvj/?{lkJDWx㴓M'm& k,";O^;ߪLձmAmk [.ئٞ㺊[K|).ZnSRA~q6PKf*XP:o6L|#Ϯ>F M?zOP-lrS[#}\\-ZrMިjVO؂oQἊv=uWw7A> u#k_ziM7W|[ņU`*=lǢ}oޞ" }G|<Ӊ#{gEuT.?k727ߝRуQnXmǛĎEwNȤAY=ji٥)XvŖZ-ՙsnd}ap(<0eA4iI酕gFj~C݂Dr|z~ ɀ];*2Ȧ^zT{ 7||7WۚA @@9gBЃ0@9gBЃ0@9gBЃ0@9gBЃ0@9gBЃ0@9gBЃ0@9gBЃ0@9gBЃ0@9gBЃ0@9gBЃ0@9gBЃp.i-7+ٙ^bA rAz*bT@7mb=Pmz:=anʒQ--n /$\[k50,rn%A[bࣇ=07pz0? ÷( nCnZRW5~unA˷J?e+z%ۿ ?$,!R/B/=oyW`wӡƚ %T%܏a)ЃS0w|jo=AZTAKT"z E۰AA8pփJLb|#z jƬ@?(7.K]fu$x|ӊMsF ҆#z`NIF=Vb5zKڅy6c;B=8BnՃP :hj|Y[";?_P-_ms0CInN*JFO;uYV/ITCn Q?|P^FHDIX i!^}.aF*/ж@ZF܁*H-9@}A}%GO=ϔ! z  "Y ߅IENDB`A/DdM:&%bbn  C JA2Figure 2 us mar 08 hoursb.L<1-@[.UnS.L<1-@PNG  IHDRztDcPLTE3f3333f333ff3fffff3f3f̙3f3333f3333333333f3333333f3f33ff3f3f3f3333f3333333f3̙333333f333ff3ffffff3f33f3ff3f3f3ffff3fffffffffff3fffffff3fff̙ffff3fffff3f̙3333f33̙3ff3ffff̙f3f̙3f̙̙3f̙3f3333f333ff3fffff̙̙3̙f̙̙̙3f̙3f3f3333f333ff3fffff3f3f̙3fDbPbKGDH cmPPJCmp0712Hs*IDATx^[ EkUepOSɀ */orqNx";/@gG ;@ox:q&/@gu^׾7h@ǍaEF @n]y{7?n=6 cB>Y~ܹ>y!>*~6x~Ǎq/jz<>^l;VYWUeQ!%#P B$/jzo ùiuO x\B<CJ! !%=Xx\B<CJ! !%=Xx\B<CJ! !%=Xx\B<CJ! !%=Xx\B<CJ! !%=Xx\B<CJ! !%=Xx\B<CJ! !%=Xx\B<CJ! !%=Xx\B<CJ! !%=Xx\B<CJ! !%=Xx\B<CJ! !%=Xx\B<CJ! !%=Xx\B<CJ! !pIt8arY =n,%"d8&\8w,.Y 'pǭr l$pX|cHOOy!><~POOy!><  ;"; &M@[@ox:q&/@ox:q&/@ox:q&/@ox:q&/@ox:q&/Yw8?55A x XS ϠN8j lq}=Ňc;u@ǯh9==^6fmRj@C_0?SKЌt=A: 1?# %nx\wP<^"M!P"K qC=JxC@7<;~@/s&u(%B|%xD!YGu! j䁀x\OP  <.F! j䁀x\OP  <.F! j䁀x\OP  <.F! j䁀x\OP  <.F! j䁀x\OP  <.F! 80$\fx|322@@<*\fx|322@@݆v{ sc8!Z lwL]I GЭ&#yi{x\w@}'6xI{TFpx q.C3y=~{L|^e>5Fqe]i:8[ۇx.Cq2_r3C{h7i_NmՕu*% [Z t'5W 1oryѣq?3w܍hZ7~|f=]/7 1ͽ1z+q:4׏ybξsffin@:&^s{|ܨn}/?w|1WѴ Dn>ӽ3YW" q.i/{(a "38 gH]F=}R[zFiuݵj^qxd@8c%#!qW6@%qzmtxxd?ý ˓ PG`qC/fͼ.Tx|,B`7s?6n՘4) 9 '彳}Rq,㵤Hs#uE7q= =sbb5Q !Ž3Fهeϳ]qR* , eK)e%uH} u_"K`qc4) ꑧՖO?GnDƍҹD)5?XSJM<^C4؛@j/_}˹zMYq7rA3spYSn<H&߁)y?ڸ'`ʂ@@j]}!pǿL;`Έ8u^@fO$&Jy!\W} ă4؛@㯽k{<Hx[yS< !3S)x a\}Gjx !W6DCzqYzg^B`Gs?oI2EADg(85׭{g'FB`q?Uz9)%&sR۞/<'Mʂ@-3^(r礘m:cM#rBWy!>S=dB4?v_qC'{6]zKll3B`A zxG|r嶱>ǿ_*@dz^VEe%yǧN,<> B2?=;O2?8gq4>-uψ'uB`I`ǟ?(_aOqzZ q/xYnx E8}}8En@ <3^C7z/pJ  ;$ ma-w_h>B?ޣ0;ׇKz{ 15y{i>=ڠϸ61;hAd%73tmO@qv?x\/=yϚ[K#ۍ4ʮJsvվCufν3FU <>nT`;f]}vE@cmcVHQǍCy nYWA B'廥gD:!& vۙ3LAs-xf<|<~f*i8i}&/@oxz=֝EEz!p <@~LȩbNs/G%3_ SN$OO8@Sg~@ĩj"N{F-3_gL 8?>uC8.F_,4rZ8 2yp?Vz@]*DOګ%-ȪA\㥑}.Za oU DlТJǿZssTךq+L'."f֭~|ͯ"\?A"__H;!)fwx|wzh"w뷂ldw#r= < Bޚ62Zje|@/Qth)g=eʚk5UDڙ+:y]K{ʱ=1淞QۖEPBCźzE)$kU@y5QgmM&̶lo[1vͽ\%=^1uzP]eT XjTeEe2|'񇟐x3JSu ݑFFE9sm:{Y2juA1 9|a={cw'rKNkTx*$SeEZXv}l9koܺ{ V+ڱLp;rߩ*\V j3z.{.&Ee9n܎o,>5]4OǸǷ;1YD,,yM@։]<큑)v. Pv}^V<^#Y XNWq?j:^M5T7[]vv<~G3ۥ߃w%KF^pi%KF^pi%KF^pi%KF^pi%KF^pi%KF^pi%KF^pi%KF^pi%ļI=M=}J>{*{J(}f=&J^SI$JvTr\$0O&aHx7:&׹Cɾ=~܋4NqnaD^ǜ"̩,74?ʻ)MFLmJzXYJ"ZLDJt(z~Luj I9I2ɢuD7*t1MsB};,ٹi<VWs*Lnq坛-h0Ĝ!giwYk$i\}QAPDx(z~Lu&)zVPBĎ4:LD&JӤTuKsir;ě=eUܴ\=e%%?n$Wä Xy QR{XPd?.(m"pWP^z L -$fꕈ0MRAF;)=m-63Sv5nCռ>H3y|}&%}VYB{XPAZ&u2>TV ʩ6Y>M`JQ=UQ= z-b:LVRdU6,&JqK4_̉ %ג>W>IW>DJSYceYྠ9Ջ'tPWiAeIʒUDDTig%T-z|<+&)q`kfOe;%JSć}eyCAiկ 34_DEU%UI*H^X$E*K3xi[0,]{4oxOe4y탈C),,!QPm t\AÅU M9iAI+aYe A<]ywu ]W +\kfmxJt]>Ud{yJSYSe q™M%M\B7T!H+H^$E+T7b\z<}vM]fL-"uMi\iiѡҔA4Yj_=Ti6Mfc4Rr dHܾl:{:Q- /r[Zr}ڂ.1oBLDJt(zZ,MJ,QTiҨÂ*KwmcC|lyo|w<~/t ȥejw0,&~ǎ-ϒ߿ø8Lެ,J' PH8^xztkEo%1Avo `2?mOfWad|5գ'VV#xlNgW٧9ĮӅknq?l÷u6-ͦܚ[ЀPc8}G/DZVV#񸽋=;3CX+ٚY4[;ML8n[0|7T{g3u}]n!p֜ժ6dwET1ϒi{Ú楏i9ZjF]؇߇#@UxȠ F$sIENDB`\+Dd_7B'\\t  C PA8Figure 3 us mar 08 days stdb*ݳ;=L Kbp*nh*ݳ;=L KbPNG  IHDRDpl<PLTE3f3333f333ff3fffff3f3f̙3f3333f3333333333f3333333f3f33ff3f3f3f3333f3333333f3̙333333f333ff3ffffff3f33f3ff3f3f3ffff3fffffffffff3fffffff3fff̙ffff3fffff3f̙3333f33̙3ff3ffff̙f3f̙3f̙̙3f̙3f3333f333ff3fffff̙̙3̙f̙̙̙3f̙3f3f3333f333ff3fffff3f3f̙3fDbPbKGDH cmPPJCmp0712Hs&IDATx^k:Fqֲ98|:yіLx|]6]@::Pb,J{O[^/1ũu% )K}tqK_֥,,5֙rޯ{=ץ'XhXh?`/W֝Ok ZHXf3[iԺT -nhl ?ֵ-,2Q1V*%Ɔ/]wa|Y`-6}altA#xC_׮5{/ .+@\umunQdž.[]UՌ'Ί[Vj~s_1nVWn@=c;S1aE^`a(5Zx%#l0hI[ƊX'mQ+Z86JcE `h(5Zx%#l0hI[ƊX'mQ+Z86JcE `h(5Zx%#l0hI[ƊX'mQ+Z86JcE `h(5Zx%#l0hI[ƊX'mQ+Z86JcE `h(5Zx%#l0hI[ƊX'mQ+Z86JcE `h(Cg:iCkG~]8v4J@c(ktؑ4_8+'fKPcof?X%V+Oޚ0VnDmZb.EÎ&Pf/=36دƎV {z' f޺ukc;@}T. PΉεcubG$&A,?Fcl.?"벼 aJwSLqi1QBƺ-M1lbl|HviOÏ|c ,2݈@q)fzXsjc (6M<60clb$!M@#Ji1FFi@ c{B]#\uf5q"cq5yP:Whzn@'!=O87"c{gܕ&6{ 'LB[~[U -cAk,LcG `0.$N0Vxn,:) XAk,LcG `0.$N0Vxn,:) XAk,LcG `0.$N0Vxn,:) XAk,LcG `0.$N0Vxn,:) XAk,LcG `0.$N0Vxn,:) XAk,LcG `0.$N0Vxn,:) XAk,LcG `0.$N0Vxn,:) XA^S]\+-  @W{fl|-,]\'6@$PdO\K>@}t~)CX!V(26;)]3 Dykc9 ;4M |c($P؇7f4 5[X-v@ coݏ}Ba, `=L<^׮&yڨ4.rcΐgM[cl6"PlEh+nT-@7/4?:3O 8(3ֽ^?{:V!Pfl).'7'6"PdN)i(+ި\4k@Û48|F'$߳민+6?@k;:'`T<5? EcE `!)+U.5OcHXryk~@mxNĸK:Rlc_&J{XdH`lE4eןMEߒc?ܣ}-ƚw$-eƺ3-[ % Y!tbEvX`x__LjF[8+ycWäb=k!EOay]@c+@ [=Oٞ1vC4&%^{:3z 0KZ2B`X{/eY7S*HKfۧMT]8VIc;[XΊ- /rNc#AlH`ؾ#'qpV\(5O`;f$xzOت8iϳ4;3O9V' r - Jc[,yI`1֧KFGc8!ycƹUo U ;a9V%Occix>cPbd0 - J`~;kry^:mg~4uOJYB`;ޠ u0[r#`7ۖYHl2 D`W>TVbmcÍ߬YqĶ"0o`l7a.yc7 vG asM'͗"0ol:+ƛ;ɨ7oW-qH`Xaź*^×)rj)K#ዓOm,YmA=wK2=JLM7O*܃*]}禩oalScd Pdlc'zze]SGIiEd??zjcK$ycS?#MǴ{>&Xw:"6=NH,w?-UnP%lE`vM0WUYiYˍ}~%}7Ūc,6;Hlc>ߏtV|npZgij|aᗗn1(2ܘrcO w8Ņ{6C '1,riqp_1# (2y>3Ђ } | jťjc)ƞ40a%ƦgߧnR,ȖxjΣw0#ƎMug|[P9vI`.Mu p>ƺwYL 76>/~ c.ycy&A#L[ĺ غ>IǺ| U*z އ˺~RSP"Pjm2[wt~)XB Pf뜲1K޹u=E B@]?3yi r6zLc9*(3gl* b?5jݏKkYcT=5 T36׫Ʀ9+> smlc]t&i~8ou抅G @5cq`CG(:Qf-f< B?7jC,DL:-ܳ믨+V:~rc xjڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrm@.]ק_!{DO@.y|KZ c&PjI>~:1#/{:lʺ1 rev-xjc|($Pf?&cYsjc (36TY12 cʃ;Ih1&xݿF("p^FT36?_o̜(8"T3߈oņc?'{~!Pدy⤸D";f=oIqNUv3z>3 cHdrc xjڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrmڬ;YXMkdJcU+G6 `ͺ*Uq$6N֪0Vrm@Ecv 9n,&P{f/#_ c׉ ! 3ί;kcXI3 Dykc9 36](?-1BՌ>6 :V!P[xMFw?.G]gMv3QUi]ՌMuش1ƶ;l# bFբYlbl|' T36W64~[X-v@5coD7&牁T3vb_%JG& 36~^ǿ{ٝ|1s&Ih00v=30Nc#b#\5q5?][07NHc(k|ZG{Eصb{5J2X6yk~@J`XCR0V\kƚ"R"X0*'`T<5? EcE `!)+U.5OcHXryk~@1R`'p|5#j,_5X5{!@R5uh! lR0V\kƚ"p\J=2W| -ZhO@PcFlnV6XF!i /WCOg/^_pwVM#}:~:-Ό "y|(|h8m0msG`sc3Gbl'/N?B) n^GY358=]a(sǗW_/=)Yjz=HC#ULJŧ7q< W8OILߩn?HI7b|k2kdߎޑ4j>c />Uy ZYdWOnOZs6|clzy;+<˸L2>z=?:F$o'_N^#iF̍ߒ&R8+c7IWZ!*d|+s+i}yVHtJ+7?J\+J]|HCzuCc'[)Eٝʶת0NWh'{H8  `,#J0VZ e @@*UX!(Xj+01%T-b2 DcEX`RUBcP"J"Vlmce+;Zej}a-[#ǞކkiuO?'hjnCkKnrDA}2u6w-!tӶUuL?ГwCn@%tOt]S^QןG!2&^Pc ԟi[czlUc,GpzW-:$^{SM۠Y 񴭱=VCctq3Ƌ<Ӧ޺~fiӋd97߀zt~6{p1gpGU<Çz;Q|>xߴ3]Ljć]Q{[O}HuWl=̉kN&ouq߄svgc:/c3aix1yv;q~U{`T.Z{pc9ޠj,zn׺Q_/x|mc{nGymk؋3ҷ?_{_5zJw*@{=b`1r9`Fz"ÓCM%npou8##w {|d멋xpf~Ύo<T{Ys N<S_oֹ捝=SL+{pAu?h8&Ga)]CnP|yPcؠ^+WymMbw.I8G/;AyND_=PcX\aW:pLKaױ2^m MqO[gYW!ÃKvy̺\q)&\qn =֙N\=+^`lT.Z{pFctr ۴q<9 bن]pVmT" l=\HuW,;|v~inh*۫鞧pGӛ-i<{milyJ0^h^^ƻ&/=ƞpi$ `쎰v&~-V7\О6:;~aZi~K0Dc T  XC&`lE$CL{?MIENDB`)Dd27%t  C PA8Figure 4 us mar 08 days logb(f, /ܷ3(En(f, /ܷ3PNG  IHDRGZWPLTE3f3333f333ff3fffff3f3f̙3f3333f3333333333f3333333f3f33ff3f3f3f3333f3333333f3̙333333f333ff3ffffff3f33f3ff3f3f3ffff3fffffffffff3fffffff3fff̙ffff3fffff3f̙3333f33̙3ff3ffff̙f3f̙3f̙̙3f̙3f3333f333ff3fffff̙̙3̙f̙̙̙3f̙3f3f3333f333ff3fffff3f3f̙3fDbPbKGDH cmPPJCmp0712Hs%@IDATx^[<@yc蛍?ؠ0Ju-BR5+s0I ]C V1 o뵘 Dz\$^@C Nק1/k9)dznwkYr:<^7e0^IH! @^W/VG0T|/&F:;]}ؔlTodI[5w}Ptzw}o'rTuV#CRÒ6&J-bdծF{!&ft3YZ@YNZgW)Xڬ"uM5,[3^:yo]ӱ jθ'h h@Ws:$IaGDF DzM]BvgV}`8]WIO'fu"F:^i-tm"NL T"u6kL̼GdMkKٜ]?ZfޖV0h]}_#af׍U3Oc TWڋ}>ڕ9B4]\0r[rY.eߵ\Z)R׋V/2~l 7lOܮ gC.IW\u}%(1q]l .^p&3Dž+PwuM~s56üUq>:%<#'r`͓*k$5c̮p]V]#WS2@% 60.@sеF'a'?ʑZ/ȁJZ=R?$}ן a'lH]K=uQfZ;ZivMnWEO, i"R]n /jbߵ1B՝1rfJG Dz1x!$2n@_M1pd@ @+D| D7_xeve8B @kAR"ua ^3}݌UlYrʾڝ]Oc5'噗}ĩҙc8q]lxD۳]=$qkM'Ap}c!a' t]ۨC(5Ŷ`ߵ]bQ%u;@Ztyikm'[t36/Nd㝦atMQZ+36/N^̺>g_pUk?vWe g7t'Wg$ei ɽ&֬?}"Q}0Znܯ0[@F]u fr5?k[Z-@F]GCVpgv5<6NڷWZoF]t;ٵȸ8jߞ> l`G`]Tݷwve8Ku2)އ{sⴤ[!4rċD]Gp?.˷87-2GKϐtV;+(3iKZ]%U+k'_fͅsX>6[}>,fU@T̮͗ ]/Qw-\OzkAt.!pZ@]+@K#縱*@ 9n еt9zkAt.!pZ@]+@K#縱*@ 9n еt9zkAt.!pZ@]+@K#縱*@ 9n еt9zkAt.!pZ@]+@K#縱*@ 9n еt9zkAt.!pZ@]+@K#縱*@ 9n еt9zkAt.!pZ@]+@K#縱*@ 9lX ](u@׮'PL瀮I}(>_{0]nt.&N/]%.hkz~̋l|iq{n:ٺw|,G RHב 2/=Iuoz˲rf&:HL|C9Onz˼"FcFWԷërgTf&r1VQ.zRI)VH)+ I%ea(fk2eװTRfqRVrFzRI)VOYC_XO*)3E#1_)+/ʨQQe!&G`*xj8 &G`*xj8 &G`*xj8 &G`*xj8 &P@׍Ǿʂ6'q} *+9,e%N([}5$.;7[o4,e%$]Ǿ6:WÚx w=;tVn:DPq屯9d6$~_~6QD't00̺~y쫜$q'cBVFI$„2屯rt }Y(":DPf]LT$\洒vЁ$„ $th/@"vp`6Y~Iqf803EIqޞY;F_S[B^F;IH@컞{ <2F[IH@aBƩs`Wܪ -vЁ$„5Φy_NJYbNKLF{IHHAB܄We!.$D$t$ uv^C%NZvС$ ]/!>@3еR!>@3еR!>@3еR!>@3еR!>@3еR!>@3еR!>@3еR!>@3еR!>@3еR!>@3еR!>@3еb)s9xTJŠ" _ux ⋮C \ 2H&xV3ibS#XJ$tMI`[޻} 56l` V}qtXi_aysnú{&{ӣ?*vMַΞXTSki/VSغl%Ev6O:nkx6~݌CQc3[5Jd߱fY[awK,^͌E$fӥ5V { [tJE"F׊e t5uXm̐8O[/3M[yl?p:]{{z9/qwL$oø~5Vl'7Ev˶u :nMYk8gZY~\Ѿ 44bf%>dy]_omJE"F׊e+?vsߔ;ϴW^˸k3sf\べnyz[CT/ktX@W3ۯ18'=j2@߆P:\:S[R_dZl(vGݞڣxk!)5YLAgusJE"F׊eF qtq66N5/qc$w6q=<.trkzH_bZjGw1=jpi'aw0w>2຃f|7X:pۑ]7[]M"|]kŲͺZYIcר]ug:λk+ܧďdvzLKKl λFٺ;NJE"F׊eNgVsK}Ozkx@u$)il>NlǥJ5]s<ֲd_ _{Z>JE"FVvV 5%̈́mK@H]f/6A%&i|4(j]*'M]/uE?@{0WIENDB`$Dd0$'|  C XA@Figure 5 dialog lengths april 9b#XlQ|kp#n#XlQ|kpPNG  IHDR5VdLPLTE3f3333f333ff3fffff3f3f̙3f3333f3333333333f3333333f3f33ff3f3f3f3333f3333333f3̙333333f333ff3ffffff3f33f3ff3f3f3ffff3fffffffffff3fffffff3fff̙ffff3fffff3f̙3333f33̙3ff3ffff̙f3f̙3f̙̙3f̙3f3333f333ff3fffff̙̙3̙f̙̙̙3f̙3f3f3333f333ff3fffff3f3f̙3fDbPbKGDH cmPPJCmp0712Hs HIDATx^[*E8c蟍?cyyl"k*+\Cg^:us4 pOxhljưt~0Z ٱJ!ě5x|Nkh/fk$x#@jXK [j[xyA2_"-5YxH8iH(&m|pq{ڄPwԜ!5lK f -@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@xp ZH@@~ ԘAy !6_LhGkL͙ 4ϭ*S @j|J%F @j|J%F @j|J%F @j|J%F @j|J%F @j|J%F @j|J%F @j|J%F `jxGV^jU@;:\jIT\jF;:^jI`j@ƇOT"@j@ƇOT"@j@ƇOT"@j@ƇOT"OtVXGvXsMl1_}EGqXTRcGi=9NZCjq(?tnp^Ƅ‚X h q'H Oa™kBR| 5B}1]t{Cc+u^rtrEMSö5Yz@}C[oj"mMMmhojΌ5v6Rr_Se77ԘRAsHMs`3VԘRAsHMs`3ڀ|3_j27 @"5d1J!3qb2!3qb2!3@45<Ad=s O5g 5D 3 HsFj.)<{#'{3,R $ 52V >tcBeDk&h"B 27M*56~#zHר;>=ܮj"+4=Qd#Oi~IjlQZ@8ц蹎"hjkn3?JC%M7K(r $ߪJB: cM^*<_y01~}C'rȈV͇fܽ,gUf5+`|(Y.w/kh{,[ZRYꭗ[ZRYꭗ@<5g|P̚z{O-K9/clHM}vGSԜ9sz{O-K<)p;Crn<_jߠ1kpoCj=x%JM@;bc$Jf3 E J ^1\aN@b?K; N įЌ:=p_㼋 5VhJjQ#v g.ޛ3-+F6';m=oiR;v،{ARAvJ9J!cft r_SLzgl\(K 5s^KX (e%>!`lJrcSm,n;9!'Cf,QeYk0q ;n^5&|$ޣ LEj,(H Z: ;tNNi@ c{3kvN $wGՆ5 FRׄ˳M0kTYFR\A9uGWh}k S3LܬxЌ).G 1|_ɚg9T@S~ѺÜ=?}Y:pvlǣ>Ϸ#O Wy7mOv_X3 kT5}"L"5C{)Ýz&=5hnx h_F fX6~Uڭ_oGn?)%!#啕]&wslhU! ϣ=^HMc9rWf0tI͎Qu̡I] 5C|vN>dդf?܏mٚInyf`7+޴8[3cM޽Skmn35{uFS3紽kl(K 6bF躏m0Cj9C)]ssh!5l%cMn omPJeMR>ʶC?Ymt)U'e0>F۸Qesshv6RlϳeFq DS5[Ƹ]-\h,lFGAQ9Qe֙+4Qۑe"@jL($hjS{9 {<34٩otj6`|Ӵ9P <O n2ޣx%> g9dQIМYCٵ懰<ٓ?.~{sS.۹B#4$R:klRI͙-l]Ht45o}&5g&l]Htbmai3I r(({qRX+TXI#&SW&>vɰ[mLR#c6B2HqE| S.MjJZ Ij\\jQiȊ0֔&5q_ޡԜ“aff)F ;8(R\͐˷py@ z`g *ɝ 1̎5Wvvʀ*"4vvqc'0u94KԠE@ -?Pk=F-Fx @j궃&5s{5 6עoP'5t 2K 10f~hf5TC 6/!wհ!ZQXCjikZodf7 >?SN}Qh#Mi_ @iO; D $W ?tbHZbRCZf-1@`-RC YK!@jXKԬ% 5nl%;'IX=l%o6RC9Ow[ ;W#wƚrΐrgp{qR%[%@jZuvo'@jdHMHvvliyڽΎ 5:O 5QUUivf;;JJԴݟ]_k{\~gCJn^swQ5Gnzeu_xc<898Ss/#=}9ׇ{cJwPlHDp17iG(5TMjvt1Iφ 9tsH@>rLOJ'oL콞;"jRpnzj2G~?+r{9o.9$ESEӃwDRGդfG_S],Ix15_Mij??}I[;S tWMjv4oӮ>d|!7^zKqF;&JM SԌF,f gnle$yMf>kfƚ>fbƌ0^}9v~}MJk2$>kh>nQO9g "R!ᣞZDf3яY=9:{X]MeEHf~~8x(1Yw}PcJq0Q4@B Title$a$5CJ OJQJ\aJ4@4 RHeader  9r 4 @4 RFooter  9r P^@"P R Normal (Web)$dd[$\$a$CJtH 8O18 Rapple-style-spanBOAB Rapple-converted-space>@R> R Footnote TextCJaJ@&@a@ RFootnote ReferenceH*~@s - Table Grid 1z:V0jj $a$6]6]p@p - Table Grid7:V0$a$>@_5_`k [ @UN o $*--T2 9S=AADoEFFGGOG8KMFNNNNOPOR{RRESg?g@gjjkkqqq!r"r$rrrrrrrcssuuuxvyvvvv w w wwwwww w!w#w&w*w+w-w0w4w5w7w:w>w?wAwDwHwIwKwNwRwSwVwYw]w^wawdwhwiwlwowswtwwwzw~wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwx,z-zvzzzzzzzz {{{{%{-{5{6{C{K{S{[{c{d{q{y{{{{{{{{{{{{{{{{{{{| ||||QRց)-156CFJNOVY]aborvz{PQ@D\ޅCfφ'>BaވɐPD$ۧѱڱBڲ-г07=vbcַnXݺ->OP[|\17-' 8SpOhEoI8*P\hGsH}$P|000000000000000@0@0@00N 0N 0N 00-0-0-0-00A 0A 0A 0A 0A00GG0GG0GG0GG0GG0GG0GG0GG0GG 0GG 0GG 0GG0GG 0GG 0GG 0GG 0GG 0GG 0GG 0GG 0GG 0GG 0 GG 0 GG 0 GG 0 GG 0 GG 0GG 0GG0GG 0GG 0GG 0GG 0GG0GG 0GG 0GG 0GG0 0 0 000000000 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0000(00000000000000(00000000 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 000000 00 00 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00000 00 0 0 0 0 00000000000000 0 0 0 0 0 0 0 00000 0 0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000@000@000@000@000@000@0@000 @_5_`k [ @UN o $*--T2 9AADoEFFGGOG8KMFNNNNOPOR{RRESgjjkkqq!r$rrrrrcssuuxvyvvvv w w wwwwww w!w#w&w*w+w-w0w4w5w7w:w>w?wAwDwHwIwKwNwRwSwVwYw]w^wawdwhwiwlwowswtwwwzw~wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwx,zzzzz {{{%{-{5{C{K{S{[{c{q{y{{{{{{{{{{{{{{{{| ||||QRց)-156CFJNOVY]aborvz{PQ@D\ޅCfφ'>BaވɐPD$ۧѱڱBڲ-г07=vbcַnXݺ->OP[8SOI@0@0@0@0@0@0@0 @0@0j@0j@0j @0@0Y@0Y@0JZ@0JZ@0JZ@0Y@0@0@0@0Y@0%=@0%=@0%= @0@0@ 0@ 0@ 0@ 0 @0@0@0@0@0@0@0@0@0@0@ 0@ 0@ 0@0@ 0@ 0@ 0@ 0@ 0@ 0@ 0@ 0@ 0@ 0 @ 0 @ 0 @ 0 @ 0 @ 0@ 0@0A 0A 0A 0A 0@0@ 0@ 0@ 0j00j00j00j00Z00Z00 00lj00kj00 j00 Z00 Z00 Z00 Z00 00 Z00 Z0 0 Z00 Z00 Z00 Z00 00  @0 Z00E @0 00  @0  @0  @0 Z00Z00Z00Z0 0 Z00 00 Z00 00 Z00 Z00 EZ00 0"0 Z00 Z0 0 Z0 0 0#0$l0#0 0#0 Z0 0  @0 Z00Z00Z00Z00 0%0'&Iij030j020j040j020j00j00j00j00Xj00j00Xj00j00XZ0 0Z0 0Z0 0j00X0-0 Z00Z00Z00Z00@P>> Z00Z00Z000Z00Z00Z00@P0>>Z00Z00Z00Z00Z00Z00~> Z0 0Z0 0Z0 0~H0|Z0 0Z0 0Z0 0@|8 |Z00Z00Z00wZ00Z00Z00@>0Z00Z00Z00Z00Z00Z00Z00@0Z00Z00Z00+E0 Z00Z00Z00@0x|Z00Z00Z00@|8 |Z00Z00Z00 @|8|Z00Z00Z00@0Z0 0Z0 0Z0 0@0Z0"0Z0"0Z0"0@0Z0$0Z0$0Z0$0@0Z0<0Z0<0Z0<0@X0Z0$0Z0$0Z0$0@0Z0l0Z0l0Z0l0@0Z0$0Z0$0Z0$0@0Z0~0Z0~0Z0~0@0Xj00j00j00Uj00Uj00 j00 j00 j00Uj00Uj00 j00 j00 j00Uj00Uj00 j00 j00 j00Uj00Uj00 j00 j00 j0 0Uj0 0 j0 0 j0 0 j0 0 j00Uj00Uj00 j00 j00 j0 0Uj0 0Uj0 0 j0 0 j0 0 00'00&j020kj020kj00j00 j00 j00Uj00Uj00 j00Uj00 j00 j00 j00 j00Uj00Uj00Uj00Uj00 j00Uj00Uj00Uj00Uj0 0 j0 0 j0 0 j0 0 j0 0 j00 j00 Uj00Uj00Uj00Uj0 0 j0 0 Uj0 0Uj0 0Uj0 0U @00 0d08ej0d070000 00 00 @0 00 00 0000000000000000000000000000 @0 00 00 00 @0 00 00 0000000000 @0 j00j020k@0@0@0@0050050050@0@00p00q0'0q0%0p00q0$0p00p00s0 0t00p00t00p0@0@0@00u00p00z0@0@0@0@0@0@0@0@000 h00 00 000000000000@0@000j0=0@0@000 j0000000000j00000000 00 yyyyy|P%QPbmAn5oz΃_)~I\cmInoz  *4>HR]hs~ 5CcqǃR5NazDސɘ[E +.=bv|!`=#]l=#],=#]l=#]%=#],=#]=#]>8=#]=#]=#]l=#]l=#]$f  =#],!=#]b "=#]$c #=#]dc $=#]c %=#]c &=#]$d '=#]dd (=#]d )=#]d *=#]$e +=#]de ,=#]0 -=#]0 .=#]1 /=#]D1 0=#]1 1=#]1 2=#]2 3=#]D2 4=#]2 5=#]2 6=#]3 7=#]D3 8=#]3 9=#]3 :=#]4 ;=#]D4 <=#]4 ==#]4 >=#]5 ?=#]D5 @=#]5 A=#]5 B=#]6 C=#]D6 D=#]6 E=#]6 F=#]7 G=#]D7 H=#]7 I=#]7 J=#]8 K=#]D8 L=#]8 M=#]j N=#]$k O=#]dk P=#]k Q=#]k R=#]$l S=#]dl T=#]l U=#]l V=#]$m W=#]dm X=#]m Y=#]m Z=#]$n [=#]dn \=#]n ]=#]n ^=#]$o _=#]do `=#]o a=#]o b=#]$p c=#]dp d=#]p e=#]p f=#]$q g=#]dq h=#]q i=#]q j=#]$r k=#]dr l=#]r m=#]r n=#]$s o=#]ds p=#]s q=#]s r=#]|; s=#]; /JJTTeeeeff&l&lrrwrwrrr%s%sssssTvTvvv{{6{6{d{d{NN,@@66OOXWٵٵZZ<<Eff$$r GGKT      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRTSUVWXYZ[\^]_/JJTTeeeeff(l(lrryryrrr's'sususVvVvvv{{8{8{f{f{PP.BB 88VZZ]۵۵\\DLLnn))t''JSZZ    !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRTSUVWXYZ[\^]_ 9$*urn:schemas-microsoft-com:office:smarttagsState=`*urn:schemas-microsoft-com:office:smarttags PlaceType=^*urn:schemas-microsoft-com:office:smarttags PlaceName8[*urn:schemas-microsoft-com:office:smarttagsCity:\*urn:schemas-microsoft-com:office:smarttagsStreet>Z*urn:schemas-microsoft-com:office:smarttags PostalCode;]*urn:schemas-microsoft-com:office:smarttagsaddress9_*urn:schemas-microsoft-com:office:smarttagsplaceBY*urn:schemas-microsoft-com:office:smarttagscountry-region `_^]\[ZY_Y^_^`__Y[_Y_Y_Y__YY_Y_Y_Y_Y_Y_Y_Y_Y_Y__YYY_Y_Y_Y_[_$YY_Y__[_[_[_`_^^_`$_Y__[Y_[_^_`^_^`5]a  + 5 = C L R '.iow{09dj4 8 < D V!Z!!!""""$$''-'7'''k(o(((z))))))**k,o,w,{,,,w-}-/ ///001162<2E2K2P3Z3e3i3333494555555 66a7d7e7j7q7z7^;h;;;U<Z<<<<<?!?D?J?fBmBDDLL]NbNcNgNhNkNrNtN{N~NNNNNNN-O2O3O7OIONOETITJTQTfThTiTmTnToTTTTTTTTTTTTTTTTTTTTTUUUU"U&U?UEUMUOUTUUUVUYU~UUUUUUUUUUUUUUUUUUVV=V?VX&XqXvXwX|X}XXXXXXXXXXYY2]6]8]D]IeNeXe]ehemerewe)028<HŁXZŅDžޅ߅ '*׆܆  *+ψԈ~\cnsčMSѝם*1]dlr{|%)ʮѮ Z\ڲlr˳γU[$+OTZ\3:[^_b5;ܹ9JҺ׺ HPnw»ɻ!&39HLMQ&3[_elu| */C\bu|157;Q\-1>HQ['+!dl %8>]fp xov~%z IMR\)3ai!(*1CJ)/3;GTZ"06cru}X^}$+?G8@PWaeqv|?G_4]a3 5 V!Z!!!""k(o(**k,o,3O7OETITSTVTTTTTTUUUUUeUjU=V?V`XbXXXXXXXddqqqq$r-rrrrrtt[u`uuuyvv-z5zRZ[QYޅ߅ ad%)47psڲ@B ѵӵMOZ\6:8;9;<?[Ӿ./\]{|ҿ$MC_\157;-1&'+F 01Kg-pOX<Epy7@BhEo34Rg Z~kp?@jh2`S)hFGvB,R \~|ORb333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333N o GGOGxUUVVaaeee?gqqqq$r.rrrrruuvx-z6zvzzz|ց@D>Bɐѱڱ0ַ߸IZs H^b fd4VT(yu R(v{ dǺ(zgTLDZ @/"jD=BAHBoEX&2)XnTYAel zdyn{J~d*)h ^`o(hH.h^`OJQJ^Jo(hHohpp^p`OJQJo(hHh@ @ ^@ `OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohPP^P`OJQJo(hHh ^`hH.h ^`hH.h pLp^p`LhH.h @ @ ^@ `hH.h ^`hH.h L^`LhH.h ^`hH.h ^`hH.h PLP^P`LhH.h ^`o(hH.h^`OJQJ^Jo(hHohpp^p`OJQJo(hHh@ @ ^@ `OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohPP^P`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohpp^p`OJQJo(hHh@ @ ^@ `OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohPP^P`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohpp^p`OJQJo(hHh@ @ ^@ `OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohPP^P`OJQJo(hHh88^8`OJQJo(hHh^`OJQJ^Jo(hHoh  ^ `OJQJo(hHh  ^ `OJQJo(hHhxx^x`OJQJ^Jo(hHohHH^H`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHh ^`o(hH.hpp^p`OJQJo(hHh@ @ ^@ `OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohPP^P`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohpp^p`OJQJo(hHh@ @ ^@ `OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohPP^P`OJQJo(hHh ^`hH.h ^`hH.h pLp^p`LhH.h @ @ ^@ `hH.h ^`hH.h L^`LhH.h ^`hH.h ^`hH.h PLP^P`LhH.h^`OJQJo(hHh^`OJQJ^Jo(hHohpp^p`OJQJo(hHh@ @ ^@ `OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohPP^P`OJQJo(hHh ^`hH.h ^`hH.h pLp^p`LhH.h @ @ ^@ `hH.h ^`hH.h L^`LhH.h ^`hH.h ^`hH.h PLP^P`LhH.h ^`o(hH.h^`OJQJ^Jo(hHohpp^p`OJQJo(hHh@ @ ^@ `OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohPP^P`OJQJo(hH zgBAoE/"v{ AeDZ fd4{J~2)Xzdyyu                                                                                                             ! G n 9-<%&)&.+ 68:NRh%^ds^;5]-n@R1-8yM-J/hDdH$ee'e8e@eOe^eneeeeeeeeeeeeeeeeeeefff-f1f4f6f9f=fAfBfIflfrfvfxf|fffffffffffffffffgg ggggg&g,g.g2g9g>g?gvvv w w wwwwww w!w#w&w*w+w-w0w4w5w7w:w>w?wAwDwHwIwKwNwRwSwVwYw]w^wawdwhwiwlwowswtwwwzw~wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwvzzzzz {{{{%{-{5{6{C{K{S{[{c{d{q{y{{{{{{{{{{{{{{{{{{{| |||ց)-156CFJNOVY]aborvz{@D>BEN.InstantFormat,w<ENInstantFormat><Enabled>0</Enabled><ScanUnformatted>1</ScanUnformatted><ScanChanges>1</ScanChanges></ENInstantFormat>@P@UnknownGz Times New Roman5Symbol3& z Arial;& z Helvetica?5 z Courier New;Wingdings"1h(׆&B#z#z$4**2qKX ?R2=Public Dialogs in Social Network Sites: What is their PurposeThelwall IT Services<         Oh+'0 , @L l x @Public Dialogs in Social Network Sites: What is their Purpose Thelwall Normal.dot IT Services24Microsoft Office Word@ X8 @@ i_#՜.+,0< hp  University of Wolverhamptonz*' >Public Dialogs in Social Network Sites: What is their Purpose Title  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`acdefghijklmnopqrstuvwxyz{|}~Root Entry FbData g1TablebWordDocument.SummaryInformation(DocumentSummaryInformation8CompObjq  FMicrosoft Office Word Document MSWordDocWord.Document.89q