Effect of external events on newcomer participation in open source online communities
First Monday

Effect of external events on newcomer participation in open source online communities by Raktim Mitra, Aditya Johri, and Oded Nov



Abstract
Newcomer participation is a strong determinant of the success of online communities, particularly of open source software (OSS) development communities. Most research on newcomer participation in online communities to date has focused on internal socialization mechanisms that shape participation. No study we are aware of has looked at the role of factors external to a community itself in determining participation levels. In this study, we investigate the effect of company announcements made by OSS steward firms, events which are external to a community, on participation in newcomer communities associated with the OSS. We studied user activity in two communities, a Java newcomer forum and a MySQL newcomer forum, around two announcements that affected each community. We found evidence that company announcements had a substantial effect and participation either increased or decreased. Through analysis of secondary sources we conjecture that announcements indirectly shape participation by influencing newcomer motivation. Announcements perceived as hostile to open source values have a significant negative effect on participation, whereas announcements that were perceived as friendly either significantly increase participation or have no significant effect. Our study shows it is important to understand external factors to identify reasons for decline or growth of participation in online communities.

Contents

Introduction
Newcomer participation in the Java developer community
Newcomer participation in the MySQL developer community
Discussion and conclusion

 


 

Introduction

In the past decade, there has been an exponential rise in both the quantity and importance of online communities represented by the enormous usage and user base of communities such as Facebook. Given the central role of online communities in our lives, it is no surprise they have been studied extensively by information science researchers. Researchers have paid particular attention to factors that motivate people to participate in OSS online communities and once participation reaches a critical mass, how is it sustained and what kind of social and governance structures emerge (Oreg and Nov, 2008; Schoberth, et al., 2003; Shah, 2006; Teo, et al., 2013). One core area of work has been newcomer socialization in these communities. Beenen, et al. (2004) and Ren, et al. (2008) showed how social psychology theories — such as common bond theory and common identity theory — can be useful for understanding, designing and managing for online communities newcomers participation (Ren, et al., 2004; Ren, et al., 2008). Socialization of newcomers was found to essential for the growth and sustainability of the community, especially to avoid the decay in volunteers’ participation, which is an essential recourse for online communities (Schoberth, et al., 2003). Several scholars have advanced design principles that promote newcomer participation in online communities (Choi, et al., 2010; Kim, 2000). For example, it has been shown that behaviors of politeness and rudeness play an important role in attracting appropriate responses in online communities (Burke and Kraut, 2008). Given the increasing prevalence of open source software (OSS) and the central role of online participation in OSS, ways to attract and socialize newcomers into OSS online communities is a critical issue in determining its success. In OSS development communities, sustained participation has been linked a number of factors, including governance structures (Shah, 2006), situated learning and identity construction behaviors of participants (Fang and Neufeld, 2009), and ideology (Stewart and Gosain, 2006).

Although this body of work has significantly improved our understanding of online communities, what seems to be missing is an understanding of how online participation may change as a result of external or exogenous events whose influence is outside the scope of community design or its governance structure. Online communities, similar to off–line communities, are embedded in larger organizational and institutional contexts. For instance, many online communities are hosted and run by commercial firms (e.g., Amazon, Usenet by Google, Yahoo Groups, Google Groups, Facebook), by professional associations, educational institutions, and other formal and information organization. In particular, OSS communities often benefit immensely from the stewardship of commercial firms that either developed the initial software or are its bigger user and supporter. Firms like IBM are known for their support of OSS, which includes hosting communities and allocating developer time. In this paper, we take a first step towards examining the changes in participation in two online communities. We identify the perceived image of the steward company as a key factor determining participation through its influence on motivation. The image of a host is especially important for communities associated with software products and community supported software products such as open source software, since these are likely to change hands through mergers or acquisitions. In the rest of the paper we first review the background literature, present both research studies, and discuss the findings and their implications for the study of online communities.

 

++++++++++

Newcomer participation in the Java developer community

In the first study, we examined changes in participation in a Java developer community triggered by two significant events: (1) open sourcing of Java by Sun (19 November 2006) and (2) acquisition of Sun (and consequently of Java) by Oracle (20 April 2009). Both events represent a change either in the nature of product or in the ownership of the product and consequently the stewardship of the associated online community.

Methodology, data collection, and analysis

The target online community to study both events was the newcomer forum for Java developers. Data were collected by parsing the public information on the forum over a four–year interval from 1 January 2006 through 31 January 2010 (to include a pre– and post–period for both the events — Open source release (November 2006) and the Sun–Oracle deal (April 2009). We compared forum participation in the eight months before and after the two events: first we examined participation data around the Sun announcement of the open source release of the Java code. We compared the number of new users who registered for the forum, and then compared two aspects of user interaction — the number of discussion threads in the forum and the number of posts in the discussions. Next, we carried out a similar analysis of the participation data in the months before and after the announcement of Sun’s acquisition by Oracle.

Duke Star Rating System in ‘New to Java’ Forum

The ‘New to Java’ forum classifies users by four different user profiles. The profiling is based purely on a feedback system that rates users on the quantity and quality of help provided on the forum. The feedback system enables the primary question asker or the help seeker the option to rate the users who provide answers. Rating is done by assigning ‘duke points’, an internal rating system, and points are stored on a cumulative basis against the help providers’ name. ‘Duke’ is the Java mascot. The ‘duke status’ of an individual user is determined on the basis of hard cutoffs set on the accumulated duke points. The status of a newcomer to the community is set to bronze by default and remains that way until the user garners up 100 duke points. Members with a ‘silver star’ rating have between 100 to 500 duke points. While gold members have amassed duke points in the 500 to 1,000 range, platinum members lie at the top of the rating scheme with more than a 1,000 duke points. The rating system is not just inclusive of the activity of members in the ‘New to Java’ forum but is a cumulative assessment of the individual user across all developer forums hosted by the company. Table 1 shows the comparative summary of different groups.

 

Comparative summary of different groups
 
Comparative summary of different groups
 
Table 1: Comparative summary of different groups.

 

Bronze members

Bronze members in the community are essentially newcomers. Bronze is the default duke status assigned to an individual who joins the community. This status continues until the individual garners 100 duke points. Since duke points are only generated by providing meaningful answers to questions asked by fellow community users, an average newcomer spends quite some time before transitioning onto the next level. On an average the number of bronze members active on the forum in a month makes up for about 90.6 percent (89.59%–91.60% confidence interval) (P < 0.0001) of the forum in terms of their numbers (averaged over a period of three years starting February 2007–January 2010). A majority of these users exhibit a ‘help seeking’ behavior (Miller, et al., 1991) and ask only questions and clarifications to their own questions. A lesser number of such individuals post replies to others questions (on a monthly basis where about 71 percent of bronze members ask just questions and clarifications, only about 23 percent post only replies to others questions without asking a question themselves while the rest ask questions and also reply to others) (P < .0001). The credibility or usefulness of these replies is often not great as reflected by the average duke star rating of the individuals who engage in such replying behavior (average duke star rating of two among bronze users who exhibited replying behavior). Although these users posts outside the ‘New to Java’ forum the difference in total posts (sum of all posts by the user including other forums as well) and posts to ‘New to Java’ is not substantial when compared to the other groups (about 50% percent of these users were only active on ‘New to Java’). In line with the onion–like social structure model proposed by (Crowston and Howison, 2005) this group of members represent the outer layer of the forum structure.

Silver members

The silver members represent the next rung of users in the forum structure. Accounting for only about four percent (3.63%–4.34% confidence interval) (P < .0001) of the number of such active members in the forum on a monthly basis, they represent the ‘in transition’ members in the forum. While a majority of them (95% ) post replies to other questions while not asking a question or clarification themselves, only about 4.5 percent of these users engage in asking questions and also replying to others (on a monthly basis). Only 0.5 percent engage in purely help seeking behavior (i.e., only ask questions while not replying to others). However none of these individuals engage in the purely help seeking behavior over time (over the three–year time period, there existed no silver member who had only asked questions and not replied). The average duke star rating for this group was found to be 200 indicating some usability of their replies. Silver members also did not restrict themselves to the ‘New to Java’ forum and posted freely on other forums as well (as indicated by a significant difference in their total post count and post to ‘New to Java’ only) and the presence of no silver member who posted exclusively at ‘New to Java’.

Gold members

According to the design principles of the community structure individuals are awarded the gold duke status on accumulating 500 duke points. These users account for about 1.83 percent of the number of active members on the forum on a monthly basis (1.62%–2.05% confidence intervals) (P < .0001). They primarily constitute the outer rung of the core members and are heavily engaged only in posting replies to the questions asked by other members in the forum (94 percent) (92.22%–95.93% confidence intervals). Gold members rarely ask questions in the ‘New to Java’ forum. With an average duke star rating of 667 they seem to have gathered substantial credibility. Furthermore, gold members are involved extensively outside the ‘New to Java’ forum as well as apparent from the difference in total posts and posts to ‘New to Java’.

Platinum members

Platinum members represent the inner core of the community structure, and participate in the highest numbers. Representing about 3.6 percent (3.13%–4.06% confidence interval) (P < 0.0001) active members in the forum per month, they represent the third largest active group interestingly more than the number of gold members. They exhibit the same behavior as that of gold members, however in a more pronounced fashion. They outnumber the gold members in terms of the number of replies provided and also in the average duke star rating (1,905 duke points). Similar to gold members, they extensively contribute posts outside ‘New to Java’ forum.

Results for event 1: Open sourcing of Java

The results of the quantitative analysis of forum participation before and after Sun’s announcement of open sourcing of Java reveal a significant difference in terms of both participation dimensions (see Figure 1): the average number of new forum users per day increased from 15.8 to 20.1 (p < 0.01, Mann–Whitney test), the average number of discussion threads per day increased from 29.4 to 42.8 (p < 0.01, Mann–Whitney test), and the average number of messages posted per day increased from 181.5 to 304.7 (p < 0.01, Mann–Whitney test). The number of discussion threads and the number of messages posted were highly correlated (r=0.96 before the announcement; r=0.92 after the announcement). Rumors of open sourcing had started appearing in the media as early as August 2006 (Figure 2), which is reflected in the data as well. Even though our media related data and quotes come from public sources for reasons of privacy we have anonymized the user names for all comment posters throughout the paper.

 

Number of messages posted (per day) on the forum in the months before and after Java was open sourced, November 2006
 
Figure 1: Number of messages posted (per day) on the forum in the months before and after Java was open sourced, November 2006.

 

 

Postings about Java's open sourcing
 
Figure 2: Postings about Java’s open sourcing.

 

Activity broken down by user status indicated a sharp increase in the number of Bronze users (Average of 776.75 before the announcement to an average of 1,020.65 after the announcement (t=-3.10, p<.01). The other groups of users also produced a steady and significant increase in activity (see Figure 3). Table 2 provides the statistics broken down by groups.

 

Activity of different groups on a monthly basis
 
Figure 3: Activity of different groups on a monthly basis (activity of Bronze members has been plotted on the secondary axis, while others on the primary axis).

 

 

Activity statistics for different groups before and after the event
 
Table 2: Activity statistics for different groups before and after the event.

 

Results for Event 2: Acquisition of Sun by Oracle

The average number of new forum users per day dropped from 15.4 to 6.3 (p<0.01, Mann–Whitney test), the average number of discussion threads per day dropped from 33.1 to 15.1 (p<0.01, Mann–Whitney test), and the average number of messages posted per day dropped from 260.9 to 128.0 (p<0.01, Mann–Whitney test). It should be noted that the number of discussion threads and the number of messages posted are highly correlated (r=0.98 before the acquisition announcement, and r=0.93 after the announcement), as they are both facets of user interaction (Figure 4). Several concerns about this acquisition were raised by developers and users, as reflected in online discussions on various forums (Figure 5).

 

Number of new users registered per day in the months before and after the acquisition of Sun by Oracle April 2009
 
Figure 4: Number of new users registered per day in the months before and after the acquisition of Sun by Oracle — April, 2009.

 

 

Postings about the Oracle acquisition
 
Figure 5: Postings about the Oracle acquisition.

 

However, the number of discussions per thread, or the number of messages posted to each questions posted on the forum (Table 3), did not see any drop and actually increased slightly although the increase was not significant (average of 8.27 messages per thread before the acquisition announcement and an average of 8.41 messages per thread after the announcement (t=-0.72, p<.3). This indicates that while there was little change in the help providing feature of the forum, there was a sharp decrease in both new user registration and the number of questions being asked. A sharp decline in the forum’s activity was observed in conjunction with the rumors of an impending buyout of Sun. The average number of distinct members active on the forum saw a sharp drop in the months leading to the announcement. The average number of users active on the forum on a monthly basis before the announcement was 907 as compared to a paltry 477.287 users after the announcement (t= 4.84, p<.01). However this decline was dominated primarily by the drop in participation among new users, while the more experienced members continued to participate in almost a similar fashion both before and after the event (Figure 6).

 

Activity statistics for different groups before and after the event (Oracle takeover)
 
Table 3: Activity statistics for different groups before and after the event (Oracle takeover).

 

 

Activity of different groups on a monthly basis (activity of Bronze members has been plotted on the secondary axis, while others on the primary axis)
 
Figure 6: Activity of different groups on a monthly basis (activity of Bronze members has been plotted on the secondary axis, while others on the primary axis).

 

 

++++++++++

Newcomer participation in the MySQL developer community

MySQL is a highly popular open source database software. It is a relational database management system that runs as a server providing multi–user access to a number of databases. The MySQL development project source code is available under the terms of the GNU General Public License, as well as under a variety of proprietary agreements. MySQL is a key part of the LAMP (Linux, Apache, MySQL, PHP/Perl/Python), the widely used open source enterprise software stack and therefore used as a solution stack for application servers. MySQL is used by a wide variety of organizations, including Yahoo!, Alcatel–Lucent, Google, Nokia, YouTube and Wikipedia. It was originally developed by David Axmark, Allan Larsson and Michael Widenius, and was then owned by a MySQL AB, a Swedish company. MySQL was first released internally in May 1995 and the Windows version was later released in January 1998 for Windows 95 and NT. MySQL 5.5 is currently available in pre–release (as of June 2010).

In January 2008 Sun Microsystems announced the acquisition of MySQL AB. Sun kept MySQL as an open source product. Oracle, which had earlier acquired Innobase (which develops the InnoDB storage engine that allows MySQL to provide various functionalities) and Sleepycat Software (the makers of Berkeley DB, a basis for another MySQL storage engine) acquired Sun Microsystems as well in January 2010. However, Oracle has committed to the dual licensing with commercial and GPL version for MySQL server until 2015. We studied a newcomer forum for MySQl developers and users. The choice in this particular forum was made for several reasons. First, newcomers indicate participation and product uptake increase or decrease. Second, MySQL is a very popular open source database management system, and therefore the findings may be more reflective of general trends in open source software and their related communities. Third, several incidents in the recent past of MySQL — its acquisition by Sun Microsystems (a proponent and major contributor of open source software and systems) and then the subsequent acquisition of the steward company Sun by Oracle (MySQL was among other few Open Source Relational Database Management Systems that were a direct competition to the Oracle Flagship), both offer useful setting for a natural experiment in which changes are being the result of external events rather than researchers’ intervention.

Methodology, data collection, and analysis

We use the natural experiment method, which is commonly used in a variety of disciplines such as sociology, economics, political science and medicine (Meyer, 1995). In a natural experiment the treatment is not administered by the experimenter, but by exogenous events outside the control of the researcher. As such, natural experiments provide an opportunity to examine data as a real–world experiment. The method is therefore particularly useful for examining the effects of public announcements on user participation. In the present study, we examine changes in participation in MySQL newcomer forum before and after two takeover announcements — the January 2008 announcement of MySQL’s takeover by Sun Microsystems, and the April 2009 announcement of Sun’s takeover by Oracle, even though the product (MySQL) does not.

The data collection procedure involved parsing information from the online Web pages and storing them as a relational database. We created three main datasets based on the activity in the forum during a four–year time window starting 1 January 2007 through 31 May 2010 (to include a pre– and post–period for both the events — Acquisition by Sun (February 2008) and the Sun–Oracle deal (April 2009). The categories were: 1) User information and 2) Discussion thread information. The User information dataset captures details about all the registered users posting within the forum during the given time frame. The Thread information table details each thread that appears within the forum. At the thread level we have gathered information like the thread ID, thread subject, post date, post time, and the post content. This exhaustive table contains information on all individual posts/messages in the forum. All information is stored in a relational database to enable easy data access and manipulation.

We compared the number of new users who registered to the forum, as well as the number of discussion threads in the forum, in the eight months before and after the acquisition announcement of MySQL by Sun Microsystems. Next, we compared the number of users and threads in the twelve months before and after the acquisition announcement of Sun by Oracle (the difference in the length of the periods analyzed is due to the availability of participation data).

Results for Event 1: Acquisition of MySQL by Sun

The results of the quantitative analysis of forum participation before and after the announcement reveal no significant difference in participation dimensions (see Figure 7).

 

New user registration before and after the Sun-MySQL acquisition announcement
 
Figure 7: New user registration before and after the Sun–MySQL acquisition announcement.

 

The average number of new forum users per day as well as the number of discussion threads decreased slightly, but not in a statistically significant way. Mixed views on the effect of the pending acquisition were also reflected in users’ postings in various discussion venues (see Figure 8).

 

Slashdot postings about the Sun acquisition
 
Figure 8: Slashdot postings about the Sun acquisition.

 

To further investigate the perception of the user community on Sun’s takeover of MySQL, we performed an analysis of major news outlets (both online and off–line) using online search tools and databases such as ProQuest. In particular, we looked at user sentiments based on their comments to online news items. Although the views were mixed, they were largely positive. Figure 9 showcases a representative sample of comments.

 

User comments on news about the Sun acquisition
 
Figure 9: User comments on news about the Sun acquisition.

 

Results for Event 2: Acquisition of Sun by Oracle

A different pattern [1] was observed following the Sun’s acquisition by Oracle announcement in April 2009 (see Figure 10): the average number of new forum users per month dropped sharply, from 423.2 to 301.9 (p<0.01, Mann–Whitney test), and the average number of discussion threads per day dropped from 669.4 to 598.1 (p<0.01, Mann–Whitney test). The drop in participation may be attributed to the perceived identity of the acquiring company, in this case Oracle (see Figure 11).

 

New user registration before and after the Oracle-Sun acquisition announcement
 
Figure 10: New user registration before and after the Oracle–Sun acquisition announcement.

 

 

Slashdot postings about the Oracle acquisition
 
Figure 11: Slashdot postings about the Oracle acquisition.

 

An analysis of the messages’ content showed that the content was similar to that before the events and there were no messages on the forum that referred to the acquisition. The questions and the responses were all related to technical issues. Once again, to further investigate the perception of the user community on Oracle’s takeover of Sun, and consequently of MySQL, we performed an analysis of major news outlets (both online and off–line) using online search tools and databases such as ProQuest. We looked at user sentiments based on their comments to online news items and the overall perception of the acquisition was negative. Figure 12 showcases a representative sample of comments.

 

User comments on news about Oracle acquisition
 
Figure 12: User comments on news about Oracle acquisition.

 

 

++++++++++

Discussion and conclusion

As the selected quotes suggest, the perceived identity of the acquiring company may be based on expectations for tangible changes (i.e., lack of investment, and eventually product obsolescence), or simply a more general perception about the company’s identity (Sun as the “cool company” vs. Oracle as a “gouging, monster”). Thus, users may decrease their participation because the values associated with the new steward company may be perceived to be incongruent with the values of the old steward company, and with which users identified. Further research may help to determine the extent to which this factor is at work. The lack of significant change after the announcement of the MySQL AB acquisition by Sun may be attributed to the relatively similar perceived identity of both companies, as supporters of free and open source software. The findings are important for companies interested in involving users in activities related to their products, and in particular, companies which are involved in the development and stewardship of open source software. The findings suggest that it is essential for such companies to actively address concerns that might arise due to changes such as acquisitions, even if such changes involve no change in the product itself.

The findings from our studies demonstrate conclusively that not all changes and interactions in online communities can be attributed to internal factors, as suggested by Wellman and Gulia (1999). External environment changes can affect a community or organization in a variety of ways. One explanation for the results observed here is the concept of environmental jolts (Meyer, 1982). Jolts are transient shocks that cause temporary disruptions that perturb their organizational inhabitants and then they subside (Meyer, et al., 2005). An important characteristic of jolts is that even though they might be expected, their exact occurrence is difficult to foresee (Meyer, 1982; Meyer, et al., 2005). They are also open to diverse interpretations within an organization and can be seen as opportunities or threats. Overall, environmental jolts rarely threaten survival of soundly designed organizations, but they can trigger responses that reveal how organizations adapt to their environments (Meyer, et al., 2005). Our research adds to the growing body of knowledge on online communities and open source software development by providing evidence of the effect of external events on participation (also see Johri, et al., 2011a, 2011b). Whereas the majority of prior work on online communities has largely focused on users’ motivations for participation, community governance structure and the design of communities to foster participation, the present study sheds light on the role of external events. Future work may be helpful in addressing specific issues such as preparedness to respond to external jolts: if the external jolt is predictable — as many are — can community leadership prepare for them? Another issue the findings raise is the possible importance of a change of perspective, that is, in cases where community activity is a critical aspect of a firm’s product, response to external events should be taken into account by company leadership before decisions are made. Strategies can be adopted that facilitate the monitoring of ongoing activities in an online community. Future research is needed to develop new ways of understanding the effect of external events on community participation.

As with any research study, this study has certain limitations. We attribute the primary exogenous effect to Oracle’s and Sun’s relative perception by the community. While this seems plausible, we recognize that it is probably not the only source of difference between the companies. Therefore, the findings should be interpreted cautiously. Since the forums we studied banned off–topic messages, these issues — even if pertinent — could not be discussed by participants. In addition to Slashdot, we performed an analysis of news sources and associated comments, but a deeper analysis of community sentiment — using methods such as surveys — could supplement our findings. End of article

 

About the authors

Raktim Mitra is currently a Project Manager at Discover Financial Services in Chicago. He received his master’s of science degree in industrial and systems engineering from Virginia Tech in Blacksburg, Virginia in 2011. His research examined newcomer socialization and effects of external events on participation in online communities.
E–mail: mitra [dot] raktim [at] gmail [dot] com

Aditya Johri is an Assistant Professor of Engineering Education, Computer Science (courtesy), Industrial and Systems Engineering (courtesy) and Science and Technology in Society (courtesy) at Virginia Tech in Blacksburg, Virginia. He studies the use of information and communication technologies (ICT) for learning and knowledge sharing, with a focus on cognition in informal environments. Dr. Johri received the National Science Foundation Early CAREER Award in 2009.
E–mail: ajohri [at] gmail [dot] com

Oded Nov is an Assistant Professor at the Polytechnic Institute of New York University in Brooklyn, New York. He received his Ph.D. from Cambridge University. His research focuses on motivational, dispositional and social aspects of information systems and social media. Dr. Nov is a recipient of the National Science Foundation Early CAREER Award.
E–mail: onov [at] poly [dot] edu

 

Acknowledgements

This work was partially supported by the U.S. National Science Foundation under award EEC#0935143. Views expressed here are those of the authors and not necessarily of the funding agency. Findings from this research were earlier presented at the 2011 Computer Supported Cooperative Work Conference and the 2011 iConference and portions of results presented here appear as short papers in the conference proceedings (Johri, et al., 2011a,b).

 

Note

1. The differences between the lengths of the two events’ observed periods were due to the availability of data.

 

References

G. Beenen, K. Ling, X. Wang, K. Chang, D. Frankowski, P. Resnick, and R.E. Kraut, 2004. “Using social psychology to motivate contributions to online communities,” CSCW ’04: Proceedings of the 2004 ACM Conference on Computer Supported Cooperative Work, pp. 212–221.

M. Burke and R.E. Kraut, 2008. “Mind your Ps and Qs: The impact of politeness and rudeness in online communities,” CSCW ’08: Proceedings of the 2008 ACM Conference on Computer Supported Cooperative Work, pp. 281–284.

B. Choi, K. Alexander, R.E. Kraut, and J.M. Levine, 2010. “Socialization tactics in Wikipedia and their effects,” CSCW ’10: Proceedings of the 2010 ACM Conference on Computer Supported Cooperative Work, pp. 107–116.

K. Crowston and J. Howison, 2005. “The social structure of free and open source software development,” First Monday, volume 10, number 2, at http://firstmonday.org/article/view/1207/1127, accessed 31 May 2013.

Y. Fang and D. Neufeld, 2009. “Understanding sustained participation in open source software projects,” Journal of Management Information Systems, volume 25, number 4, pp. 9–50.http://dx.doi.org/10.2753/MIS0742-1222250401

A. Johri, O. Nov, and R. Mitra, 2011a. “‘Cool’ or ‘Monster’? Company takeovers and their effect on open source community participation,” iConference ’11: Proceedings of the 2011 iConference, pp. 327–331.

A. Johri, O. Nov, and R. Mitra, 2011b. “Environmental jolts: Impact of exogenous factors on online community participation,” CSCW ’11: Proceedings of the 2011 ACM Conference on Computer Supported Cooperative Work, pp. 649–652.

A.J. Kim, 2000. Community building on the Web. Berkeley, Calif.: Peachpit Press.

A.D. Meyer, 1982. “Adapting to environmental jolts,” Administrative Science Quarterly, volume 27, number 4, pp. 515–537.http://dx.doi.org/10.2307/2392528

A.D. Meyer, V. Gaba and K.A. Colwell, 2005. “Organizing far from equilibrium: Nonlinear change in organizational fields,” Organization Science, volume 16, number 5, pp. 456–473.http://dx.doi.org/10.1287/orsc.1050.0135

B.D. Meyer, 1995. “Natural and quasi–experiments in economics,” Journal of Business and Economic Statistics, volume 13, number 2, pp. 151–161.

S. Oreg and O. Nov, 2008. “Exploring motivations for contributing to open source initiatives: The roles of contribution context and personal values,” Computers in Human Behavior, volume 24, number 5, pp. 2,055–2,073.

Y. Ren, R.E. Kraut, and S. Kiesler, 2008. “Applying common identity and bond theory to design of online communities,” Organization Studies, volume 28, number 3, pp. 377–408.

T. Schoberth, J. Preece, and A. Heinzl, 2003. “Online communities: A longitudinal analysis of communication activities,” HICSS ’03: Proceedings of the 36th Hawaii International Conference on System Sciences.

S.K. Shah, 2006. “Motivation, governance, and the viability of hybrid forms in open source software development,” Management Science, volume 52, number 7, pp. 1,000–1,014.

K.J. Stewart and S. Gosain, 2006. “The impact of ideology on effectiveness in open source software development teams,” MIS Quarterly, volume 30, number 2, pp. 291–314.

H.J. Teo, A. Johri, and R. Mitra, 2013. “Visualizing and analyzing productive structures and patterns in online communities using multilevel social network analysis,” paper presentated at CSCL 2013: Tenth International Conference on Computer Supported Cooperative Learning (17 June, University of Wisconsin–Madison).

B. Wellman and M. Gulia, 1999. “Net–surfers don’t ride alone: Virtual communities as communities,” In: M.A. Smith and P. Kollock (editors). Communities in cyberspace. New York: Routledge, pp. 167–194.

 


Editorial history

Received 24 March 2013; revised 26 May 2013; accepted 28 May 2013.


Copyright © 2013, First Monday.
Copyright © 2013, Raktim Mitra, Aditya Johri, and Oded Nov.

Effect of external events on newcomer participation in open source online communities
by Raktim Mitra, Aditya Johri, and Oded Nov.
First Monday, Volume 18, Number 6 - 3 June 2013
http://firstmonday.org/ojs/index.php/fm/article/view/4642/3689
doi:10.5210/fm.v18i6.4642





A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2017. ISSN 1396-0466.