Communities, Technologies and Participation: Notes from C&T 2009

Joe McCarthy

Strands Labs Seattle

CCT2009 Participation was the overriding theme at the 4th International Conference on Communities and Technologies (C&T 2009) held at Penn State University in State College, Pennsylvania at the end of June. We can design and deploy technology to support a community, but how do we truly engage that community and motivate its members to participate? The approximately 125 attendees presented and discussed a number of ways to encourage greater participation - especially civic participation - over the course of the 3-day conference (and the preceding day-long workshops). One way some participants at the conference have broadened potential engagement via technology within the C&T community is through the use of social media: a number of presenters have uploaded their C&T 2009 slides to SlideShare (tagged with cct2009), a number of photographers have uploaded their C&T 2009 photos to Flickr (tagged with cct2009) ... and a number of tweets from / to / about the conference can be found via the #cct2009 hashtag on Twitter. I hope I am further broading participation by sharing these notes.

David McDonald from NSFDavid McDonald opened up the conference with an engaging keynote in which he introduced the new U.S. National Science Foundation (NSF) program on Social-Computational Systems (SoCS), which promotes an agenda of embedding more social intelligence into computational systems. The idea is to combine collective human intelligence with socially intelligent computing, making computers first-class participants in a new breed of emergent intelligence. Although he stipulated that the intention is to "let people do what people do best, and let computers do what computers do best", I got the distinct impression that one goal of the program is to render unto computers what is social ... which may lead to the [unintended] consequence of rendering unto humans what is computational. As a former AI researcher, who believes that a little AI goes a long way - and a lot of AI often goes a short way - I am, no doubt, rather biased (and perhaps jaded). In any case, the program strikes me as a resurgence of strong AI in HCI clothing (and others voiced concerns about the prospective domination of computers in this proposed partnership during the Q&A after his talk).

On a lighter note, Dave began his talk with an amusing delineation of the "four stages of 'invited' talk" - idealism & excitement, realization, despair, and finally, resolution - and proceeded to highlight the stages of NSF sponsorship of communities and technologies over the years. He noted that many communities arise as a side effect - someone tries something, people like it, and form a community around it (reminding me of the notion of object-centered sociality) - and posed a thought experiment to encourage us to think about charting new applications of community (and technology) ... I found his idea of FlightBagWatch.com - where voyeurs might login and help conduct bag inspections at the Transportation Security Administration (TSA) X-ray machines in U.S. airports (reminding me of the Texas Border Watch program) - particularly provocative.

In the first paper session I attended at the conference, Marla Boughton presented "Supportive Communication, Sense of Virtual Community and Health Outcomes in Online Infertility Groups", in which she looked at the differing impacts on the sense of virtual community (SOVC) offered by emotional vs. informational support in an online infertility support group. Observing emotional support by members of the community turned out to be the most significant positive factor - surprisingly, even more than providing or receiving emotional support in SOVC ("'tis better to observe than to give or receive" (?!)) - while observing informational support turned out to be the most significant negative correlation in SOVC (I forget whether she reported on giving / receiving informational support).

Dana Rotman shared some observations and insights into the world of video blogging ("vlogging") in "The Community is Where the Rapport Is - On Sense and Structure in the YouTube Community" [slides]. Grounded theory analysis of users' feelings and interaction patterns revealed repeated themes, among which was the distinction between YouTubers (people who post vlogs) and Tourists (who don't). She found that comments constituted the "most important interaction mechanism", a finding very much in alignment with my own thinking - and blogging - about comments as validation (or as my friend Noah Kagan so pithily put it, "comments make me orgasm"). This finding was particularly interesting given that she also found considerable ambivalence in the vlogger community about whether people who [only] post comments qualify as full-fledged YouTubers or mere Tourists (and I find the ambivalence all the more interesting given my own musings about whether bloggers are more likely to post comments on other blogs than non-bloggers). Other interesting findings include what she called triangulation - vloggers expose their thoughts, feelings and face and [so] comments are all based around an individual (vs. community bulletin board, forum) - and the preference of Gmail over YouTube's own direct 1:1 communication mechanisms. In addition to stimulating content of Dana's talk, the form of her slides - using a tag cloud for the slide titles - was very creative, so I highly recommend viewing them on SlideShare.

WikipediaLogo Pnina Shachaf reported an experiment on "Answer Quality on the Wikipedia Reference Desk" [paper | slides], in which she compared the the quality of answers produced [collectively] on the Wikipedia Reference Desk (WRD) to those produced by individual professional reference librarians. The experiment was motivated by an interesting series of responses to a question she posted on the WRD asking about the prefix following pico- (one trillionth) in the international system of units, (it is femto-, one quadrillionth) and the discussion about linguistic origins of that term which followed. The collective wisdom of WRD on this topic was vacillating back and forth between Spanish and Italian (and the professional librarian she consulted also posited one of these two sources - I forget which), but eventually, over the course of discussions in WRD, the Danish origin of the word was established. Other motivations include Andrew Keen's scree against Wikipedia and other forms of social media, The Cult of the Amateur, and the ongoing debate(s) about the relative accuracy of Wikipedia vs. more traditional encyclopedias. Pnina decided to conduct an experiment to test the relative quality of WRD vs. professional reference librarians more rigorously. Using Nvivo 7 for the content analysis of 434 messages involving 77 transactions (where a transaction = a question + one or more answers) among 170 unique users (of whom 122 were "seasoned" Wikipedia users), she found that Wikipedia is generally doing as well or better than professional librarians. In particular, WRD was, on average, more responsive (4 hours vs. 18-21 hours), more complete (63% vs. 47%), and verifiable (88% vs. 53%) than professional librarians. The accuracy was the same (55%), and the explicit satisfaction - as measured by messages of unsolicited thanks - was similar (20% vs. 16-20%) [and I'll pass on the opportunity to extend this into an unsolicited rant about how many people are, or appear to be, so ungrateful (ubiquitous ingratitude?) ... a topic I've previously ranted about in the context of community, football and food).] Among the possible explanations she gave for the relatively high quality of WRD answers were: technology (interactive nature of WRD), the likelihood that many WRD volunteers have relevant professional background, type types of questions asked on WRD, and the collaborative nature of social reference, i.e., what she called amalgamated answers - iterative elaborations by both the requester and the responders (WRD questions generated an average of 4.6 responses).

Lev Gonick gave the second keynote of the conference, "From Digital Campus to Connected Community", providing an overview of the OneCommunity project that he is helping to lead in Cleveland, Ohio. As with with the earlier keynote, Lev led off with some light-hearted fare - a satirical "Hastily Made Cleveland Tourism Video" ... which either inspired or was inspired by a contest for a positive Hastily Made Tourism Video ... the former has been viewed over 2M times ... approximately 2000 times more than the winning entry of the positive videos.

Apollo11-NASA-as11_36_5355 Internet_map_800 Lev proceeded to show a photo of the earth taken during the Apollo 11 mission in July 1969 ([similar to that] shown on the left), noting that this photo produced a new cognitive map of what place meant [I found myself musing on the term "placemeant"], and what community meant - as did another momentous - though more terrestrial (but perhaps transcendental for some) - event in 1969 (Woodstock) - and how a more recent map of the internet ([similar to that] shown on the right) similarly [re]frames our conception of community ... at least within the context of a conference on communities and technologies. Continuing the progression from the larger to the smaller, and from more abstract to more concrete, Lev devoted much of the rest of his talk to describing OneCommunity:

OneCommunity is a connected-community venture, informed by a mission to be a big bold 21st-century community-oriented project that delivers advanced information technology capabilities to achieve community priorities for economic development, learning, job training, research support, preeminence and distinction.

Logo_OneCommunity As the CIO of OneCommunity, he emphasized the importance of technology infrastructure (the "glue") but also highlighted that the challenge for us (technologists) is to find a way to get out of the way (of the community that will be using the technology) ... rather than indulging our fondness for complexity, a fondness which is not generally shared by "Joe Sixpack". OneCommunity is partnering with non-industry organizations (universities, libraries, hospitals) so that they can more effectively give away control in community networking projects, which he sees as an essential element in their prospects for success. Impactful community applications developed thus far in the 5 year old project have focused on health and education, including a Green Computing Initiative to collect, refurbish and re-install computers in Cleveland schools, and OneClassroom, a collaboration between Cleveland schools and the Cleveland Clinic to provide students with remote birds-eye views of - and interaction opportunities with those engaged in - surgical procedures conducted at the clinic.

During the Q&A session that followed, some of the questions revolved around the issue of the sharp economic decline that Cleveland has suffered in recent years (which was reflected in the video shown at the outset). Acknowledging this, Lev noted that "we start from pain" and "progress can be measured directly proportional to the real or perceived state of crisis".

Doug Schuler gave a talk in the following session on "Communities, Technology, and Civic Intelligence" in which he defined civic intelligence as "how smart a society is as-a-whole in relation to its problems". Given the increasing economic, environmental and social problems we are witnessing, posed the timely and provocative question: "will we be smart enough, soon enough?" In contrast to Tim O'Reilly's framing of web 2.0 as a business revolution, harnessing collective intelligence, Doug suggested that civic intelligence represents a social revolution in communication, enhancing civic intelligence ... and [so] we have to choose whether we want to promote a paradigm of people in service to the computer industry vs. the computer industry in service to people. Doug finished off with a couple of plugs - for the Public Sphere project and a new book based on one of the project's latest areas of focus, Liberating Voices: A Pattern Language for Communication Revolution - and an exhortation: don't rely on the emergence of civic intelligence as a side effect.

CCT2009-10Karim Lakhani opened up the second day of the conference with a keynote on "Knowledge Reuse and Novelty in Community Settings", in which he offered some interesting insights into and experiences with collaborative and competitive projects, as well as some that represented a combination of the two approaches. Karim offered Wikipedia ("the free encyclopedia that anyone can edit") as an example of collaborative endeavors, Innocentive (where Seekers post challenges and Solvers compete to produce - and be paid for - the best solutions) as an example of competitive endeavors, and Threadless (where t-shirt designs are submitted and scored by a community of mutually supportive designers) as a middle ground. He claimed that similar motivations appear to be at work for both competitive and cooperative efforts, and has been investigating the relative costs and benefits of each approach in a series of experiments.

One such experiment - that represents a mixture of competitive and cooperative processes - was a one-week MATLAB "wiki-like" programming contest in which the rules specified that participants could view and modify each others' code and see each others' relative standings (though I'm not sure how these standings were computed - e.g., via explicit voting or via more implicit [re]use of code). The contest generated 4402 entries, of which 181 were leaders at some point. The winning entry was based on code borrowed by 30 other participants. One of my favorite quotes from the presentation - not sure if it was from Karim or the winning contestant - was "A successful competitor is a creator, a recycler and a talent scout." In assessing the factors that lead to success in such a contest, Karim and his colleagues found 3 factors that positively correlated with individual success: contributing new knowledge (knowledge = code), combining existing knowledge in novel ways and conformance (which, I believe, relates to the adherence of the code to stylistic guidelines). 2 factors not correlated with individual success were complexity and using existing knowledge (presumably without novel contributions), though these 2 factors - along with the other 3 - were positively correlated with collective success ... I believe the non-novel use of existing knowledge may contribute to collective success due to the implicit voting represented by the use of that knowledge ... but I must admit that I fell behind in my notes, so I'm not sure about some of these details.

The rewards for the MATLAB experiment were fame, glory and a t-shirt. A second experiment was run within the TopCoder platform (which provides financial incentives to winners, as well as a coveted "TopCoder rating" which acts as an unofficial certification used by many software companies in their assessment of candidates), using a problem in computational biology. Participants were partitioned into three "treaments": full competitive, full collaborative and hybrid. The collaborative treatment resulted in the fewest entries, lowest effort and best performance (possibly surpassing the state of the art in computational biology); the competitive treatment resulted in the most entries, highest effort, and worst performance; the hybrid treatment, as one might expect, had results between these two extremes.

During the Q&A, Karim noted noted that most competitive systems operate despite massive failure rates, and [so] learning, signaling and intrinsic motivations are important factors. Among the gems contributed by the audience were Ben Shneiderman's recommendation of Alfie Kohn's book, No Contest: The Case Against Competition (or perhaps he was referring to Kohn's essay with the same title ... which should not be confused with Karim's earlier reference to an article in Nature by Marek Kohn, The Needs of the Many), and Jenny Preece's observation that we should be mindful of cultural differences, e.g., Americans tend to be very individualistic - and thus more competitive - than people from other cultures. [Note: Paul Resnick has also posted some notes on Karim's keynote.]

wConnect Mary Beth Rosson presented "wConnect: A Facebook-Based Developmental Learning Community to Support Women in Information Technology", in which she shared some lessons she and her colleagues learned in developing an online community to support the development of women in computer and information sciences. They started out using Bridgetools to create the first version of wConnect, which, while not very successful, helped them understand that they had to "embed wConnect activities within social activities that members enjoy in their everyday lives, and use those interactions to bootstrap the developmental learning community" (as opposed to a "build it and they will come" approach). Their second version used Facebook Groups which simplified many aspects (especially programming), but introduced a number of complications as well, e.g., limited customization, concerns about privacy, marginalization of wConnect members who were not [active] Facebook members, and the merging of personal and professional identities. One of the challenges they faced was how to lower the barriers to enrollment and authentication, while ensuring that only members of their target community - women who are in or thinking about careers in computer and information sciences - have access.

Ibm-Welcome-Center-02Bridget Blodgett gave a talk entitled "And the Ringleaders were Banned: An Examination of Protest in Virtual Worlds", offering a brief history of virtual world protests, starting with reactions to a 1993 cyberrape in LambdaMoo, a text-only virtual world, to more recent social uprisings in modern 3D virtual worlds including EVE Online and Ultima Online (I was surprised to discover that there is so much criticism of Ultima Online, the topic has its own Wikipedia page!). My favorite example was one that bridged gaps between online and offline: protests against IBM in Second Life (I was slightly less surprised to discover there is an entire blog devoted to the movement ... and even less surprised to find a YouTube video). In the fall of 2007, IBM removed bonuses and reduced wages for unionized Italian employees, who then went on strike in Second Life (SL), recruited up to 2000 virtual protesters from all over the world (including bananas and triangles, as well as more human-like avatars), managed to shut down IBM islands in SL, and disrupt IBM business (a real IBM meeting in SL). From what I can gather, the protest achieved its goals (in real life), and there is even a SL museum dedicated to the protest. The other examples of protests Bridget shared were all against companies that host virtual worlds. IBM is a global company with a large stake in the physical world, although I should note that IBM has been very active in Second Life (and a recent report suggests that it has reaped a significant ROI through its use of Second Life). Given my longstanding misgivings about the pervasiveness and permeability of games worlds (not that SL is, strictly speaking, a game), I was heartened to see participation in a virtual world yield positive impact in the real world.

Speaking of online and offline participation, the next session I attended was a panel on "Making Participation a Priority", moderated by Ben Shneiderman, in which Gerhard Fischer, Jenny Preece and Marc Smith shared their perspectives on participation.

During the audience participation portion of this panel on participation, Jack Carroll - the chair of C&T 2009 - voiced further caution, noting that those who have been conducting research into communities and technologies [in the U.S.] have benefited from being funded like computer scientists, and asking "if we succeed in making social participation an important component of future research, will we be funded like sociologists?" [Sociology research receives far less funding in the U.S. than Computer Science research.] This observation may well be part of the what is driving some of the computational prioritization in the NSF program that Dave McDonald had introduced in his opening keynote. Shelly Farnham attempted to shift the conversation back toward a more positive perspective, noting that while there are some risks to broadening the scope of participation, greater participation is, generally speaking, A Good Thing.

[Update: Ben has since pointed me to a recent article he and Jenny have coauthored - and about which Marc has blogged - that provides much more information on some of the topics discussed in the panel, exploring the progression from reader to contributor to collaborator to leader for [some] users of social media. I'll include a graphic representation of the Reader-to-Leader Framework (Figure 1 from the paper (also included in Marc's blog post)) and a reference to the full paper below:

Reader-to-Leader_Framework

Preece, Jennifer and Shneiderman, Ben (2009) "The Reader-to-Leader Framework: Motivating Technology-Mediated Social Participation," AIS Transactions on Human-Computer Interaction (1) 1, pp. 13-32. Available at: http://aisel.aisnet.org/thci/vol1/iss1/5.]

The next day started off with a panel on "Community technology to support geographically-based communities", moderated by Marcus Foth (who co-organized the Digital Cities 6 workshop at C&T 2009 and will be chairing C&T 2011), and included Paul Resnick, Fiorella de Cindio, Keith Hampton and me.

There was time for a few questions, but unfortunately, I did not take careful notes. The only thing I remember from the Q&A session was Doug Schuler urging us not to forget the importance of human animation, citing as an example a woman in one of his classes [at The Evergreen State College in Olympia, Washington] last semester who always got people talking on her bus (which sounds like a great example of civic participation, but I wondered how the practice was perceived by other bus riders ... and whether there is such a thing as bus etiquette).

SocialLifeCover The next session I attended was a set of papers on Placed Community, [also] chaired by Marcus Foth. Keith Hampton led off this session with "The Social Life of Wireless Urban Spaces: Internet Use, Social Networks, and the Public Realm", a project in which Keith and his colleagues followed in the footsteps of William Whyte's classic work (and book (and video)) on "The Social Life of Small Urban Spaces" to see how the introduction of WiFi into these spaces affects their usage(s). Keith noted a confluence of trends that constrain diversity: homogeneity within the mass media, privatism, and changes to the physical form and use of urban public spaces. He defined three "realms" of social interaction that affect the public sphere: private (the home), parochial (neighborhoods, workplaces, and many third places, where people are generally familiar with other inhabitants of the space) and public (parks, plazas and other public spaces, where people are generally not familiar with many of their co-inhabitants). Whyte's insightful studies were conducted in an era before the Internet, and before the advent of portable devices that could be used to connect to - and through - the internet, such as mobile phones and laptop computers, so Keith and his colleagues set out to see how these recent technological developments have affected the social life of these small urban spaces.

Keith distinguished between co-located acquaintanceships (people sharing and interacting in physical space) and copresent acquaintanceships (people interacting online with others who whom they are not sharing physical space). Among the interesting findings: 10% of the people using wireless Internet in public spaces were observed to have an extended interaction with a stranger, compared with 13% of people people using more traditional media (books, newspapers), 5% of people using mobile phones, and 0% of people using portable music players, making iPods the most effective portable involvement shields. However, 24% of wireless Internet users were "infrequently attentive" to their surroundings, compared to 15% of print media users, 10% of portable music device users, 7% of portable gaming device users and 3% of mobile phone users. In close observations of the wireiess Internet use - and Keith showed a number of fabulous photos showing just how close his observers were able to get to their subjects - it appeared that 29% visited a social networking website, 8% contributed to a blog, and 43% consumed online news or political information.

There is a considerable amount of research into - and controversy about - the impact of Internet use on society and culture, with some claiming that Internet use erodes the sense of community while others claiming that it enhances the sense of community. Keith cited a 2006 study by McPherson, et al., "Social Isolation in America: Changes in Core Discussion Networks over Two Decades", showing that the average American had an average of 2.08 people with whom he / she discusses important matters, and 24.6% of Americans have no such "discussion ties" at all. Although wireless Internet users may interact less with co-located people in public settings, they enjoy an average of 3.82 discussion ties, and only 3.5% report having no discussion ties. One possible explanation Keith offered for this was that wireless Internet use may better balance opposition and like-mindedness to maximize tolerance, deliberation and discursive participation than exposure to provocative and contested public settings. In concluding his presentation, Keith noted that public spaces do not appear to constitute a public realm for wireless Internet users, but may offer new opportunities for engagement - with copresent but not co-located others - in the public sphere ... and while serendipity is nice, it probably is not vital to a public space.

Unfortunately, due to my frantic last-minute preparations for my own presentation at the end of the session, I do not have any notes from the second presentation - "Facilitating Participatory Decision-Making in Local Communities through Map-Based Online Discussion", by Bo Yu.

My presentation was on "Supporting Community in Third Places with Situated Social Software" (and I've posted the slides on SlideShare), in which I reported on some studies we conducted on the adoption, use and impact of our Community Collage (CoCollage) application - a system that shows a dynamic collage of photos and quotes uploaded to a web site by cafe patrons and staff on a large computer display in the cafe, providing a new channel for awareness, interactions and relationships among people co-located there. In his influential book, "The Great Good Place: Cafes, Coffee Shops, Bookstores, Bars, Hair Salons, and Other Hangouts at the Heart of a Community", the sociologist Ray Oldenburg distinguishes third places from first places (home) and second places (work), characterizing them as "'homes away from home' where unrelated people relate", offering access to the "full spectrum of local humanity" in environments that foster inclusive sociability and ease of association. General purpose technology - particularly laptops with WiFi connections - has negatively impacted the sociability in many such places in the U.S., leading some coffeehouses to turn off their WiFi access points on weekends. CoCollage was designed for the specific situation of a third place, leveraging existing offline "technologies" in such places (photos, art, sketches, posters, flyers) to bring some of the richness of online social networks into the physical spaces we share with others. The paper presents some results of a study showing that CoCollage had a significant impact on the dependency component of place attachment (the extent to which people rely on the cafe to have their needs met) and the neighboring component of sense of community (the extent to which people visit each others' homes and do each other favors), and includes a number of examples of open text responses to survey questions asking what people like and don't like about CoCollage.

The final keynote of the conference was by Mark Finkle, a community evangelist at Mozilla, on "Mozilla: Working with the Community", in which he distinguished between the Mozilla Foundation (guidance), the Mozilla Corporation (thrust), and the Mozilla Community (fuel). Mark said that Mozilla wants to push forward an agenda that is not corporate, but advocates for end users. He offered the following interesting statistics about Mozilla's community of communities:

One of the most interesting statistics Mark mentioned was that new beta releases of Firefox attract a community of 800,000 volunteer testers, or 0.25% of the user population (!). Dan McComb recently pointed me to a fabulous series of blog posts on Community by the Numbers, and I've been thinking a lot about how many active members a community needs to be successful (and how to define "active" ... and "successful"), so this provided a useful data point.

Mark shared a number of lessons he has learned from the Mozilla approach to "design by community", emphasizing the needs to listen to the community, lead the community and let the community play and explore. One of Mark's observations about etiquette - "Be respectful until you're respected" - drew a large number of questions from some in the audience during the Q&A session; Mark clarified that he didn't mean to imply that one should no longer be respectful after one has become respected.

One of the things I like about the Communities and Technologies conferences is that they represent a community of communities, but in a more intimate setting than, say, the much larger, multi-track CHI conferences. However, I should note that, due to the dual track nature of the conference, I only saw (and am thus only sharing notes on) about half the paper presentations, and so I want to offer my apologies to all the other authors who were presenting work at the conference - even with only two tracks, every session required a tough decision. I also want to offer my thanks to the conference chair, Jack Carroll, the program committee, student volunteers, presenters and other participants for co-creating a great conference! I enjoyed meeting and reconnecting with a number of remarkable people, and - as I hope my notes show - I learned a lot of interesting, relevant and useful information about communities, technologies and participation!

[This article is an edited version of notes from C&T 2009 originally posted on Joe McCarthy's blog.]