CSCW 2006: Notes on the ACM Conference on Computer Supported Cooperative Work

The ACM Conference on Computer Supported Cooperative Work (CSCW 2006) was held in Banff, Alberta, Canada, last week. The conference, which celebrated its 20th year anniversary this year, included 62 papers and notes, dozens of demonstrations and interactive posters, three panels and a pair of keynotes, and was attended by approximately 400 people. Unfortunately, due to air travel problems -- my 9pm connecting flight from Denver flew all the way to Calgary, circled the fogged-in airport once, then flew all the way back to Denver ... another flight the next morning did manage to land -- I missed the opening keynote and the following session, but I managed to arrive in time for the remainder of the Monday's sessions, and stayed thru the closing keynote on Wednesday.

Unlike UbiComp 2006, for which I posted separate entries each day (and then some), I will split my blog entries on CSCW into two: this one, which will primarily report some of the highlights of the talks I saw, and another in the near future, which will include more commentary.

Shilad Sen (University of Minnesota) presented some work he had done with his colleagues at IBM Research (during an internship) on FEEDme: A Collaborative Alert Filtering System [slides], in which they explored ways of determining when to alert people about changes in different types of resources -- message, file, chat, screen snapshot, folder, task -- that are shared within the IBM Activity Explorer application. An web interface enabled people to rate alerts (thumbs up or thumbs down), and these ratings (for 6385 alerts received by 34 people over 29 days) were used to assess different filtering mechanisms, with the collaborative naive Bayes classifier -- combining personal and group ratings -- achieving the highest accuracy (73%) ... although as the authors note, higher accuracy is probably required for long-term acceptance.

Kimberley Tee (University of Calgary) presented a paper on Providing Artifact Awareness to a Distributed Group through Screen Sharing [video], an effort to translate the kind of informal awareness of activities provided through the artifacts on one's physical desktop into virtual desktops, for the benefit of remote collaborators.  The paper described a component added to the CommunityBar application that regularly updates thumbnail images of users' screens in a frame in the CommunityBar, so that users in a workgroup can see what's on each others' desktops. In order to accommodate varying privacy concerns, the component includes capabilities for users to control the frequency of updates, the zoom level, the distortion level and the window regions shown to others.[Code and more info can be found on the CB Screen Sharing project page].

Jakob Bardram (IT University of Copenhagen) presented "AwareMedia - A Shared Interactive Display Suppoting Social, Temporal and Spatial Awareness in Surgery", the latest in a series of reports on the iHospital project, a long-term effort to design technology to assist collaboration in a life-critical hospital setting. I posted a few notes on Jakob's colleague Thomas Hansen's presentation on some of their experiences at UbiComp 2006. At CSCW, Jakob described an electronic whiteboard application, AwareMedia, that included real-time scheduling information, a video feed from operating theatres, context information about who was present in each operating theatre, and an instant messaging capability, and highlighted its impact on temporal, spatial and context-mediated social awareness.

Bonnie Nardi (University of California, Irvine) presented "Strangers and Friends: Collaborative Play in World of Warcraft", in which she and her co-author participated in and observed a variety of collaborative activities and organizational units within the massively multiplayer online role-playing game (MMO[RPG]), World of Warcraft. These activities included fleeting lightweight encounters such as acts of kindness -- e.g., "buffing" (casting a spell to help another player), "kill assist" (helping another player kill a monster) and escorting lower level players out of harm's way -- and acts of violence ("ganking" and "corpse killing"), moderately structured collaborations such as parties, raids and interactions with "friends", and more highly structured collaborations such as participation in guilds, joining battlefields, and engaging in duels and trades.  There are also random acts of fun, where people can use emote commands for delightful and unpredictable actions (e.g., spontaneous dances), helping to create a less inhibited ambience ig (in-game) vs. irl (in real life).

I have yet to participate in or observe any MMORPGs myself, and yet I am increasingly fascinated by this increasingly popular sphere of human activities, and the benefits enjoyed by those who do choose to participate ... and [so] may post a future blog entry delving into more details of the Nardi and Harris paper, and related work. However, as I mentioned in my notes from the most recent Microsoft Social Computing Symposium (which focused on games worlds), I am concerned about whether the benefits realized in these immersive virtual worlds translate into benefits that are realized in the physical world ... or whether the participants in MMOs get -- and give -- their fill of goodwill online, and thus feel less inclined to engage in actions that are of benefit to friends and strangers offline.

Scott Counts (Microsoft Research) presented "Sandboxes: Supporting Social Play through Collaborative Multimedia Composition on Mobile Phones", an application designed to promote interactivity, flexibility and cohesiveness in the creation and sharing of multimedia content via mobile phones. Creators (sharers) can add or manipulating text, symbols (e.g., a smiley), sounds and photos; viewers (sharees?) can chose to playback multimedia compositions in two modes, slideshow or historical. In trial deployments, users co-opted the platform for unanticipated types of social gaming, such as tugs of war in manipulating multimedia items on the canvas and one-upping each other in posting the most humorous photos.

Cliff Lampe (Michigan State University) presented "A Face(book) in the Crowd: Social Searching vs. Social Browsing", in which he and his colleagues conducted studies of Facebook users. They found that most users of Facebook are far more likely to use the online social networking service for social searching -- finding out more about someone online that they have already encountered offline -- than for social browsing -- finding someone online that they want to encounter offline. Among the interesting statistics they compiled were: the percentage of freshmen surveyed who had Facebook profiles increased from 85% to 96% from September 2005 to January 2006; most users report using Facebook at most 30 minutes per day (after initial profiles are created); and only 5% of users thought that their professors viewed their profiles (Cliff reported the collective jaw dropping that occured when he announced to his class of 200 freshmen that he had looked at every student's profile). [I can't find an online version of this paper, but here's another, more comprehensive, paper on the topic -- the CSCW paper was a "Note" -- by the same authors: Spatially Bounded Online Social Networks and Social Capital: The Role of Facebook, by Nicole Ellison, Charles Steinfield and Cliff Lampe, presented at the Annual Conference of the International Communication Association (ICA), June 19-23, 2006 in Dresden, Germany]

Susan Wyche (Georgia Tech Work to Play Lab) presented "Technology in Spiritual Formation: An Exploratory Study of Computer Mediated Religious Communication" [slides], a report on how the leaders of megachurches are using technology to communicate with their congregations. Susan and her co-authors found a variety of religious practices in which senior pastors used technology, including administration, education, counselling, preparation of sermons, processing prayer requests and audio / visual support for presentations. What struck me about all of the examples described in the study was that they support the "command and control" paradigm that characterizes most large organizations (be they military, commercial or religious), rather than the "listen and participate" model that is one of the distinguishing feature of Web 2.0 technologies ... which [hardly surpisingly] did not appear to figure prominently in the religious institutions the authors studied. In an earlier post on spiritual computing, I pondered the differences between religion and spirituality. I believe this paper focuses more on former than the latter, as did a paper on techno-spiritual practices at UbiComp 2006 by Genevieve Bell. I am happy to see religious practices, an important realm of human activity, recieve more attention at these conferences, however, as technology further empowers independent and collective action outside of traditional structures, I hope we will see more work on the use of technology to support extrareligious spiritual pursuits in the future.

Kathy Lee (Stanford University) presented "What Goes Around Comes Around: An analysis of del.icio.us as social space" [slides], in which she investigated the perceptions and effects of social presence for the users of the del.icio.us shared bookmarking system. Kathy described a "return bookmark" mechanism (that I never knew about) wherein one user can subscribe to another user's tags, and when the subscriber reads and tags an item originally tagged by the other user, the original tagger is notified, reinforcing a sense of presence -- and audience. Recipients of return bookmarks are twice as likely as other taggers to annotate their bookmarks, and while one questioner raised the issue of causality (vs. correlation), this does seem like another instance of [the awareness of] the audience providing positive feedback to the performer (similar to the correlation of the number of comments by blog readers to the probability of future blog posts by a blog author).

Paul Dourish (University of California, Irvine) presented "Re-Space-ing Place: 'Place' and 'Space' Ten Years On", a reflective piece on a seminar paper he and Steve Harrison published at CSCW '96 on "Re-Place-ing Space: The Role of Place and Space in Collaborative Systems". The aim of the earlier paper was to take ideas from architecture to infuse meaning [place] into [virtual] spaces. In the new paper, Paul questions his earlier adoption of the layer-cake model of socio-technical systems (proposed by Rob Kling) as not sufficiently accounting for the interactions between the social and technical, and raises issues of spatial practice, power geometries, and views space as collective and cultural, with technology acting as a lens through which it is understood.

A.J. Brush (Microsoft Research) presented "Revisiting Whittaker & Sidner's 'Email Overload' Ten Years Later", a study in which she and her colleagues investigated how email, and the technologies and techniques for processing it, have changed since the publication of a seminal paper on email overload at CSCW 1996. The earlier study was based on 20 users at Lotus, and the newer study was based on a profile of 6000 users at Microsoft. What was most surprising was how little has changed -- the email clients look the same, the average number of new messages received per day hasn't changed significantly (49 in 1996 to 87 today), and the average size of inboxes hasn't changed (approximately 1000). There were a number of interesting points raised in the Q&A session, regarding the impact of multiple email addresses (e.g., personal gmail accounts to get around corporate-imposed quotas, or to partition one's digital selves), the practices around email (e.g., frequency of using email during the day), viewing inboxes as a space, and the preference for email vs. other communication tools such as instant messaging or social networking services. In keeping with the retrospective theme, Paul Dourish shared one of the most interesting quotes I heard at the conference, from an earlier observation of information overload:

One of the diseases of this age is the multiplicity of books; they doth so overcharge the world that it is not able to digest the abundance of idle matter that is every day hatched and brought forth into the world.

-- Barnaby Rich (1580-1617), writing in 1613 (!); quoted by de Solla Price in his 1963 book "Little Science, Big Science."

Rounding out an afternoon of reflections on CSCW was a panel on "Twenty Years of CSCW - What Have We Learned?", with Irene Grief, Jonathan Grudin, Tom Malone, Judy Olson and Lucy Suchman. Irene introduced the panel, providing a history of the conference, and noting that two of the panelists (Jonathan and Judy) have continued to come to CSCW conferences, and two (Tom and Lucy) have not. Jonathan presented more historical background and some interesting statistics about the institutions that dominate CSCW: each year, 50% of the authors come from 6 institutions, and over the twenty-year history of the conference, 50% of the authors have come from a total of 12 institutions ... evidence of an in-group and out-group for this research community. Tom asked the question of why attendance at CSCW (and UbiComp) has been relatively flat during a period where the CHI conference has grown, and though he has not attended CSCW in many years, toward the end said he was hopeful about the future of CSCW. Judy noted a growing "professionalism" in CSCW ... and yet also provided -- or played -- one of the most playful episodes of the conference: a hilarious Prairie Home Companion skit on Linda Wertheimer struggling to communicate over a telephone link with a remote correspondent. Lucy raised the issue of US-centrism and said that she has shifted her focus to other areas, highlighting some of the work of her students along the way.

Pam Hinds (Stanford University, and co-chair of the conference) presented "Structures that Work: Social Structure, Work Structure and Coordination Ease in Geographically Distributed Teams", which reinforced the importance of social ties (in collocated and distributed teams), and emphasized the importance of more transparent work structures and the potential benefits of technology that tracks and makes visible emergent interdependencies.

Bill Buxton (Microsoft Research) presented the closing plenary, ambiguously, or perhaps ambivalently, entitled "Tufte Doesn't Understand the GENESYS of PowerPoint", and announcing that "I am not a scientist anymore, I'm an entertainer". Bill's presentation was, indeed, entertaining, providing more questions than answers. After getting bogged down for a while in an entirely unnecessary (and, I believe, misguided) dismissal of Linux, he went on to raise some important issues with respect to design, collaboration and technology. Claiming the blackboard, invented in 1801, was the most innovative technological development in the history of education, he went on to propose a wager with the CSCW community: "I will buy drinks for everyone at CSCW in 7 years if the following is not true: It will be as cheap (in cost per square feet) in 7 years to buy 100dpi displays as it is to buy whiteboards today". Bill also offered some retrospection, noting the brilliance of Genesys, a line-driven animation system created by Ron Baecker for his Ph.D. Thesis in 1969, and the surprising lack of progress beyond this work ... until recently, when this feature was added to PowerPoint ... and given a somewhat more playful platform in Line Rider. Bill finished up with a number of insightful, potentially inciteful, and inspiring ideas:

• The most important ingredient to science is play (Albert Einstein)
• The worst thing in the world is a precious idea
• The worst person to have on your team is someone who thinks his idea is precious
• Good ideas are cheap, they are not precious
• The key is not to come up with ideas but to cultivate the adoption of ideas

The idea that Bill wants to cultivate is that we can change -- and have changed -- our culture through design, and the goal Bill wants to achieve is to transform Microsoft from the world's leading software company into the world's leading design company in the next five years (although he didn't offer to buy anyone drinks if this does not come to pass).

Bill's talk, as well as several of the other presentations and informal discussions at CSCW, brought up alot of issues for me. Rather than making a long blog entry even longer, I will compost some of my thoughts in a separate entry in the near future.

[Update: Jorge Aranda, from University of Toronto, has posted some notes from the conference, including: "CSCW seems like a misnomer, then, for research that is really about the interaction among people through any kind of technology – from the ways airplane pilots use paper notes, to how World of Warcraft players form in guilds and throw virtual parties with conga lines."]