sociolinguistic reader

8/14/2019 Sociolinguistic Reader

1/14

IEEE TRANSACTIONS ON LEARNING TECHNOLOGIES, MANUSCRIPT ID 1

Video Collaboratories for Research andEducation: An Analysis of Collaboration

Design PatternsRoy Pea and Robb Lindgren

AbstractWeb-based video collaboration environments have transformative potentials for video-enhanced education and forvideo-based research studies. We first describe DIVER, a platform designed to solve a set of core challenges we have identified

in supporting video collaboratories. We then characterize five Collaboration Design Patterns (CDPs) that emerged from

numerous collaborative groups who appropriated DIVER for their video-based practices. Collaboration Design Patterns (CDPs)

are ways of characterizing interaction patterns in the uses of collaboration technology. Finally we propose a three-dimensional

design matrix for incorporating these observed patterns. This representation can serve heuristically in making design

suggestions for expanding video collaboratory functionalities, and for supporting a broader constellation of user groups than

those spanned by our observed CDPs.

Index TermsVideo, collaboration, CSCL, collaboratory, design patterns, Web 2.0, metadata, video analysis.

1 INTRODUCTION

E argue in this paper that the proliferation of digi-tal video recording, computing, and Internetcommunications in the contexts of social sciences

research and learning technologies has opened up dra-matic new possibilities for creating video collaborato-ries. Before describing our vision for video collaborato-ries, and our experiences in designing and implementingthe DIVER platform for enabling video collaboration, webriefly sketch the historical developments in video useleading to the opportunities at hand.

Throughout the 20th century, film and later videotechnology has been an influential medium for capturingrich multimedia records of the physical and social worldsfor educational purposes [1] and for research uses in thesocial sciences [2]. While K-20 education is still largelydominated by textual representations of information anduses of static graphics and diagrams, during the 21 st cen-tury the education and research communities are moreregularly exploiting video technologies, and in increa-singly innovative ways. For example, with the develop-ment of more learner-centered pedagogy, uses of videoare expanding from teachers simply showing videos tostudents to approaches where learners interact with,create, or comment on video resources as part of their

knowledge-building activities [3], [4], [5], [6]. In research,with the proliferation of inexpensive digital consumervideocameras, and software for video editing and analy-sis, individual researchers and research teams are captur-ing more data for studying the contextual details of learn-ing and teaching processes, and the learning sciences

community has begun experimenting with collaborativeresearch infrastructures surrounding video datasets [7],[8], [9], [10].

We view the Web 2.0 participatory media culture illu-strated by media-sharing community sites [11] as exem-plifying how new forms of collaboration and communica-tion have important transformative potentials for moredeeply engaging the learner in authentic forms of learn-ing and assessment that get closer to the experiences ofworldly participation rather than more traditional de-contextualized classroom practices. Video representationsprovide a medium of great importance in these transfor-mations, in capturing the everyday interactions betweenpeople in their physical environments, during their en-gagements in cultural practices, and using the technolo-gies that normally accompany them in their movementacross different contexts. For these reasons, video is im-portant both as a medium for studies of learning and hu-man interaction, and for educational interventions. Inresearch the benefits of video have been well evidenced.Learning scientists have paid increasing attention overthe past two decades to examining human activities innaturalistic socio-cultural contexts, with an expansion offocus from viewing learning principally as an internal

cognitive process towards a view of learning that is alsoconstituted as a complex social phenomenon involvingmultiple agents, symbolic representations, and environ-mental features and tools to make sense of the world andone another [12], [13].

This expansion in the central focus for studies of learn-ing, thinking, and human practices was deeply influencedby contributions involving close analyses of video andaudio recordings from conversation analysis, sociolin-guistic studies of classroom discourse, anthropologicaland ethnographic inquiries of learning in formal and in-formal settings, and studies of socially-significant non-

xxxx-xxxx/0x/$xx.00 200x IEEE

W

The authors are with the H-STAR Institute and the School of Educationat Stanford University, Stanford, CA 94305. E-mail: {roypea, rob-blind}@stanford.edu.

anuscript received 9 Dec. 2008.


2/14

2 IEEE TRANSACTIONS ON LEARNING TECHNOLOGIES, MANUSCRIPT ID

verbal behaviors such as body language or kinesics, ges-ture, and gaze patterns. This orientation to understandinglearning in situ has led researchers to search for tools al-lowing for capture of the complexity of real-life learningsituations, where multiple simultaneous 'channels' of in-teraction are potentially relevant to achieving a deeperunderstanding of learning behavior [2]. Uses of film andaudio-video recordings have been essential in allowingfor the repeated and detailed analyses that researchershave used to develop new insights about learning andcultural practices [14], [15]. In research labs throughoutdepartments of psychology, education, sociology, linguis-tics, communication, (cultural) anthropology, and hu-man-computer interaction, researchers work individuallyor in small collaborative teamsoften across discip-linesfor the distinctive insights that can be brought tothe interpretation and explanation of human activitiesusing video analysis. Yet there has been relatively littlestudy of how distributed groups make digital video anal-ysis into a collaborative enterprise. Nor have there beentools available that effectively structure and harvest col-

lective insights.We are inspired by the remarkable possibilities for es-

tablishing video collaboratories for research and foreducational purposes [7], [16]. In research-oriented videocollaboratories, scientists will work together to share vid-eo datasets, metadata schemes, analysis tools, coding sys-tems, advice and other resources, and build video analys-es together, in order to advance collective understandingof the behaviors represented in digital video data. Virtualrepositories with video files and associated metadata willbe stored and accessed across many thousands of fede-rated computer servers. A large variety of types of inte-ractions are increasingly captured in video-data, with

important contexts including K-20 learningas in ratioand proportion in middle-school mathematics or collegereasoning about mechanics; parent-child or peer-peersituations in informal learning; surgery and hospitalemergency rooms and medical education; aircraft cock-pits or other life-critical control centers; focus-groupmeetings or corporate workgroups; deaf sign-languagecommunications; and uses of various products in theireveryday environments to help guide new design (in-cluding cars, computers, cellphones, household ap-pliances, medical devices), and so on. Corresponding op-portunities exist for developing education-centered videocollaboratories for the purposes of technology-enhanced

learning and teaching activities that build knowledgeexploiting the fertile properties we have mentioned ofaudio-video media. It is our belief that enabling scientificand educational communities to develop flexible and sus-tained interactions around video analysis and interpreta-tion will help accelerate advances across a range of dis-ciplines, as the development of their collective intelli-gence is facilitated.

We recognize how this vision of widespread digitalvideo collaboratories used throughout communities forresearch and for education presents numerous challenges.The process of elucidating and addressing these chal-lenges can be aided considerably by exploring emerging

efforts to support collaborative video practices. In thispaper we describe the features of a particular digital vid-eo collaboratory known as DIVER. Using the large vo-lume of data that we have collected from DIVER users weare able to describe the substantial challenges associatedwith establishing collaboration around digital video usingexamples from real-world research and educational prac-tices. This dataset also permits us to extrapolate futurecollaboration possibilities and the new challenges theycreate. We end by presenting dimensions for organizingour vision of digital video collaboraties that we hope willprovide entry points for researchers and designers to en-gage in its further realization.

2 DIVER:DIGITAL INTERACTIVE VIDEOEXPLORATION AND REFLECTION

2.1 The Need for Supporting Video Conversations

We can distinguish three genres of video in collaboration.The first is videoconferencing, which establishes synchron-ous virtual presence (ranging from Skype/iChat video on

personal computers to dedicated room-based videoconfe-rencing systems such as HP's Halo) where video is themedium, as collaboration occurs via video. The second isvideo co-creation (e.g., Kaltura, Mediasilo)where video isthe objective, and the collaboration is about making vid-eo. The third is video conversationswhere video is thecontent, and the collaboration is about the video. We feelthat video conversations are a vital video genre for learn-ing and education because conversational contributionsabout videos often carry content as or more importantthan the videos themselvesthe range of interpretationsand connections made by different people, which pro-vides new points of view [8], [9] and generates important

conceptual diversity [17]. For decades, video has beenbroadcast-centricconsider TV, K-12 education or corpo-rate training films, and e-learning video. But with thegrowth of virtual teams, we need multi-mediated collabo-ration infrastructure for sharing meaning and iterativeknowledge building across multiple cultures and pers-pectives. We need video infrastructure that is more inte-raction-centricfor people to communicate deeply, pre-cisely, and cumulatively about the video content.

In our vision of video collaboratories, effectively sup-porting video conversations requires more than the capa-bilities of videoconferencing and net meetings. One re-quirement concerns a method for pointing to and anno-

tating parts of videosanalogous to footnoting for textwhere the scope of what one is referring to can be madereadily apparent. We are beginning to see this capabilityemerge with interactive digital video applications online.In June 2008, YouTube enabled users to mark a spotlightregion and create a pop-up annotation with a start andend time in the video stream. Flash note overlays on topof video streams are also provided in the popular Japa-nese site Nico Nico Douga launched in December 2006,where users can post a chat message at a specific momentin the video and other chat messages other users haveentered at that time point in the video stream togetheracross the video as it plays. Similar capabilities of deep


3/14

PEA AND LINDGREN: COLLABORATION DESIGN PATTERNS IN USES OF A WEB-BASED VIDEO RESEARCH AND EDUCATION PLATFORM 3

tagging of video were illustrated in the past few years byBubblePly, Click.tv, Eyespot, Gotoit, Jumpcut, Motion-box, Mojiti (acquired by News Corporation/CBS jointventure Hulu), Veotag and Viddler.

While the virtual pointing requirement is necessary

for supporting online video conversations, it is not suffi-cient. It is important to distinguish between simple anno-tation and conversationthe former can be accomplishedwith tools for coupling text and visual referents, while thelatter requires additional mechanisms for managing con-versational turn-taking and ensuring multi-party en-gagement with the target content. While there are numer-ous software environments currently supporting videoannotation, the number of platforms that support videoconversations is much smaller1. We focus here on a soft-ware platform called DIVER that was developed in ourStanford lab with the objective of supporting video con-versations. In addition to providing a unique method for

pointing and annotating, DIVER also possesses functio-nality for facilitating and integrating multi-user contribu-tions.

2.2 DIVER as an Example of a VideoConversations Platform

DIVER is a software environment first developed as adesktop software system for exploring research uses ofpanoramic video records that encompass 360-degree im-agery from a dynamic visual environment such as a

1VideoTraces [18] is one system that we are familiar with that strives tosupport rich asynchronous discourse with video using gestural and voiceannotations in a threaded format.

classroom or a professional meeting [19]. The Web ver-sion of the DIVER platform in development and use since2004 allows a user to control a virtual camera windowoverlaid on a standard video record streamed through aweb browser such that the user can point to the parts of

the video they wish to highlight (Fig. 1). The user canthen associate text annotations with the segments of thevideo being referenced and publish these annotationsonline so that others can experience the users perspectiveand respond with comments of their own. In this way,DIVER enables creating an infinite number of new digitalvideo clips and remix compilations from a single sourcevideo recording. As we have modified DIVER to allowdistributed access for viewing, annotation, commentaryand remixing, our focus has shifted to supporting colla-borative video analysis and the emerging prospects fordigital video collaboratories. DIVER and its evolving ca-pabilities have been put to work in support of collabora-

tive video analysis for a diverse range of research andeducational activities, which we characterize in the nextsection.

We refer to the central work product in DIVER as adive (as in dive into the video). A dive consists of a setof XML metadata pointers to segments of digital videostored in a database and their associated text annotations.In authoring dives on streaming videos via any web-browser, a user is directing the attention of others whoview the dive to see what the author sees; it is a processwe call guided noticing [16], [19]. To author a dive withDIVER, a user logs in and chooses any video record in thesearchable database that they have permission to access

Fig. 1. The DIVER user interface. The rectangle inside the video window represents a virtual viewfinder that is controlled by the users

mouse. Users essentially make a movie inside the source movie by recording these mouse movements.


4/14


(according to the groups to which they belong). A divecan be constructed by a single user or by multiple userseach of whom contributes their own interpretations thatmay build on the interpretations of the other users.

By clicking the Mark button in the DIVER interface(see Fig. 1), the user saves a reference to a specific point inspace and time in the video. The mark is represented by athumbnail image of the video record within the DIVER

worksheet on the right side of the interface. Once themark is added to the worksheet, the user can annotatethat mark by entering text in the associated panel. A pan-el is also created by clicking on the Record button, anaction that creates a pointer to a temporal video segmentand the video content encompassed within the virtualviewfinder path during that segment. Like a mark, a rec-orded clip can be annotated by adding text within its as-sociated DIVER worksheet panel. The DIVER user canreplay the recorded video path movie or access the pointin the video referenced in a mark by clicking on itsthumbnail image. DIVER is unique among deep videotagging technologies in enabling users to create and anno-

tate panned and zoomed path movies within streamingweb video.In addition to asynchronous collaborative video analy-

sis using DIVER, multiple users can simultaneouslyaccess a dive, with each user able to add new panels ormake comments on a panel that another user has created.Users are notified in real time when another user hasmade a contribution. Thus users may be either face-to-face in a meeting room or connected to the same webpageremotely via networking as they build a collaborativevideo analysis. There is no need for the users to be watch-ing the same portions of the video at the same time. Asthe video is streamed through their browsers, users may

mark and record and comment at their own pace and ac-cording to their own interests.

The DIVER user may also create a compilation or "re-mix" of the contents of the dive as a stand-alone presenta-tion and share this with collaborators by sending emailwith a live URL link that will open a DIVER server page(see Fig. 2) and display a DIVER player for viewing thedive including its annotations.

The current architecture of WebDIVER has three parts:(1) a video transcoding server (Windows XP andFFMPEG); (2) application servers that are WAMP XP-based (Windows XP, Apache Web Server, MySQL data-base for FLV-format videos, thumbnail images, XML filesand logs, and PHP); and (3) the streaming media server(Flash Media Server 2, Windows Server 2003). Progress isunderway in moving WebDIVER into Amazon ElasticCompute Cloud (Amazon EC2), using Amazon's S3 sto-rage service, converting to Linux, and replacing AdobeFlash Media Server with open-source Red5 so that we canhave 100% open source DIVER.

2.3 Socio-technical Design Challenges for Video

CollaboratoriesThe web version of DIVER was designed specifically toaddress a set of core problems in supporting the activitiesof participants involved in distributed video collabora-tions [20]. These problems have surfaced in various multi-institutional workshops convening video researchers [21],[22], [23], [15], and many of the concerns relate to thefundamental problem of coordination of attention, inter-pretation and action between multiple persons. HerbClark [24], in characterizing non-technology mediatedcommunication, described as common ground whatpeople seek to achieve as they coordinate what they are

Fig. 2. DIVER remix player for viewing dive content outside of the authoring environment (on a webpage or blog).


5/14


attending to and/or referring to, so that when commentsare made, what these comments refer to can be appro-priately inferred, or elaborated. We expand the commonground aspect of communicative coordination to the needto refer to specific states of information display in usingcomputer tools, including digital video in collaborativesystems.

Of course a video conversation system such as DIVERdoes not fully resolve the common ground problem anymore than pointing does in the physical world. Whilepointing is a useful way of calling attention to something,as Clark and others such as Quine [25] have pointed out,there can still be referential ambiguity present in a point-ing act even for face-to-face discourse (e.g., is she refer-ring to the car as an object or the color of the car?). How-ever, it is possible to design a software platform withfunctionality that makes it possible for participants tonegotiate the identity of a referent and its meaning over pro-gressive conversational turns. Marking points of interestin a video using DIVERs virtual viewfinder and creatingpersistent references to video events are important steps

forward in addressing these coordination challenges. Auser who offers an interpretation on a movie moment in aDIVER worksheet can be confident that the referent oftheir commentsat least in terms of spatial and temporalpositionwill be unambiguous to others who view thedive. This guided noticing feature provides the basis forsubsequent interpretive work by group members to aligntheir loci of interpretive attention (e.g., What is this? Whatis happening here?). In the next section we examine theeffects of introducing this capability to users with varyinginterests and goals for working with video.

We summarize here the core socio-technical designchallenges we addressed in the design of DIVER [20]:

1. The problem of reference. When analyzing video, howdoes one create a lasting reference to a point in space andtime in the dynamic, time-based medium of a videorecord (a video footnote)?2. The problem of attentional alignment, or co-reference. Howdoes one know that what one is referring to and annotat-ing in a video record is recognized by one's audience?Such coordination is important because dialog about avideo referent can lead to conversational troubles if one'saudience has a different referent in mind. This process ofdeixis requires a shared context.3. The problem of effective search, retrieval and experiencing ofcollaborative video work. If we can solve the problems of

allowing users to point to video and express an interpre-tation, and to establish attentional alignment, new digitalobjects proliferate beyond the video files themselves (di-ves) that need to be searchable and readily yield the us-ers' experience of those video moments that matter tothem.4. The problem of permissions. How does one make the shar-ing of video and its associated interpretations more avail-able, while maintaining control over sensitive data thatshould have protection for human subjects or digitalrights management?5. Integrating the insights of a collective group. How can wesupport harvesting and synthesizing the collective intelli-

gence of a group collaboratively analyzing video?6. The problem of establishing coherent multi-party video-anchored discourse. Consider a face-to-face conversationalinteraction, as in a seminar, where the rules of discoursethat sustain turn-taking and sense-making as people con-verse are familiar. In an academic setting, video-, film andaudio-recordings, as well as paper, may be used. Tradi-tionally these are used asymmetrically, as the facilita-tor/instructor prepares these records to make a point orto serve as an anchor for discussion, and controls theirplay during the discourse. Computer-facilitated meetingsfor doing video analysiswhere each participant has acomputer and is network-connected with other partici-pants and to external networks for information accessbring new challenges in terms of managing a coherentgroup discourse.

3 COLLABORATION PATTERNS IN DIVER

We now investigate what collaboration design patternsare used by groups of educators and researchers when

they have access to DIVER's web-based video collabora-tion platform. The DIVER software platform allows usersto flexibly exchange analyses and interpretations of a vid-eo record with relatively few constraints on their collabo-ration style or purpose. Once the DIVER software wasweb-enabled and made publicly accessible in 2004, wetook the approach of supporting any users with needs forengaging in video exploration as part of their profession-al, educational, or recreational practices. We sought toaccommodate a range of video activities, but rarely didwe actively recruit users or make customizations to sup-port specific applications. The result has been a large anddiverse international user base. There have been approx-

imately 3000 unique users who have registered with DI-VER and approximately 200 private user groups thathave been created to support the activities of a particularproject, event, or organization. Since DIVER was createdin an academic setting the majority of users are affiliatedwith schools and universities; however, there have alsobeen users from the private sector, and even within a sin-gle setting, we have found DIVER to be used for a num-ber of different purposes.

To conduct this analysis we examined every dive thatwas created by each of the user groups and characterizedthe way that each group had organized their discourse:what interface features were used most frequently, whatannotation conventions the group adopted, how conver-sational turn-taking was managed, etc. These characteri-zations were agnostic as to who the particular partici-pants were in the collaboration (e.g., high school studentsor researchers), and we regularly observed behaviors thattranscended such user categories. In reviewing these cha-racterizations we were able to extract a small set of dis-tinct patterns that describe the ways groups collaboratedto achieve their particular video learning objectives. Wethink of these patterns as akin to Collaborative Design Pat-terns (CDPs) [26]a conceptual tool for thinking abouttypical learning situations. In particular, CDPs have beenemployed in the literature on computer-supported colla-


6/14


borative learning to characterize interaction patternsaround the use of learning technologies such as handheldcomputers [27]. Collaboration patterns we observed withDIVER were emergent, and not behaviors that we pre-scribed or that were dictated by the software, yet theywere still comprised of the standard CDP elements (e.g., aproblem, a context, a solution approach, etc.). Conceptua-lizing our observed patterns as CDPs is useful because itwill allow us to make design suggestions for supportingspecific kinds of user groups and to elicit patterns of be-havior that have shown themselves to be particularly ef-fective for such groups.

Here we briefly describe the five most notable collabo-ration patterns we observed in our analyses of group ac-tivity in DIVER. Our dataset encompasses numerous in-stances of each of these patterns, but we will describeeach with a single example to elucidate its key features.1. Collective Interpretation

Making interpretations of human behaviorparticularly of learningis a complex enterprise thatbenefits from multiple perspectives [28], [29]. The impor-

tance of collecting multiple interpretations is known in-stinctively by most, and so when an individual sets out toconduct an analysis of a video event he or she will oftenrecruit the input of others who may bring novel insightsbased on their differences in knowledge and experience.Coordination of multiple perspectives, however, has longbeen a challenge for video analysts, especially in the daysof VHS tapes where a single individual had playback con-trol and had to somehow harvest the insights of a roomfull of contributors [14]. The DIVER platform has affor-dances for sharing and comparing insights in a morecoordinated manner, and we have observed numerousgroups use DIVER in an attempt to achieve interpretive

consensus, or to refine points of contention.An example of this pattern of activity occurred within

the context of a teacher credential program at a large eastcoast university. Students in a course on teaching andlearning were provided with sample video of a scienceteacher working with her students to help them under-stand animal classification systems. The student-teacherswho watched the video were tasked with describing howthe instructional approach of the teacher in the video inte-racted with her students learning processes. By givingeach student-teacher independent access to the focalevent in the DIVER system they were able to anchor theirinterpretations directly with video data rather than hav-

ing to rely on their memory of what they had seen. Apublic record of these interpretations is maintained sothat another contributor can offer a response hours ordays later.

At one point in the science classroom video, the teach-er asks one of her students about what they have learnedfrom an animal classification activity. In the video thestudent says: It helped me realize what things have incommon and what they dont have in common. One ofthe student teachers (ST1) selected this clip from the vid-eo and moved it to the DIVER worksheet, which led tothis exchange with another student-teacher (ST2):

ST1:This student found the activity helpful in terms of un-derstanding what an evolutionary biologist does. He seemsto have gotten the big picture.

ST2: I don't know if I agree that he has gotten the big pic-ture. Though I admit that I could be thrown off by his affect,which is pretty flat, it seems like he is just saying the mostbasic things--we were putting out species in different cate-

gories, they [biologists] categorize certain species or ani-mals, I found it helpful because it helped me realize whatthings have in common and what don't. Is this the depthyou were hoping for? How could we move him to the nextlevel?

Importantly, the contribution of ST2 likely changedthe interpretation of this students reflection, not just inthe mind of ST1, but for all group participants who arenow thinking about how the instructional interventionhad failed to achieve deep student understanding. HadST2 simply stated: I disagree that the student has gottenthe big picture in an un-anchored group discussion of

the event there would have been a greater chance thatinterpretive disagreement would have persisted. On thecontrary, ST2 grounded her assertions in the actual class-room events, effectively shifting the analytical focus forsubsequent interpretations.

We observe this pattern of collaboration in DIVER fre-quently. It typically emerges from groups who have theopen-ended task of determining what is going on in avideo event, ranging from field data collected by a re-search team to a funny video that someone posts on You-Tube. The pattern often starts with one person taking apass at making their own interpretation, and then otherschime in with support or criticism. DIVER gives this dis-course structure, such that an outside observer can fairlyeasily view a dive and ascertain the general consensus.2. Distributed Design

There can be considerable value in collecting and con-sidering video of the user experience in the designprocess [30], but the logistics of incorporating this videointo a design teams workflow can be tricky. We haveencountered a number of well-intentioned design groupswho collect hundreds of hours of video of users, only forthis video to sit unwatched on a shelf of DVDs. Video cancapture notable affordances or systematic failures in theuser experience, but identifying these trends and com-municating them to the rest of the team at critical pointsin the design process is an arduous, time-consuming task.

It is not typically feasible for an entire design team toconvene and spend hours watching videos of user testing.We have, however, observed the members of a number ofdifferent design initiatives use DIVER as a tool for effi-ciently integrating video insights into their collaborativethinking and design reviews.

The distributed design pattern was observed in a smallresearch team at a west coast university working on aprototype touch-screen interface for preschool children.The researchers were trying to create a tool that allowedchildren to construct original stories using stock videofootage (e.g., baby animals in the wild). Access by thedesign team to children of this age was limited, but there


7/14


was a real need for the team to understand how the child-ren would react to this kind of interface and whether itwas possible to scaffold their developing storytelling abil-ities with a novel technology. The design team arrangedfor a few pilot user sessions to be conducted at a localnursery school, but neither the lead researcher on theproject nor the software engineer who was responsible forimplementing interface changes were able to attend. Thesolution was for two graduate students to run the pilotsessions and immediately upload the session videos toDIVER so that the entire team could quickly formulatedesign modifications that could be implemented for thenext iteration (Fig. 3).

Fig. 3. DIVER worksheet showing an instance of distributed design.

In this instance, one team member took a first pass atsegmenting and highlighting the important events in the

video. This allowed the rest of the team to focus in on themoments that had potential design implications. In sever-al instances, a team member would make a design rec-ommendation based on something that they observed inthe video, and one of the team members present at thesession would respond with a clarification of what hadoccurred and possible alternative design schemes. Thediscourse in DIVER quickly transitioned from an investi-gation to a design plan that was adopted by the softwareengineer and implemented on a short timeline.

As a pragmatic matter, design teams in all areas of de-velopment will typically rely on data summaries and ag-gregations of user-test findings to inform the design

process. While understandable, this practice can distancedesigners from the valuable insights to be derived fromobserving key moments in video recordings of users. Asystem for supporting asynchronous references of specificspace-time segments of videosuch as the one found inDIVERseems to alleviate some of the coordinative bar-riers and permit the integration of user data with the de-sign process. Recognizing this potential, a number of DI-VER user groups found success generating new designsfor learning and education through their discussions ofvideo-recorded human interactions.3. Performance Feedback

A challenge of giving people feedback on a perfor-

mance, whether a class presentation or some display ofartistic skill, is that the feedback offered is typically sepa-rated from the performance itself, such as a verbal evalua-tion given a week later or a review that has been writtenup in a newspaper. The effectiveness of the feedback inthese cases relies on the memory of the performance bothfor the persons giving the review, as well as the personreceiving the feedback. If, for example, a student does notrecall misquoting a philosopher in their presentation for alaw school course, they are not likely to be receptive tohaving this pointed out by their professor or classmates.Associating performance feedback with actual segmentsor images from the performance has an intuitive utility,but the approach to implementing a feedback system thateffectively makes these associations and integrates inputfrom multiple individuals is less clear. Using DIVER asmeans to deliver feedback on video-recorded perfor-mance was one of the most common collaboration activi-ties that we observed among our users. We were particu-larly impressed by the range of performance types (e.g.,films produced by undergraduates; K-12 student teach-

ing) for which users attempted to use DIVER for convey-ing suggestions and making specific criticisms.

A good example of the performance feedback patternis its use in a prominent U.S. medical school where aneffort was made to improve the manner by which stu-dents communicated important health information topatients. Medical students interning at a hospital wereasked to record their consultation sessions with patientsand submit these recordings to their mentors using DI-VER. Experienced medical professionals offered thesestudents constructive feedback on how they were com-municating with their patients and provided suggestionsfor improvement. In some cases, there was also the op-

portunity for the students to respond to the feedback withquestions or points of clarification. In one instance a stu-dent has uploaded her interview with a patient who hascome in with various ailments including abdominal pain.She asks numerous questions in her attempt to narrowdown the underlying problem. In this dive she has donesome self-evaluationmarking segments and makingcomments where she believes that she could have donesomething better. Additionally her mentor (M) haswatched the video and marked his own segments uponwhich he bases his evaluation. He has marked one mo-ment in the video in particular and made the followingcomment:

M:Good check here on the timing of her GI symptoms inthe bigger picture - you are now making sure you knowwhere these symptoms fit into the time course of this com-

plicated history.

Note that the evaluator used the word here ratherthan having to describe the referent event in detail, as onewould likely have to do if they were delivering an entire-ly written or verbal assessment. Targeted feedback of thiskind should help to minimize misunderstandings or ge-neralizations. People will sometimes over-react to nega-tive feedback on their performance, concluding hastilythat she hated it or I cant do anything right, but with


8/14


feedback that is linked to specific behaviors, there is abetter chance that the evaluations will be received con-structively. In a related vein, a recent paper describesproductive uses of DIVER for changing the paradigm ofcommunication skills teaching in oncology to a more pre-cise performance feedback system rather than one princi-pally based on observation [31].

Additionally we observed users administer feedbackas a group, such as when an entire class was asked to re-spond to a student presentation. In this case the studentswere aware of and were able to coordinate their feedbackwith that offered by their classmates, leading to less re-dundancy. Users in such a design pattern also have theopportunity to mediate each others feedback, perhaps bygiving support for anothers comments with additionalideas for improvement or by softening criticism that maybe seen by some as overly harsh.4. Distributed Data Coding

Some attempts at making interpretations of video re-cordings of human activity are more structured thanthose that we observed in the distributed interpretation

pattern. In research settings in particular, the categories ofactivity that are of interest are often clear, while it is theidentification of those categories and formal labeling ofevents in the data that must be negotiated. In their expe-rience coding video for the TIMSS studya cross-culturalstudy of math and science classroomsStigler and col-leagues [32] reflect on two lessons they learned aboutcoding video data: (1) the videos must be segmented inmeaningful ways to simplify and promote consistency incoding, and (2) the construction and application of thesecodes requires input from multiple individuals with dif-fering expertise. Both of these issues can be facilitated bya system like DIVER that structures video discourse

around segments selected by any number of participants.While only a few user groups implemented a formal cod-ing scheme in DIVER, their interaction style took on adistinct pattern that is useful to consider for thinkingabout possible applications and design needs.

One group of users that utilized DIVER to managetheir distributed coding work was a team of seven re-searchers working as part of a large Center dedicated tothe scientific study of learning in both formal and infor-mal environments. This team conducted over 20 inter-views with families in the San Francisco Bay area on howuses of mathematics arise in their home life. The researchgroup was interested in how families of diverse back-

grounds organize their mathematical practices and toanalyze how differences in social conditions and re-sources in the home support these practices. The videosfrom all the interviews were uploaded into DIVER, butfor all of the two-hour videos to have been logged andcoded by the entire group would have been overly bur-densome and ultimately counter-productive. Instead thegroup reached consensus on a coding scheme by review-ing a subset of the interviews as a group, and then as-signed two different team members to code each of theremaining videos. The codes that they agreed upon werelabels for specific instances that they were interested instudying as part of their analysis, such as gesture

which denoted a family member had made some physicalgesture to illustrate a mathematical concept, or values which was used when someone made an expression oftheir familys values when discussing a mathematicalproblem that arose in their home life. The consistent useof these codes in DIVER was particularly useful becausethe software allows users to search for keywords in theannotations of the video segments. The results of a searchare framesfrom multiple dives where the word or code wasused, meaning that this research team was able to quicklyassemble all the instances of a particular code that hadoccurred across all the interviews, regardless of who hadassigned the code. This facilitated the process of makinggeneralizations across families and drawing conclusionsthat addressed their hypotheses.

While other tools for video analysis and coding exist(such as commercial systems ATLAS.ti, NVivo and Studi-oCode), most are not available online nor do they supportmultiple users in collaborative analyses. In its currentstate, DIVER is probably not sufficient for large-scale vid-eo codingthe research group described above, for ex-

ample, supplemented their analysis with a FileMaker Prodatabase that held multiple data fields and detailed codespecificationsbut more sophisticated coding capabilitiesare not difficult to add and are currently in development.What is notable about the distributed coding pattern thatwe observed in existing DIVER users is that the key cod-ing needs of segmenting and supporting multiple pers-pectives [32] were both supported and utilized.5. Video-Based Prompting

Most of the collaboration patterns discussed so farhave consisted of fairly open-ended interpretive work.Even the distributed coding pattern, though using a set ofpredefined categories, still required video segmentation

and the reconciliation of coded events where there wasdisagreement. Some of the activity that we observed inDIVER, however, was far more constrained. We observedsuch a pattern most frequently in formal educational con-texts such as classrooms where instructors had specificquestions about a video event that they wanted their stu-dents to try and answer. For example, a film studies pro-fessor had a set of questions that he wished to pose to thestudents in his class about classic films like GodardsBreathless. He used DIVER to distribute these questionsbecause he wanted his questions and his students answersto be anchored by clips from the actual film. In some ofthe DIVER user groups there was an instructor or facilita-

tor that initiated a dive by capturing a clip from a largervideo source and using the worksheet to pose a questionabout some aspect of the clip.

An especially interesting application of DIVER andgood example of the video-based prompting pattern wasan undergraduate Japanese language course at a westcoast university. The instructor of this course had col-lected a number of videos of informal spoken Japanesefrom various sources such as interviews and televisionprograms. These videos demonstrated certain styles andforms of the language that the instructor wished her stu-dents to experience and reflect upon. These videos wereuploaded into DIVER and the instructor created home-


9/14


work assignments where students would have to respondto 4 or 5 questions, each associated with a video segmentselected by the instructor. In this class, the students wereall able to see each others responses, which turned out tobe a benefit in the eyes of the instructor because it stimu-lated the students to think deeply and try and make anoriginal contribution to the analysis. Fig. 4 is a screenshotof the worksheet used for one of these homework as-signments.

Fig. 4. A DIVER worksheet used in a Japanese language course.

The instructor has posted questions for her students about informal

language conventions used in the video.

Instructors and facilitators of various kinds will fre-quently use visual aids such as video for encouragingcritical thinking by their students. A web services plat-form like DIVER allows students and participants to takethese visual aids home with them, make independent

interpretations, and then contribute their input in a struc-tured forum. The important feature of this pattern is thatany interpretation is necessarily associated with videoevidence of the target phenomena. This practice has po-tential long-term benefits in that it teaches people to ade-quately support their arguments by making direct links toavailable data.

4 MAPPING THE SPACE OF COLLABORATIVE VIDEOANALYSIS TASKS

There are numerous learning situations that can be en-hanced with video analysis, resulting in different patterns

of collaboration, each with different needs in terms ofinterface supports and structure. DIVERs open-endedplatform allowed many of these patterns to emerge or-ganically, but it was clear that several of these collabora-tive practices could be enhanced with additional featuresand capabilities. While DIVER has taken modest stepsforward on this frontadding coding, transcription, andclip trimming functionality, for examplewe recognizethat there is still much progress to be made to fully ex-ploit the learning potential of digital video for researchersand for education.

In order for the community of learning technology re-searchers and developers to adequately address this de-

sign problem, we have attempted to map out the space ofcollaborative video practices. In extracting the five colla-boration patterns from our dataset, we recognized a fewsalient dimensions that we can abstract from these prac-tices for defining this space. These dimensions capture allof the practices we observed, but more importantly theysuggest a number of other practices that we did not ob-serve. Perhaps due to limitations in DIVER or video tech-nologies in general, or simply incidental to the needs ofthe DIVER user community to date, there were severalpossible and promising collaboration activities that havenot yet manifested themselves in mainstream video prac-tices. By specifying the features of this space in its entire-ty, our hope is that we can more fully address the designneeds of existing practices as well as cultivate fledglingpractices that could have a significant impact on the cul-ture of learning technology. The three dimensions thatdefine this space are the style of discourse, relationship to thesource material, and target outcome.1. Discourse Style

In discussing and interpreting the events within a vid-

eo, the needs of some groups are best met by employing amore informal structure with fewer constraints on partic-ipation. Exploratory analyses or discourse around videofor purely social purposes are likely to adopt this moreconversational style. Other groups, however, such as incourses, have specific aims for their collaboration andparticipatory roles that must be maintained during thecourse of discussing the video. These groups will adopt amore formal discourse style that may involve promptingparticipants for desired input or limiting contributions tospecific times or locations within the video.2. Relationship to Source Material

Digital video technologies have made tremendous ad-

vances over the last decade, particularly in the capacityfor capturing, uploading and sharing video at rapidspeeds and with relatively little effort. The ubiquity andease of transmission of digital video has changed the tra-ditional relationships one has with the video they watchfor recreation or utilize in a professional capacity. On oneend of this dimension is an insiderrelationship, meaningthat the video used for the collaborative activity is videowith which the group has strong familiarity. It could bevideo that someone in the group recorded or video inwhich the group members themselves are featured. Withinsider videos the group often has some degree of controlover how the video was recorded and for what purpose.

If a group has an outsiderrelationship with a source videothey typically have less control over factors such as edit-ing style and camera orientation. These are videos thatmay have been obtained from a secondary source or wereselected from an archive collection. It is less likely thatgroups discussing outsider videos will possess back-ground information or be able to fill in comprehensiongaps stemming from contextual features and events thatwere not recorded.3. Target Outcome

Any group of individuals that sets out to explore avideo record asks themselves: What do we want to getout of this? As should be apparent from our review of


10/14


collaborative video activities in DIVER, there are numer-ous possible answers to this question. We have identifiedfour general types of outcomes that, while not exhaustiveor mutually exclusive, describe the bulk of possible colla-boration objectives: design, synthesis/pattern-finding, evalua-tion, and analysis/interpretation. Groups are working to-wards design outcomes when they are discussing videowith the aim of conceiving or improving upon someproduct, process, or organizational scheme. Synthesis orpattern-finding outcomes come from attempts to reachconsensus on the big picture and recognize importanttrends and commonalities. Evaluation outcomes are criti-ques of video products and recorded events with the cor-responding aim of either delivering feedback to a specificgroup or for demonstrating critical competence (as in a K-12 teacher performance assessment). Unlike synthesisoutcomes, groups with an analysis or interpretation objective are trying to break things down and typicallyare attempting to understand what is happening behindthe scenes or in minds of the actors that is causing theevents seen in the video.

Explicating these dimensions results in a matrix of col-laboration activities represented in Figure 5. We cautionthat the states of each dimension that we have identifiedhere do not have hard boundariesit is perfectly feasi-ble that a collaboration activity around video couldstraddle the line between a formal and informal discoursestyle, for example. However, we believe that this repre-sentation provides a useful depiction of the range of pos-sibilities and the various needs of groups that fall withinthis space. To clarify these possibilities and needs evenfurther, we have offered examples of activities that em-body the defining characteristics of each cell. Some ofthese examples are actual activities that we have observed

in DIVER and some are hypothetical activities that sharethe same characteristics of those that we have observed.Other examples in this representation (in grey typeface)are hypothetical activities with characteristics that wehave not yet observed directly in DIVER but are possiblepresuming that the right configuration of situational va-riables exists. All the examples offered in Figure 5 aresituated within a particular user group (teenagers, indus-try professionals, etc.), but again we reiterate our convic-tion that these activity patterns could be implemented inany number of educational and research settings.

These activities should be of particular interest to de-signers and educators because they present powerful

learning opportunities with digital video that simplyhave not been realized with current toolsets. In the mean-time, there is still a great deal of work that can be done toaddress the activity patterns that we have observed,whether they are minor modifications to a software plat-form like DIVER or the development of an entirely newtechnology that targets a specific subset of the space wehave defined here.

5 DESIGNING FOR VIDEO-BASED COLLABORATIONS

When a group or a team is planning to embark on aproject using video as an anchor [33] for their collabora-

tive activity, they need to have in mind the target out-comes of their work, and as we have seen, such envi-sioned outcomes may involve design, synthesis/pattern-finding, evaluation, and analysis/interpretation. Nowthat we have a handle onto not only actual design pat-terns from uses of the DIVER video platform, but a three-dimensional heuristic matrix for generating possible de-sign patterns which vary in terms of target outcome, rela-tionship to video sources (insider/outsider), and dis-course style (formal/informal), we can visit the questionof what forms of supportive structures might serve asuseful design scaffolding for the activities of video-basedcollaborative groups.1. Designs for Target Outcomes

We can envision a number of support structures forgroups whose objective is to produce one of the four out-comes weve identified in the matrix. For groups that areaiming at design, it would be useful to have platform ca-pabilities that supported the processes of iteration andrevision. Videos that capture product development atdifferent stages, for example, could be tagged as such in a

video database. Support for design argumentation [34]could be integrated, linking to video evidence. A collabo-ration platform for supporting design could also includefeatures for tracking progress, such as milestone comple-tion markers that are anchored by video of a successfuluser-test. Additionally, video could be accompanied by adesign canvas that allows users to make free-formsketches or models that are shared with the group. Ratherthan the worksheet in DIVER, for example, users couldhave a sketch space where they could manipulate imagescaptured on the video to convey new ideas.

For synthesis and pattern-finding objectives, there is aneed for tools that aid in collecting and building relation-

ships between instances of a focal interest. With rapidadvances underway in computer vision technology [35], itis becoming feasible that recognition algorithms could beused to automatically identify key objects, faces, headposes, and events in video (e.g., hand-raising in a class-room discourse; uses of math manipulatives; etc), andgenerate tagging metadata for videorecords. Modules fora video research platform that could reliably flag suchmoments of analytic content automatically and makethem readily available for subsequent analyses wouldsave immense time and effort.

Evaluation outcomes could be aided with tools that mi-nimize redundancy and make provision of feedback more

efficient. Rather than free-form text input, it may be ap-propriate to provide a template for a rating system orchecklist schemas. To avoid the recurrence of the samecomment, users could be enabled with a way to showsupport for an existing comment or poll functionalityabutton that says something to the effect of I agree withthat. Another need that we have observed in groupsdoing evaluation is the integration of documents and other materials associated with a performance. Someone be-ing evaluated as part of the teacher credentialing process,for example, may be asked to submit lesson plans andexamples of student work in addition to video of theirlesson. A platform that allows evaluators to dive into and


11/14


Fig. 5. A representation of possible design patterns based on 3 dimensions of collaborative activity: target outcome, discourse

style, and relationship to the source material. Cells with black print indicate collaboration patterns that we have observed onthe DIVER platform. Cells with grey print describe hypothetical scenarios that have not yet been observed but are plausible

given the appropriate user group and software environment.


12/14


annotate these supplemental documents and connectthem to the video event could provide for a more com-prehensive workflow for assessment processes.

In analysis activities, participants are typically lookingfor ways to contribute the most insightful information inthe most efficient manner possible. While text can be agood way of sharing interpretations, there are some sce-narios where voice annotations would be more effectiveat conveying a nuanced perspective [18]. There are alsosome situations where an analysis is best supported withan existing document or representation. In this case,having the ability to simply link a video event to a URLor to embed an object in the analysis would be advanta-geous. Finally, some analyses require one to look at anevent from multiple vantage points. Having the abilityto easily synchronize multiple-video sources and viewsimultaneous playback is a feature requested by a num-ber of our user groups. Researchers of human-computerinteraction, for example, may want to study how some-one uses a new car prototype using video streams of thefront-view, the rear-view, and the drivers face and

posture.2. Designs for Relationships to Video Sources

Both insider and outsider relationships with sourcevideo would likely benefit from different metadata ca-pabilities. For insiders who have shot their own videoand wish to share it with a small group or broadcast it tothe whole world, it is important to be able to control theinformation, or metadata, that people have about thatvideo. Where was the video recorded? Who is featuredin the video? For what purpose was it recorded? This isalso a potential opportunity to specify how the video canand should be used (e.g., granting Creative Commons li-censure). This is true not only for the video itself, but for

the analysis that one performs on the video. These ana-lyses can themselves be works of intellect and artisticexpression, and so attribution of this work is important.

For outsiders, it would be highly valuable to haveaccess to any metadata that was provided for video agroup is using. Geographic data about the video, forexample, could facilitate mashups of video and mappingtools such as Google Earth. If a group was working witha large archive of video, it could be useful to have a plat-form that uses timestamp data to organize and createvisualizations based on when the video was recorded.3. Designs for Discourse Style

If the desired type of formal discourse for a certain

group is known in advance, it would be possible for avideo collaboration platform to structure this type ofdiscourse using templates or other interface constraintsto script activities [36]. Some forms of collaborationhave explicit rules for how these activities should beconducted, and there is potential for the interface to as-sist in regulating this activity as a form of distributedintelligence [37], [38]. Besides templates, this effect canbe achieved by assigning participants different roleswith associated permissions and capabilities, or therecan be features that support divisions of labor by direct-ing participants to work on different parts of the videotask.

Groups that desire a more informal or conversationaldiscourse style will likely desire fewer constraints andmore flexibility for communicating and constructingnew insights. This may include features that allow formore social interaction such as chat or networking capa-bilities (e.g., you may want to connect with someone ifyou know that they have extensive experience doing acertain type of video analysis). It may also include toolsfor doing more free-form interpretive work, such asbuild-as-you-go coding schemes. These types of capa-bilities may allow for the emergence of novel designsand interpretations that would not develop in a moreconstrained setting.

6 CONCLUSION

In this paper we have articulated a vision for the ge-nerative promise of video research and education plat-forms for supporting the work practices of collaborativegroups. In particular, the DIVER video platform embo-dies a new kind of communication infrastructure for

video conversations by providing persistent and searcha-ble records of video pointing activities by participants tospecific time and space moments in video so as to devel-op common ground in technology-mediated conversa-tions among distributed teams. Rather than speculateabout the opportunities for using the DIVER platform,we empirically examined how roughly 200 globally dis-tributed groups appropriated the DIVER technology toserve their needs. We found five dominant collaborativedesign patterns: collective interpretation, distributed de-sign, performance feedback, distributed data coding,and video-based prompting. Abstracting from the fea-tures of these collaboration patterns, we were able to

identity a set of three dimensions that we argue providesa tri-partite design space for video collaboration groups:discourse style (formal/informal), relationship to videosource (insider/outsider) and target outcome (design, syn-thesis, evaluation, and analysis). These dimensions wereused heuristically to articulate a three-dimensional ma-trix for conceptualizing video collaborative groups, andthen used to spawn concepts for dimension couplingsunrealized in DIVER collaborative groups to date, and torecommend new socio-technical designs for better serv-ing the design space of these groups. We invite the learn-ing technologies community to refine and advance ourconceptualizations of collaboration design patterns forvideo platform uses in research and education, and thecreation of systems that support the important needs forvideo conversations in the work practices of educatorsand researchers.

AcknowledgmentsSpecial thanks to DIVER software engineer Joe Rosen forhis exceptional contributions to every aspect of the DI-VER Project since 2002, and Michael Mills, KennethDauber, and Eric Hoffert for early DIVER design innova-tions. DIVER, WebDIVER, and Guided Noticingare trademarks of Stanford University for DIVER soft-ware and services with patents awarded and pending.


13/14


We are grateful for support for The DIVER Project ingrants from the National Science Foundation (#0216334,#0234456, #0326497, #0354453) and the Hewlett Founda-tion.

REFERENCES

[1] P. Saettler, The Evolution of American Educational Technology. Charlotte,NC: Information Age Publishing, 2004.

[2] Goldman, R., Pea, R. D., Barron, B. & Derry, S. (2007). (Eds.). Video Re-search in the Learning Sciences. Mahwah, NJ: Lawrence Erlbaum Associates.

[3] T. Chambel, C. Zahn, and M. Finke, "Hypervideo and Cognition: DesigningVideo-based Hypermedia for Individual Learning and Collaborative Know-

ledge Building," In Alkhalifa, E.M. (Ed.), Cognitively Informed Systems: Uti-lizing Practical Approaches to Enrich Information Presentation and Transfer.Hershey, PA: Idea Group, pp. 26-49, 2006.

[4] R. Goldman, "Video Perspectivity Meets Wild and Crazy Teens: A Design

Ethnography." Cambridge Journal of Education, vol. 34, no. 2, pp. 157-178,June 2004.

[5] D. Schwartz and K. Hartman, "It's Not Television Anymore: Designing Digi-

tal Video for Learning and Assessment." In R. Goldman, R. Pea, B. Barron &

S. J. Derry (Ed.), Video Research in the Learning Sciences. Mahwah, NJ: Erl-baum, pp. 335-348, 2007.

[6] C. Zahn, R. Pea, F. W. Hesse, M. Mills, M. Finke and J. Rosen, J., AdvancedDigital Video Technologies to Support Collaborative Learning in School

Education and Beyond. In T. Koschmann, D. Suthers, & T-W. Chan (Eds.),Computer Supported Collaborative Learning 2005: The ext 10 Years.Mahwah, NJ: Erlbaum, pp. 737-742, June 2005.

[7] R. Pea, and E. Hoffert "Video Workflow in the Learning Sciences: Prospects

of Emerging Technologies for Augmenting Work Practices," In R. Goldman,

R. Pea, B. Barron & S. J. Derry (Ed.), Video Research in the LearningSciences. Mahwah, NJ: Erlbaum, pp. 427-460, 2007.

[8] B. MacWhinney, "A Transcript-Video Database for Collaborative Commen-

tary in the Learning Sciences." In R. Goldman, R. Pea, B. Barron & S. J. Der-

ry (Ed.), Video Research in the Learning Sciences. Mahwah, NJ: Erlbaum, pp.537-546, 2007.

[9] R. Goldman, "Orion, an Online Collaborative Digital Video Data Analysis

Tool: Changing Our Perspectives as an Interpretive Community." In R.Goldman, R. Pea, B. Barron & S. J. Derry (Ed.), Video Research in the Learn-

ing Sciences. Mahwah, NJ: Erlbaum, pp. 507-520, 2007.[10] R. M. Baecker, D. Fono, and P. Wolf, "Towards a Video Collaboratory." In R.

Goldman, R. Pea, B. Barron & S. J. Derry (Ed.), Video Research in the Learn-

ing Sciences. Mahwah, NJ: Erlbaum, pp. 461-478, 2007.[11] H. Jenkins, K. Clinton K., R. Purushotma, A.J. Robinson and M. Weigel,

Confronting the Challenges of Participatory Culture: Media Education for the21st Century. Chicago, IL: The MacArthur Foundation, 2006. Retrieved No-vember 30, 2008, from http://tinyurl.com/y7w3c9

[12] J.D. Bransford, A.L. Brown and R.R. Cocking (Eds.),How People Learn:Brain, Mind, Experience, and School (Expanded Edition). Washington, DC:National Academy Press, 2000.

[13] J. Bransford, N. Vye, R. Stevens., P. Kuhl, D. Schwartz, P. Bell, A. Meltzoff,

B. Barron, R. Pea, B. Reeves, J. Roschelle & N. Sabelli, Learning Theories

and Education: Toward a Decade of Synergy. In P. Alexander & P. Winne

(Eds.),Handbook of Educational Psychology, 2nd edition. Mahwah, NJ: Erl-baum, pp. 209-244, 2006.

[14] B. Jordan and A. Henderson, Interaction Analysis: Foundations and Prac-tice,J. Learning Sci., vol. 4, no. 1, pp. 39-103, Jan. 1995.

[15] S. Derry, R. Pea, B. Barron, R.A. Engle, F. Erickson, R. Goldman, R. Hall, J.

Lemke, M.G. Sherin, B.L. Sherin and T. Koschmann, "Guidelines for con-ducting video research in the learning sciences,"J. Learning Sci., 2009 (inpress).

[16] R. Pea, Video-as-data and Digital Video Manipulation Techniques for Trans-forming Learning Sciences Research, Education, and other Cultural Practic-

es. In J. Weiss, J. Nolan, J. Hunsinger, & P. Trifonas, (Eds.), The Internation-al Handbook of Virtual Learning Environments. Heidelberg: Springer, pp.1321-1393, 2006.

[17] S. E. Page, The Difference: How the Power of Diversity Creates BetterGroups, Firms, Schools, and Societies. Princeton: Princeton University Press,2007.

[18] R. Stevens, Capturing ideas in digital things: A new twist on the old problem

of inert knowledge. In R. Goldman, R. Pea, B. Barron, & S. J. Derry (Eds.),

Video research in thelearning sciences. Mahwah, NJ: Erlbaum, pp. 547-563,2007.

[19] R. Pea, M. Mills, J. Rosen, K. Dauber, W. Effelsberg and E. Hoffert, The

DIVERTM Project: Interactive Digital Video Repurposing,IEEE Multimedia,vol. 11, pp. 54-61, 2004.

[20] R. Pea, R. Lindgren and J. Rosen, Cognitive Technologies for Establishing,

Sharing and Comparing Perspectives on Video Over Computer Networks,

Social Science Information, vol. 47, no. 3, pp. 355-372, Sep. 2008.

[21] M. Lampert and J. Hawkins,ew Technologies for the Study of Teaching.Report to the National Science Foundation from a workshop held June 911,1998, NSF Grant #REC- 9802685. Ann Arbor, MI, June 1998.

[22] B. MacWhinney and C. Snow, Multimodal Studies of Instructional Discourse. Report to the ational Science Foundation, 1999. Available at:http://www.nsf.gov/sbe/tcw/events 991203w/.

[23] R. D. Pea and K. Hay, CILT Workshop on Digital Video Inquiry in Learningand Education, ovember 2526, 2002 (Report to the National Science

Foundation based on NSF #0124012). Palo Alto, CA: Stanford Center for In-

novations in Learning, 2003.[24] H.H. Clark, Using Language. New York: Cambridge University Press, 1996.[25] W.V.O. Quine, The Inscrutability of Reference. In D. Steinberg and L.

Jacobovits (Eds.), Semantics: An Interdisciplinary Reader. Cambridge: Cam-bridge University Press, pp. 142-154, 1971.

[26] C. DiGiano, L. Yarnall, C. Patton, J. Roschelle, D. Tatar and M. Manley,

Conceptual Tools for Planning for the Wireless Classroom,J. Computer As-sisted Learning, vol. 19, no. 3, pp. 284-297, Sep. 2003.

[27] T. White, Code Talk: Student Discourse and Participation with Networked

Handhelds, Computer-Supported Collaborative Learning, vol. 1, no. 3, pp.359-382, Aug. 2006.

[28] S. Wilson, The Use of Ethnographic Techniques in Educational Research,Review of Educational Research, vol. 47, no. 2, pp. 245-265, Spring 1977.

[29] R. Goldman-Segall,Points of Viewing Childrens Thinking: A Digital Ethno-graphers Journey. Mahwah, NJ: Erlbaum, 1998.

[30] J. Buur and K. Bagger, Replacing Usability Testing with User Dialogue,Communications of the ACM, vol. 42, no. 5, pp. 63-66, May 1999.

[31] A.L. Back, R.M. Arnold, W.F. Baile, J.A. Tulsky, G. Barley, R.D. Pea and

K.A. Fryer-Edwards, "Faculty development to change the paradigm of com-munication skills teaching in oncology,"Journal of Clinical Oncology, The

Art of Oncology, March 2009, to appear.[32] J.W. Stigler, R. Gallimore, and J. Hiebert, Using Video Surveys to Compare

Classrooms and Teaching Across Cultures: Examples and Lessons From the

TIMSS Video Studies,Educational Psychologist, vol. 35, no. 2, pp. 87-100,June 2000.

[33] J.D. Bransford, R.D. Sherwood, T.S. Hasselbring, C.K. Kinzer and S.M.

Williams, "Anchored instruction: Why we need it and how technology can

help." In D. Nix & R. Sprio (Eds), Cognition, education and multimedia.

Hillsdale, NJ: Erlbaum Associates, pp. 163-205, 1990.[34] S. Buckingham Shum, "Design Argumentation as Design Rationale," The

Encyclopedia of Computer Science and Technology, Vol. 35, Supp. 20, pp.95-128, Marcel Dekker Inc: NY, 1996.

[35] D. G. Lowe. "Distinctive image features from scale-invariant keypoints,"International Journal of Computer Vision, vol. 60, no. 2, pp. 1573-1405, 2004.

[36] P. Dillenbourg and P. Tchounikine, "Flexibility in macro-scripts for computer-

supported collaborative learning,"J. Comp. Assisted Learning, vol. 23, no. 1,pp. 1-13, 2007.

[37] R. D. Pea, "Practices of distributed intelligence and designs for education." In

G. Salomon (Ed.).Distributed cognitions.New York: Cambridge University

Press. pp. 47-87, 1993.[38] J. Hollan, E. Hutchins and D. Kirsch, "Distributed Cognition: Toward A New

Foundation for Human-Computer Interaction Research,"ACM Transactionson Computer-Human Interaction (TOCHI), Volume 7, Issue 2, 174-196,2000.

Roy Pea is Professor of the Learning Sciences at Stanford Univer-sity and H-STAR Institute co-director (http://hstar.stanford.edu). Hehas published extensively on learning and education fostered byadvanced technologies, including scientific visualization, onlinecommunities, digital video collaboratories and mobile learning. Heis Co-PI of the LIFE Center (http://life-slc.org), funded by the Na-tional Science Foundation as one of several national Science ofLearning Centers, and was co-author of the 2000 National Acade-my Press volume How People Learn. He is a Fellow of the NationalAcademy of Education, the American Psychological Society, andthe American Educational Research Association.

Robb Lindgren received a B.S. in Computer Science from North-western University in 2000 and is currently a doctoral candidate in


14/14


Learning Sciences and Technology Design at Stanford University.He is currently researching learning and interactive media technol-ogies at the LIFE Center while completing his dissertation on pers-pective-based learning in virtual worlds.

sociolinguistic reader

Documents