1 Computer-Supported Cooperative Work (CSCW)

2 Outline Review of Rapid Prototyping Review of Rapid Prototyping Definitions of CSCW & group work Definitions of CSCW & group work Implementation issues Implementation issues Success/Failures Success/Failures Media Media

3 Review of Rapid Prototyping UI tools good for testing more developed UI ideas UI tools good for testing more developed UI ideas Two styles of tools ? Two styles of tools ? “Prototyping” vs. UI builders“Prototyping” vs. UI builders what is the difference?what is the difference? Both types generally ignore the “insides” of application -> this is research Both types generally ignore the “insides” of application -> this is research

4Collaboration Current work environments Current work environments several people working on personal workstationsseveral people working on personal workstations Frequently people need to cooperate Frequently people need to cooperate create/modify documents, drawings, designscreate/modify documents, drawings, designs Two key ways Two key ways at different times (asynchronously)at different times (asynchronously) see changes previous workers have made see changes previous workers have made simultaneously (synchronously)simultaneously (synchronously) actions taken by user must be seen immediately actions taken by user must be seen immediately

5 Computer-Supported Cooperative Work (CSCW) Def.: “the study of how people work together using computer technology” Def.: “the study of how people work together using computer technology” Examples of systems: Examples of systems: shared databases/hypertextshared databases/hypertext video conferencingvideo conferencing chat systemschat systems real-time shared applicationsreal-time shared applications collaborative writing, drawing, games collaborative writing, drawing, games

6 Groupware Groupware denotes the technology that people use to work together Groupware denotes the technology that people use to work together “systems that support groups of people engaged in a common task (or goal) and that provide an interface to a shared environment.”“systems that support groups of people engaged in a common task (or goal) and that provide an interface to a shared environment.” CSCW studies the use groupware CSCW studies the use groupware “CSCW is the study of the tools and techniques of groupware as well as their psychological, social, and organizational effects.”“CSCW is the study of the tools and techniques of groupware as well as their psychological, social, and organizational effects.”

7 Types of Cooperation Focused partnerships Focused partnerships users who need each other to complete a taskusers who need each other to complete a task often a document or image to work on often a document or image to work on e.g., joint authors of a paper e.g., joint authors of a paper Lecture or demo Lecture or demo person shares info. with users at remote sitesperson shares info. with users at remote sites questions may be asked questions may be asked may wish to keep history and be able to replay may wish to keep history and be able to replay

8 Types of Cooperation (cont.) Conference Conference group participation distributed in spacegroup participation distributed in space at same time or spread out over time at same time or spread out over time Structured work process Structured work process a set of people w/ distinct roles solve taska set of people w/ distinct roles solve task e.g., hiring committee accepts applications, reviews, invites top for interviews, chooses, informs e.g., hiring committee accepts applications, reviews, invites top for interviews, chooses, informs aka “work flow” or “task flow”aka “work flow” or “task flow”

9 Types of Cooperation (cont.) Meeting and decision support Meeting and decision support meeting w/ each user working at a computermeeting w/ each user working at a computer e.g., PDA Brainstorming tool e.g., PDA Brainstorming tool Tele-democracy Tele-democracy online town hall meetingsonline town hall meetings

10 Groupware Taxonomy AsynchronousSynchronous Same placeproject scheduling,games, in/out boardclassrooms, ATC Distributed , netnews,chat rooms, writing video conf., netmeeting RCS, CAD,

11 Key Issues Group awareness Group awareness Multi-user interfaces Multi-user interfaces hard to design/conduct controlled experimentshard to design/conduct controlled experiments Concurrency control Concurrency control consistency and reconciliationconsistency and reconciliation Communication & coordination Communication & coordination can’t see each other -> lose visual cuescan’t see each other -> lose visual cues floor controlfloor control

12 Key Issues (cont.) Latency Latency e.g., user points at an object and talke.g., user points at an object and talk Security and privacy Security and privacy more... more...

13 Asynchronous Implementation Issues Each user may have own copy of data Each user may have own copy of data Must integrate changes at some point Must integrate changes at some point example: programmers working on sourceexample: programmers working on source Problems when conflicts between changes Problems when conflicts between changes lock portions of worklock portions of work keeps state well defined, although doesn’t stop semantically incompatible changes keeps state well defined, although doesn’t stop semantically incompatible changes resolve conflicts via integration mechanismresolve conflicts via integration mechanism

14 Synchronous Implementation Issues >=Two users working on same data, at the same time, in cooperation >=Two users working on same data, at the same time, in cooperation Extend Model View Controller (MVC) Extend Model View Controller (MVC) views & copies of the model are distributedviews & copies of the model are distributed Propagate command history Propagate command history must resolve conflicts among N historiesmust resolve conflicts among N histories at what level are commands?at what level are commands? mouse position not good enough (e.g., different font sizes, etc.) mouse position not good enough (e.g., different font sizes, etc.)

15 Social Issues Can these technologies replace human-human interaction? Can these technologies replace human-human interaction? can you send a “handshake” or a “hug”can you send a “handshake” or a “hug” how does intimacy survive?how does intimacy survive? Are too many social cues lost? Are too many social cues lost? facial expressions and body language for enthusiasm, disinterest, angerfacial expressions and body language for enthusiasm, disinterest, anger will new cues develop? e.g., :)will new cues develop? e.g., :)

16 Groupware Successes ubiquitous (your grandparents have it?)ubiquitous (your grandparents have it?) Newsgroups and mailing lists Newsgroups and mailing lists Videoconferencing Videoconferencing growing slowly but steadilygrowing slowly but steadily

17 Groupware Successes (cont.) Lotus Notes Lotus Notes integrates , newsgroups, call tracking, status, DB searching, document sharing, & schedulingintegrates , newsgroups, call tracking, status, DB searching, document sharing, & scheduling very successful in corporationsvery successful in corporations will the Web erode? Notes is more structuredwill the Web erode? Notes is more structured

18 Groupware Failures Shared calendars Shared calendars making a come back? web-based?making a come back? web-based? Why does groupware fail? (Grudin) Why does groupware fail? (Grudin) disparity between workers & beneficiariesdisparity between workers & beneficiaries threats to existing power structuresthreats to existing power structures insufficient critical mass (Web reduces)insufficient critical mass (Web reduces) violation of social taboosviolation of social taboos rigidity that counters common practice or exceptionsrigidity that counters common practice or exceptions

19 Success/Failure of Groupware Depends on competing alternatives Depends on competing alternatives collaborators down the hall or across country?collaborators down the hall or across country? If users are committed to system, etiquette & conventions will evolve If users are committed to system, etiquette & conventions will evolve tend to arise from cultural & task backgroundtend to arise from cultural & task background users from different orgs or cultural contexts may clashusers from different orgs or cultural contexts may clash Synchronous systems that work well for 2 users may be less effective w/ more users Synchronous systems that work well for 2 users may be less effective w/ more users

20 Media Video: Rich, but problems with gaze, gesture, non-verbal communication. Video: Rich, but problems with gaze, gesture, non-verbal communication. Audio: Conveys meaning well but not necessarily location Audio: Conveys meaning well but not necessarily location Text: Good for synchronous or asynchronous communication Text: Good for synchronous or asynchronous communication Ink: Good for expressing ideas and brain-storming Ink: Good for expressing ideas and brain-storming

21 Video Eye contact problems: Eye contact problems: Offset from camera to screenOffset from camera to screen “Mona Lisa” effect“Mona Lisa” effect Gesture has similar problems: trying pointing at something Gesture has similar problems: trying pointing at something

22 Audio Good for one-on-one communication Good for one-on-one communication Bad for meetings. Spatial localization is normally lost. Can be put back but tricky. Bad for meetings. Spatial localization is normally lost. Can be put back but tricky.

23 Turn-taking, back-channeling In a face-to-face meeting, people do a lot of self-management In a face-to-face meeting, people do a lot of self-management Preparing to speak: lean forward, clear throat, shuffle paper Preparing to speak: lean forward, clear throat, shuffle paper Unfortunately, these are subtle gestures which don’t pass well through today’s technology Unfortunately, these are subtle gestures which don’t pass well through today’s technology Network delays make things much worse Network delays make things much worse

24 Breakdowns Misunderstandings, talking over each other, losing the thread of the meeting Misunderstandings, talking over each other, losing the thread of the meeting People are good at recognizing these and recovering from them “repair” People are good at recognizing these and recovering from them “repair” Mediated communication often makes it harder Mediated communication often makes it harder E.g. often escalates simple misunderstandings into flaming sessions E.g. often escalates simple misunderstandings into flaming sessions

25 Usage issues Communication in the real world has both structured & unplanned episodes Communication in the real world has both structured & unplanned episodes meeting by the Xerox machinemeeting by the Xerox machine Much face-to-face communication is really side-by-side, w/ some artifact as focus Much face-to-face communication is really side-by-side, w/ some artifact as focus

26 Solutions Sharing experiences is very important for mutual understanding in team work Sharing experiences is very important for mutual understanding in team work Context-based displays (portholes) work well Context-based displays (portholes) work well Video shows rooms & hallways, not just people or seats Video shows rooms & hallways, not just people or seats

27 Solutions Props (mobile presences) address most of these issues. They even support exploration. Props (mobile presences) address most of these issues. They even support exploration.

28 Solutions Ishii’s Clearboard: sketching + presence Ishii’s Clearboard: sketching + presence

29 Face-to-Face: the ultimate? It depends It depends Conveys the maximum amount of information, mere presence effects are strong. But… Conveys the maximum amount of information, mere presence effects are strong. But… People spend a lot of cognitive effort managing perceptions of each other People spend a lot of cognitive effort managing perceptions of each other In a simple comparison of F2F, phone and , most subjects felt most comfortable with the phone for routine business contact In a simple comparison of F2F, phone and , most subjects felt most comfortable with the phone for routine business contact

30 Face-to-Face: the ultimate? Kiesler and Sproull found -only programming teams were more productive than +F2F teams in a CS course Kiesler and Sproull found -only programming teams were more productive than +F2F teams in a CS course There you want coordination, commitment, recording There you want coordination, commitment, recording Conclusion: Match the medium to the mission Conclusion: Match the medium to the mission

31 CSCL: Computer-Supported Collaborative Learning Sub-area of CSCW concerned with learning & collaboration Sub-area of CSCW concerned with learning & collaboration Peer interaction is a powerful source of learning, especially in universities Peer interaction is a powerful source of learning, especially in universities Three powerful models: Three powerful models: TVI, DTVI: recorded instructor, team reviewTVI, DTVI: recorded instructor, team review Peer instruction: pauses for group discussionPeer instruction: pauses for group discussion PBL: Problem-based learning, team problem-solvingPBL: Problem-based learning, team problem-solving

32 Livenotes Designed to include other learners perspectives into note-taking Designed to include other learners perspectives into note-taking

33 Review CSCW vs. groupware CSCW vs. groupware Taxonomy based on space and time Taxonomy based on space and time Key issues Key issues awareness, multi-user UIs, concurrency, communication & coordination, latencyawareness, multi-user UIs, concurrency, communication & coordination, latency Implementation and social issues Implementation and social issues extend MVCextend MVC are social cues lost?are social cues lost? Successes ( ) & failures (scheduling) Successes ( ) & failures (scheduling)