Community Awareness And Visualization: Shi Shi

Community Awareness and Visualization Shi Shi

1

‘Social’ Systems: Designing Digital Systems that Support Social Intelligence THOMAS ERICKSON  Social Computing Group IBM T. J. Watson Research Center 



2

Social Intelligence • What is it? The ways in which groups of people manage to produce coherent behaviour directed towards individual or collective ends. • Why should we be interested in this sort of social intelligence? - Produce better quality solutions - Produce solutions more quickly 

3

Social proxy A social proxy for a group chat in the Babble system: (a) an active chat; (b) after chat has ceased. a

b

IBM Babble http://socialcomp.com/pro jects/babble/

4

Social proxy IBM Loops

http://socialcomp.com/projects/loop s/ 5

Lecture proxy

Figure 4. Three instances of the lecture proxy: (a) the norm; (b) an audience member interrupting; and (c) many audience members speaking, which violates the norm.

The lecture proxy can serve as an aid in either enforcing a return to the norm, or signaling the group that perhaps it is time to shift to a different mode of interaction. 6

Some concept pieces – The conference call proxy

Three instances of a conference call proxy: (a) going around the table (b) asking questions, and (c) signaling that there are difficulties hearing the speaker

7

The Auction Social Proxy

a

b

c Three instances of the auction proxy: (a) people viewing information about to-be-auctioned item; (b) two bidders; (c) the end game with many bidders. 8

A Task Proxy for Organizationwide Activities

A n in d ivid u a l

*

N o sta te h a s b e e n e n te re d

T h e u se r w h o is vie w in g th e ta sk

In p ro g re ss

Group’s manager

co m p le te d

9

A Task Proxy for Organizationwide Activities

A task proxy for a third line organization of about 140 people divided into four departments and fourteen work groups. 10

A Task Proxy for Organizationwide Activities • The task proxy as envisioned here enables two sorts of things: First, it permits the overall status of a

task to be visualized. This permits either centralized or decentralized management of the task. Second, the task proxy provides a contextualized means of communication that is tied to the task and its state 11

Plot-polling: Collaborative Knowledge Visualization for Online Discussions Alex Ivanov, Thomas Erickson, Dianne Cyr School of Interactive Arts and Technology, Simon Fraser University,  IBM T. J. Watson Research Center,  Faculty of Business, Simon Fraser University  [email protected], [email protected], [email protected] 



12

Design of the plot-poll

http://www.sjwinfo.org/ 13

Design of the plot-poll

14

Design of the plot-poll

http://www.sfu.ca/~aivanov/demo.htm

15

Design of the plot-poll

Once a dozen or so users had contributed, successive clicks by other users would have an increasingly small effect on the size of bubble, thus making the act of participation at best unsatisfying, and at worst invisible.

16

Design of the plot-poll

Latest version of plot-poll http://www.venlafacts.org/

17

Other format for poll

Color represents much better/better/same/worse… Bar thickness represents frequency variable But, but it relies entirely on colors to show a pattern—a problem for those who are color-blind or have monitors with different calibrations.

18

Studying Cooperation and Conflict between Authors with history flow Visualizations Fernanda B. Viégas  MIT Media Lab  Cambridge, MA 02139 USA  [email protected] .edu 

M a rtin W a tte n b e rg IB M R e se a rch C a m b rid g e , M A 02142 U SA m w a tte n @ u s. ib m . c om

K u sh a lD a ve IB M R e se a rch C a m b rid g e , M A 02142 U SA kd a ve @ u s. ib m . co m

19

Wikipedia On Wikipedia, content can be added or changed at any time by anyone on the Internet. To many, this approach—so vulnerable to mistakes, ignorance and malice—seems a flatly ridiculous way of producing a serious reference tool. The mystery of Wikipedia is that despite the obvious potential drawbacks of its openness, it has enjoyed significant success. 

20

history flow

21

The technique in history flow

22

history flow Text panel

Contribute d authors

Visualizat ion 23

PATTERNS OF COOPERATION AND CONFLICT • They used the history flow method to examine in detail more than 70 different Wikipedia page histories. Their examination revealed several common patterns of collaboration and negotiation. 

24

Pattern 1 Vandalism and repair • Mass deletions – one common form of vandalism : deletion of all contents on a page

•

history flow for “Abortion” page, versions equally spaced.

25

Pattern 1 Vandalism and repair • 2. Offensive copy: insertion of vulgarities or slurs. • 3. Phony copy: insertion of text unrelated to the page topic. e.g. Chemistry page -- “Windows 98 readme” file. • 4. Phony redirection: Often pages contain only a redirect link to a more precise term (e.g. “IBM” redirects to “International Business Machines.”), but redirects can also be malicious. • 5. Idiosyncratic copy: adding text that is related to the topic of the page but which is clearly one-sided, not of general interest, or inflammatory; these may be long pieces of text. e.g. cat – unix cat command

26

Pattern 2 Negotiation • zigzag pattern of an edit war Zigzag patter n

“Chocolate” page spaced out by number of versions; we can see the zigzag pattern of an edit war.

27

Pattern 3 Anonymous contributions

“Brazil” page showing abrupt growth and few anonymous contributions. 28

Pattern 4 First mover advantage • The initial text of a page tends to survive longer and tends to suffer fewer modifications than later contributions to the same page. Our hypothesis is that the first person to create a page generally sets the tone of the article on that page and, therefore, their text usually has the highest survival rate. 29

Pattern 5 deletion and insertion more frequently than moving • One explanation may be that the editing window of Wikipedia pages is by default 25 lines long, making it hard for one to see long articles in their entirety. Moving text

history flow for “Abortion” page

30

Using Community Visualization to Stimulate Participation in Online Communities Ju lita V a ssile va 1, Lingling Sun 2 1Computer Science Department, University of Saskatchewan, Canada 2Solutions AB TELUS Business Transformation, Edmonton, Canada. [email protected], [email protected]

31

P2P PAPER-SHARING COMMUNITY

Motivational visualization - fixed version (user names and IP address are blacked out to preserve the privacy of the participants). 32

Customizable visualization for a class-support community

The dynamic design of the visualization (user names blacked out). 33

Visualizing Reciprocity in an Online Community to Motivate Participation Kadhambari Raghavun and Dr.Julita Vassileva

Online communityWISETales

Visualization of Reciprocity in WISETales

Demo

http://homepage.usask.ca/~kas411/visual



Six Claims about Visualization Design • 1. Everyone sees the same thing; no user customization - Share the knowledge - Support people being held accountable for their actions - Leads to useful social phenomena( feelings of obligation, peer pressure) 

39

Six Claims about Visualization Design • 2. Portray actions, not interpretation - minimizing the amount of interpretation - user understand the context better

40

Six Claims about Visualization Design • 3. Social proxies should allow deception - Like face to face interaction, some

deception is important(politeness, vital social skills)

41

Six Claims about Visualization Design • 4. Support micro/macro readings - Small persist components - Accrete into recognizable patterns at muliple levels - Fine structures

42

Six Claims about Visualization Design • 5. Ambiguity is useful: suggest rather than inform - Provide grist for inferences - Our users are comfortable with making guesses - e.g It is much more important for users to be able to tell whether there are 3 or 7 people present, than whether there are 103 and or 107 present.

• 43

Six Claims about Visualization Design • 6. Use a third-person point of view - Show users own activity as others would see it

44

More interesting visualizations

B aby N am eM apper h ttp :// n a m e m a p p e r. b a b yn a m e w i za rd . co m / n a m e m a p p e r/

B a b y N a m e Vo ya g e r h ttp :// n a m e m a p p e r. b a b yn a m e w i za rd . co m / n a m e m a p p e r/

T h in kin g M a ch in e 4 h ttp :// w w w . tu rb u le n ce . o rg / sp o tli g h t/ th in kin g / ch e ss. h tm l

45

Thank you

46