Ekkli - Effective large-scale discussions

talk given at the Online Deliberation Research Group

Transcript

1. Effective large-scale discussions Ekkli Maps Feb, 2012 Sunday, February 26, 2012
2. WHO? Udi h Bauman / @dibau_naum_h Tal Yaron / @talyaron “practicants” Architect & developer @Ekkli Chief-scientist @Ekkli Sunday, February 26, 2012
3. AGENDA? Demo Design Model Project Sunday, February 26, 2012
4. PROJECT Sunday, February 26, 2012
5. MISSION? Enable effective large-scale discussions Before After Sunday, February 26, 2012
6. MISSION? By addressing the problems of: Congestion & Aggregation Before After Sunday, February 26, 2012
7. LARGE SCALE? Moving gradually, but aspiring to the largest scale: global discussions Sunday, February 26, 2012
8. TYPE OF PROJECT? Open-source - Dual model: Free for public discussions, Commercial for private ones $ Sunday, February 26, 2012
9. STRATEGY? Iterations of: talking with users, building a solution for them, measuring how much it works & learning from it Sunday, February 26, 2012
10. MODEL Sunday, February 26, 2012
11. DISCUSSION? Bunch of people exchanging text messages Sunday, February 26, 2012
12. DISCUSSION? Bunch of people exchanging text messages, towards achieving some goals Goal 3 Goal 2 Goal 1 Start Sunday, February 26, 2012
13. GOALS? Part of a wider context: the participants have targets they’re trying to achieve Sunday, February 26, 2012
14. CONTEXT? The situation in which the participants are, in respect to the discussion Sunday, February 26, 2012
15. CONTEXT? The discussion can be forward-looking: solving a problem, making a decision, planning Sunday, February 26, 2012
16. CONTEXT? Or, it can be backward-looking: reviewing stuff, analyzing root-causes & reaching conclusions Sunday, February 26, 2012
17. CONTEXT? The discussion can be looking on current state: synchronizing state/progress, sharing gossip/knowledge Sunday, February 26, 2012
18. TEXT MESSAGES? (written or spoken) Sunday, February 26, 2012
19. TEXT MESSAGES? Can be classified into Speech Acts Sunday, February 26, 2012
20. TEXT MESSAGES? In our model, we have 2 types: Stories & Opinions Story Opinion Sunday, February 26, 2012
21. STORIES? A text describing a topic for discussion “Whenever the class models change, they trigger a change event, which is handled by the diagram widget, that redraws the diagram.” Example Sunday, February 26, 2012
22. OPINIONS? Judgements of aspects of stories “This design is good, because it will be easy to maintain.” Example Sunday, February 26, 2012
23. OPINIONS? Can be represented as a combination of speech acts “This design is good, because it will be easy to maintain.” Good Easy Maintain Example Sunday, February 26, 2012
24. STORIES? Cannot be represented as a combination of speech acts “Whenever the class models change, they trigger a change event, which is handled by the diagram widget, that redraws the diagram.” Example Sunday, February 26, 2012
25. TEXT MEDIUM? Because, till now, discussions were done F2F Sunday, February 26, 2012
26. TEXT MEDIUM? Easy to produce, difficult to consume & aggregate Sunday, February 26, 2012
27. TEXT MEDIUM? Does it cut also for online large-scale discussions? Sunday, February 26, 2012
28. DESIGN Sunday, February 26, 2012
29. IF NOT TEXT, WHAT? Need something x1000 better to solve the congestion = ? Sunday, February 26, 2012
30. PICTURES! Automatically convert text to visual representation Story diagrams Opinion icons Sunday, February 26, 2012
31. VISUAL DISCUSSION? Interactive, zoomable picture Sunday, February 26, 2012
32. VISUAL DISCUSSION? Consists of 3 layers Context Stories Opinions Sunday, February 26, 2012
33. PDF 2 JPG? I don’t know... Story Picture How do you convert “Whenever the class models change, they trigger a change event, which is handled by the diagram widget, that redraws the diagram.” Sunday, February 26, 2012
34. Starting with a simple use-case: Design PDF 2 JPG? How do you convert Story Picture “Whenever the class models change, they trigger a change event, which is handled by the diagram widget, that redraws the diagram.” Sunday, February 26, 2012
35. Use Machine Learning to “guess” the correct meaning, and let users manually fix the mistakes PDF 2 JPG? How do you convert Story Picture “Whenever the class models change, they trigger a change event, which is handled by the diagram widget, that redraws the diagram.” Sunday, February 26, 2012
36. AGGREGATION? Sunday, February 26, 2012
37. AGGREGATION? Sunday, February 26, 2012
38. EFFECTIVENESS? Sunday, February 26, 2012
39. ADDED VALUE? Capturing knowledge
    
    
    
    
    
    
     
              Sunday, February 26, 2012
40. ADDED VALUE? Learning across discussions & even organizations Suggesting past insights Sunday, February 26, 2012
41. ADDED VALUE? Helping with the wider context Sunday, February 26, 2012
42. DEMO Sunday, February 26, 2012
43. INITIAL PROTOTYPE Will be tested in several beta sites Sunday, February 26, 2012
44. INVITATION? udi@ekk.li Sunday, February 26, 2012
45. Q&A Sunday, February 26, 2012