HCI Capstone Project 2013

The slides for my MS HCI capstone project.

Transcript

1. Greetings from... Atlanta
2. Yard Genius Flickr user netcfrance
3. The Idea Flickr user rosipaw
4. • Homeowners (primarily) • Limited experience and knowledge • Limited confidence in their skills Who are the users?
5. “I want to have a nice yard; I just don’t want to spend all my time on it. It’s not my first priority.” - P1 (Female, 31) Flickr user gatesofmemphis
6. Current Sources • Home improvement store employees • Internet searches, blogs • Magazines (Southern Living, Martha Stewart) • Gardening books • Local newspaper columns
7. “I planted azaleas, but I put them in too late and they burned up in the sun.” - P2 (Male, 33) Flickr user martinlabar
8. • Determine suitable plants for user location • Automatically identify tasks based on context • Design layout for the yard Tasks
9. Design Flickr user Josh Ward Garden Design
10. Challenge • Skill, Rule, Knowledge behavior framework (Rasmussen, 1983) • Put knowledge in the world (Norman, 2002)
11. Design Space • Context (Dourish, 2004) • Tablet application • Focus on tasks or layout?
12. Design Space • Don’t overwhelm with info and tasks • Consistency to increase learnability • Interaction of layout screens
13. Home Yard Layout/To Do Add Structure Add Plant Get Recommendation Search List Filter Plant Page To Do List Manual Add Mark Complete Shopping List Manual Add Add From Task/ Plant Page Manual Configuration: Location Develop Initial Flow
14. Lots of quick, rough sketches
15. Refine sketches for prototype
16. Prototype Flickr user Funny-Leo
17. Goal • Rich interactions • [T]he main drawback of conventional sketching has to do with its limitations in capturing time, dynamics, phrasing - the temporal things that lie at the heart of experience. (Buxton, 2007) • Test on tablet
18. Mixed Fidelity • [-] Visual • [+] Breadth of functionality • [+] Depth of functionality • [+] Richness of interactivity • [-] Richness of data-model • (McMurdy et al., 2006)
19. Method • Coded in HTML • Used the following libraries • Twitter Bootstrap • jQuery
20. Demo
21. Evaluation Flickr user kevin dooley
22. Method • Four users • Test prototype on iPad • Eleven tasks
23. Tasks 1. Check weather 2. Check to dos 3. Add to do 4. Delete to do 5. Mark to do complete 6. Add planter to yard 7. Find plants by least effort 8. Add plant 9. Are begonias recommended 10.Check shopping list 11. Add item to shopping list
24. Quantitative • Performance • Task completion (binary) • Number of errors • Satisfaction • Systems Usability Scale (Sauro and Lewis, 2012)
25. Qualitative • Observed user behavior • Think out loud protocol
26. Task Completion 0 1 2 3 4 T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11 Number of Users Completing Task Percent of Users Task
27. Errors per Task 0 1 2 3 4 5 6 7 T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11 Errors per Task by Participant Errors P1 P2 P3 P4
28. SUS Participant Score P1 72.5 P2 87.5 P3 35 P4 82.5 68 is average and the top 10% of applications score above 80.3 (Sauro and Lewis, 2012)
29. Issues Issue Severity Principle Task 1 Home page: Not obvious that you can view weather 3 Visibility Task 3 Calendar page: Users clicked on date to add new item to date 2 Conceptual model, Affordances Calendar obscured by modal dialog for adding new to do item 1 Visibility Task 5: Calendar page and home page: Users click only on check mark to mark item complete when whole item is touch target 3 Affordances Task 6: Landscape page: Users attempted to touch and place before attempting to drag and drop 1 Conceptual model, Affordances Task 7: Plant page: Not clear if effort means least amount of work 2 Conceptual model Not clear if sort by most or least effort 2 Visibility Task 9 Plant page: Not clear what recommended heading means 2 Conceptual model Overall: Not clear that clear that tasks are inferred by items added 2 Conceptual model
30. “It had that pinwheel thing that looks like settings. There wasn’t anything else that looked like it might have been weather.” P4 (Female 25)
34. “Clicking didn’t do it, so I dragged... I assumed after I clicked it, it would highlight so I could see which one I selected and then touch on the screen where I want to put it.” P1 (Male 33)
35. “Did I put all these todos in or did you infer them for me?” P2 (Female, 31)
36. Questions?