EBoard 12: Prototyping
This class will be recorded! Its use will be limited to members
of the class. Please do not share with others.
Approximate overview
- Preliminaries
- Notes on Investigation 5.
- Prototyping.
- [Break]
- Group formation for Investigation 6.
- Time to work on Investigation 6.
Administrative stuff
General Notes
- Happy Monday! I hope you had a wonderful weekend. Weekly one-word share.
Unproductive. Fun [+7]. Short. Happy. Baseball. Busy [+2].
Tiring [+1]. International Exhausting. Boring. Mothers.
Adventurous. Grading/Grating.
- You should have received email from me earlier today with tentative grades.
- Let me know if you have questions.
- Remember that you can make up any grade you’re not satisfied with.
However, you will need to email me to let me know that you’ve
uploaded something new.
- I’ve cancelled Investigation 7. We’ll do something similar (but lighter)
in class.
- Although the final is listed on the schedule, there is no final.
Ideally, you’ll have everything done by the last day of class.
- We’ll talk a bit about Investigation 5.
UI Rants
Upcoming Activities
- Wednesday at 7:00 p.m. Writers @ Grinnell Distinguished Author Lecture.
https://www.prairielights.com/live/alexander-chee-writers-grinnell-distinguished-author-event
- Thursday at 5:00 p.m.: Advice and guidance for a data science career.
https://grinnellcollege.webex.com/grinnellcollege/j.php?MTID=mdce020bcd5f926a1549020e215edc659
Work for Friday’s class.
- Have prototypes ready to test.
I’ll plan focus-group-style testing for the first half of class.
- Readings:
- Task:
- Find a usability or accessibility checklist that you’d like to use.
Post a link to it and a short comment on why you like it.
Work for Monday’s class. (Our last class.)
- Reading
- Task:
- Apply a usability checklist to a specified site.
Work for Thursday the 20th
Q&A
How will we test a paper prototype during class?
The “user” is going to direct one of the group on what to touch.
Can we do a storyboard or set of photos instead?
Certainly. Any kind of prototype you think is appropriate is fine.
Easy for you, understandable for your audience.
Reminder: In-class will be focus groups. In effect, you become the
person they direct.
Investigation 5
Naming Files
- This is my fault, since I didn’t tell you to give the files particular
names. But …
- Teams downloads using whatever name you’ve given.
- So choosing good file names is a nice UI issues for your instructor.
Aligning files
- It’s nice not to have to rotate images.
Prior art
I may not have been clear. I wrote (or copied)
Those on the front should come from ideas that already exist in
the world, whether available as products, described in research
papers, or imagined in fiction. Consider these your literature
review and prior art search, where you map the space of solutions
that have already been designed for your problem. Provide some
kind of citation when possible.
Number of ideas
Some of you said, “I’ll be graded on 24. I should only submit 24.”
That seems dangerous, particularly since I did ask you to submit 31.
- Some folks submitted redundant ideas.
- Some folks miscounted.
Prototyping
Important points, take one
- Value of low-cost prototypes
- Not too invested
- More willing to try different options. (Maybe take risks.)
- Iterative design has been shown to be better
- Try to identify mistakes early in the process; the later you identify
mistakes, the more expensive it is to fix them.
- Focused feedback - On what is important (big picture) rather than
less important (“I think you have the wrong shade of green.”)
- Value of multiple different prototypes
- Better feedback
- More consideration of options
- Gather diverse strengths and know those stengths
- We seem to be seeing consistent themes across the weeks
- Thinking about roles is a lot like contextual inquiry
- “Deliver what the customer wants, not what the designer wants”
- Iterative design
- Plan ahead
- Different kinds of prototypes
- Role
- Look-and-feel
- Implementation
- Integration
Important points for user testing
- Know your audiences
- Engineers vs. marketers vs. ….
- Your testers
- Your investors
- Your actual users
- Practice.
- Help your subjects think aloud.
- Have fun?
- Strive to understand both successes and failures.
- Take notes / recordings. “It doesn’t matter how good your testing
is if you don’t remember it!”
Are we at the stage in which electronic prototypes match the cost/benefit ratio of paper prototypes? Why or why not?
- Users react differently to electronic and paper prototypes. Using
an electronic prototype may lead people to critique those small
details that paper prototypes let us avoid.
- We can make electronic prototypes that look like paper prototypes
(e.g., using a graph paper background)
- People may be less willing to critique electronic versions.
- Are there generational issues? Gen Millen knows that tech can be
changed. Da Boomers assume that tech is finished. [The return of
audience.]
- For some people, the electronic prototypes are a lot easier to build.
- Drag and drop
- On the sketch assignment, it would have been nice not to draw
so many washers and dryers. Yay copy and paste.
- The ability to copy and paste may encourage you to put in too
much detail.
- Sam notes that he has a lot of 4x6 post-it notes and he’d be happy to
get a bunch to you (somehow).
- Electronic UIs could be more inclusive / accessible.
When testing, how do we ensure that we get useful/accurate feedback?
- Make sure to work with a wide variety of people.
- Lots of testing.
- Rely on important roles (greeter, facilitator, computer, and observers).
- Ensure that the participants know that there are no right or wrong
answers. Build trust.
- Ask the right questions to participants / give them the right tasks.
- If we focus on “the right questions”, will we get confirmation bias?
- Some questions might be “What did you find the most difficult?”
- Be careful about guiding the user too much. (Identify/design the
tasks before designing the UI.)
- Be consistent across different users. “Stick to the script.”
- Think out loud.
- Think carefully about what they are doing and ask followup questions.
Break
I6 Planning
Testing multiple interfaces
As we move from testing one UI (as in the usability tests) to testing multiple options for the same task (as in I6), what are issues we should consider?
- Worry that people may be biased toward the first/last option. Ask for
feedback after each option we are testing.
- In real life: Run them in different orders.
- Don’t have your own preference in advance.
- Avoid presenter bias (e.g., by having one person present all the UIs)
- Simple tasks, probably only one or two.
- Need to be explicit about asking for improvements.
- Make it clear that you’re not just interested in hearing what is
best, but more what they like and don’t like (or find usable or
not usable) about each.
- Going through the sequence may lead to some bias. “The first UI
had a ‘Fire’ menu. Where’s that menu?”
- We’ll learn after we’ve tried.
Focus groups
What other issues are there to consider for focus groups rather than just individuals?
- Have one person do all of the options.
- Find ways to get mulitple perspectives.
- We’ll learn after we’ve tried.
Other details
- How many groups? (Proposal: 4x4 + 2x3) (vs 6x3 + 4x1) (vs 11x2)
- Topics/Problems
- Laundry Rooms
- Mentor Sessions
- Job Hunting
- Bathrooms
- How should we form groups?
- Express preference + random selection
- Completely random selection
- Chat amongst ourselves
- How should we structure Friday’s class?
- 10 min, prelims
- 10 min, group prep
- 25 min, first test panel (half the groups)
- extended break
- 5 min setup
- 25 min, second test panel (switch roles)
- ? min, discuss usability/accessibility heuristics
Group Discussion
Please set up meetings in the Investigation 6 Channel so that Mai
and I can check in.
Topics
- Review the assignment
- Agree upon broad picture of interfaces for this project
- Other logistics (when are you interviewing people, what roles and
will you rotate, collaboration on writeups, timeline)
- ???