SIMS 213: User Interface Design

1
SIMS 213 User Interface Design Development

• Marti Hearst
• Tues, Jan 23, 2007

2
Today

• Today
• Coordinate Project Topics and Teams
• Discuss Sullivan Reading
• Next time
• The User-Centered Design Process

3
User Interface TimeLine

• Xerox Alto
• Circa 73
• http//www.digibarn.com/collections/systems/xerox-
  alto/
• Xerox Star
• Circa 81
• http//members.fortunecity.com/pcmuseum/alto.html
• Unofficial Apple timeline
• http//fp3.antelecom.net/gcifu/applemuseum/index.h
  tml
• Apple Lisa
• Lisa 1 and 2, circa 83-85
• http//en.wikipedia.org/wiki/Apple_LisaSoftware
• http//fp3.antelecom.net/gcifu/applemuseum/lisa2.h
  tml
• http//www.digibarn.com/collections/screenshots/Ap
  ple20Lisa/

• Apple Macintosh
• 84-86
• http//www.digibarn.com/collections/screenshots/Ap
  ple20Macintosh20v1-120-201984/
• Apple 2gs
• 86-92
• http//fp3.antelecom.net/gcifu/applemuseum/iigs.ht
  ml
• Mac II
• Circa 87-90
• http//fp3.antelecom.net/gcifu/applemuseum/mac2.ht
  ml
• Windows 3.0
• Released 90
• http//en.wikipedia.org/wiki/ImageWindows_3.11_wo
  rkspace.png
• Windows 95
• Released 95
• http//en.wikipedia.org/wiki/Windows_95

4
Discussion

• Kent Sullivan, The Windows 95 User Interface A
  Case Study in Usability Engineering, CHI 1996.
• A case study of the iterative design process.
• What are some of the important points?
• Only 12 people on the entire Win 95 interface
  design team!

5
Sullivan Case Study

• Interesting points about early users
• Beginning users had problems with
• Finding a non-visible program, double-clicking to
  activate, using hidden menus, finding minimized
  windows, file system hierarchy
• Initially didnt want to introduce too many new
  things
• However, initial studies on 3.1 and the 95
  prototype indicated that the starting point was
  just too poor, and more radical revision was
  needed.
• http//acm.org/sigchi/chi96/proceedings/desbrief/S
  ullivan/kds_txt.htm

6
Sullivan Case Study

• Had an offsite and rethought everything.
• The need to rapidly iterate
• Waterfall design method (compartmentalized UI
  done at the end during quality assurance stage)
  would produce an unusable product.
• Jettison the giant design spec
• Encouraged closer interaction and hallway
  conversations

7
Sullivan Case Study

• Had an offsite and rethought everything.
• Method Adopted
• Find most common tasks, and test on those
• Compare to a baseline (Windows 3.1)
• Use rapid prototyping tools
• Keep testing with users.
• Focus on scaffolding (beginner -gt
  intermediate-expert)
• Give up on too much backwards-compatibility with
  3.1
• Continually inform outsiders as well as team
  members about the design

8
Sullivan Case Study

• Fine tuning with usablity tests
• Holistic interface test in the lab
• 64 phases of lab testing
• Many participants! (560 total)
• Longitudinal field study
• Other testing on components
• No detail of the orginal design for W95 was
  retained.
• About Windows 3.1
• a good demonstration of what happens when
  usability is not taken into account