| Kurzkommentar |
DESCRIPTION
Designing easy-to-use user interfaces is a challenging problem-solving task. The number of possible designs in any meaningful design problem is too large to explore manually. For example, there are 10^26 possible virtual keyboard designs even in the simplest formulation of the keyboard arrangement problem. Ideally, the analysis of well-studied parts of a design could be automatized, allowing the designer to focus on the truly novel aspects of the problem. This goal can be achieved by models of human-computer interaction that make predictions on optimal designs. They can also be used to adapt the user interface to users' capacities dynamically during runtime.
This course familiarizes students to predictive models of human-computer interaction that capture well-studied aspects of interaction and they learn to use computational methods to optimize and adapt user interfaces for improved user performance. Whereas user interface design has been traditionally largely based on manual work and trial and error, this approach allows aggressive exploration of user interfaces.
After introductory lectures, student teams (pairs) design and complete a mini-project. The project goal is to develop an optimized user interface using the principles and methods taught in the course. This involves defining and representing the design space, acquiring a predictive model and distribution of user actions, building an optimizer, implementing the outcome(s), and testing them with users (light-weight). The outcome is presented to the whole department in an open demo session.
INSTRUCTORS
Dr. Antti Oulasvirta (http://www.mpi-inf.mpg.de/~oantti/)
Dr. Gilles Bailly (http://www.gillesbailly.fr)
Guest lecturers
PARTICIPATION
This is a course meant for advanced, computationally oriented computer science / visual computing students with some exposure to human-computer interaction. The two hard prequisites are 1) a previous course related to HCI, such as on the topics of user interface design or information visualization; 2) programming skill. Background in optimization, behavioral sciences, or Matlab are considered as advantages but are not necessary.
Participation is limited to a maximum of 10 students. Interested students are asked to send 1) brief letter stating their relevant background and interest toward the course and 2) course transcript. Please email to Antti Oulasvirta at firstname.lastname@mpii.de by March 23, 2013.
MEETINGS April 16 - June 26
The weekly meetings start on the first week of the semester and take place on:
- Tuesdays 9.00-12.00 am.
Place: building E1 7, room 008
CONTENTS (Tentative)
1. Organization:
- Welcome and introduction
- Overview of overall course requirements
- Resources for and structure of course project
- Selection of topics and student teams (pairs)
2. Introductory lectures with open discussion:
- The problem of user interface design: Definition and a history of approaches
- User interface description languages, designer tools, prototyping tools, automation of design and evaluation
- Adaptive and intelligent user interfaces
- Predictive models of motor control, perception, attention, memory
- User simulations, cognitive models, compound models
- User interface optimization basics
- Introduction to resources in students' disposal
3. First presentation and evaluation of student plans (1 meeting)
4. Lectures on specific application topics:
- Keyboards
- Menu interfaces
- Gestural interfaces
- Continous control
- Command language interfaces
- Dialogue interfaces
- Graphical representations
- Hypertext and WWW
5. Second review of student plans (1 meeting).
6. June: Independent student project work. No meetings during this period, but instructors are available for meetings.
7. Student projects present outcomes openly to the department. Written feedback from instructors.
WEEKLY PEN & PAPER EXERCISES
Instructors will provide a pool of exercises on the course homepage, each with a background reading. Students must return one exercise per week to the meeting on paper.
REQUIREMENTS AND GRADING
For passing the course, students must have
- been present in all course meetings. Max. 2 absences are allowed!
- returned acceptably 75% of the pen&paper exercises.
- completed the project work in collaboration with the team and according to instructions.
- presented the project in an open demo session at the end of the course.
Grading is based on:
- Project process and outcome (50%)
- Project plan (20%)
- Exercises (20%)
- Contribution to discussion at meetings (10%)
|
