Wednesday, January 21, 2009

Usability, HCI, Task Analysis

Usability

1. (Jennifer Hogle) According to Smith and Ragan (2005), the goal of instructional design is to create instruction that is "effective", "efficient", and "appealing". On page 27 of the Leventhal & Barnes chapter, usability is defined as, "the extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use." What comparisons can be made between the usability and instructional design? How can understanding the process of instructional design help us in understanding usability?

*If you have bad ID, there probably wouldn't be good usability.
*If there is good ID, there is probably going to be good usability.
*Usability is a subset of ID. ID is the bigger umbrella. It's possible that in you ID process you would decide not to use computers at all. (Of course, usability can refer to non-computer based tools too.) But usability, although part of the whole ID process in some aspects, is more focused on the final product. It seems like a smaller part of the entire ID.
*Is there ever a time when good ID will violate good usability?
Maybe when it is done on purpose -- making something hard so that the user has to struggle a little to learn more. (Cognitive Load) But this would be intrinsic cognitive load, not extrinsic.

2. (Joanna Gibb) Nielsen & Shackrel are similiar in that their definition of 'usability' is a sytem; is valid or usable if it is found useful. How many systems are usable, easy to learn, but have no use once they have been learned? How useful will the final project be once it's completed? Easy to access only, or worthwhile and applied once it's been presented?

*There are times when it seems a system has been created that really is of no use to the the user. They did not task match. There wasn't the need for it. They didn't use participatory design.

HCI

1. (Kevin Dolan) The study and research of HCI is evident back to the 1980's - even as a multidisciplinary science and investment topic. Can we determine that there is measurable progress in the evolution of HCI over the years; i.e., is the software and computer development world improving the effort to connect humans and computers or are we just getting better at paying lip service to this area of interaction?

*Companies are having to pay attention to what customers want and need and want. If they don't pay attention to HCI they don't sell as much and people will go to another product.
*Web 2.0 has made user wants and needs more obvious. People are "talking" online about their opinions, wants, needs, etc. and they can't be as easily ignored or unknown.
*It seems that education is a niche that gets ignored. HCI principles aren't used as well because of limited funding and because they don't feel like they have to. Things are also designed top-down and then mandated to the users. They aren't driven by marketing, but the irony is they are in fact wasting money if they make a system that people will really use.

2. (W Scott Slade) If there is too much fragmentation, too many theories, too many methods, too many application domains and too many systems, is simplification possible?

3. (Shannon Ririe) Is there/should there be a committee or organization that defines what HCI is or isn't?


*Remember: You are not the user! You can't ever think like a novice again. It is really important to involve your user in the design process.


Tasks

Usability for Ed Tech
•Not just whether system is used (cf., Eason)
•System Functions should include
–Outcome of use(!)
•Task Characteristics should include
–Desirable difficulty
•Task match will need to include relevance to mental models/cognitive processes

Usability Engineering
•Works best early in design/development
•Good interface won’t solve all your problems
•HCI Usability Engineering builds on foundation of:
–Task
–User
–Context of use

Developing Useful Systems
•Slightly more than 30%of the code developed in application software development ever gets used as intended
•Likely because developers do not understanding what users need

Maybe also because the basic tasks were made usable and the rest was not

Maybe because the designer got excited after the basic user needs were designed and added a whole bunch of other stuff

Maybe the extra/advanced stuff too difficult to find users to test

The questions is do you want it this way so that as users become more proficient they can move on? or does the extra stuff get in the way?

Useful Systems Support Tasks
•Task ≠ User can use the system
•Tasks:
–Specific
–Observable
•Will you know if you’ve successfully accomplished it?
–Reflects end-goal of a user session
–Relate to key aspects of system components
•What do you think are the key parts of the system?
•What will be frequently used?

–Hugh Beyer and Karen Holtzblatt, "Contextual Design: A Customer-Centric Approach to Systems Design," ACM Interactions, Sep+Oct, 1997, iv.5, p. 62.

What is a Task Analysis?
•Analysis of how a task is performed
–Detailed description of behaviors in interface
•Highly detailed
•Step-by-Step
•Procedural (ignore mental processes for now)

How is Task Analysis Useful?
•Specify problems/gaps in process
•Highlight unnecessary or inconsistent steps
•Specifies procedural aspects of key tasks
•Requires concrete analysis of user actions


For next week:
•DUE:HCI Exercise: Task & Task Analysis Exercise - Something in education - Keyboarding for kids? - goal is to login and practice a lesson
•Read:2 articles on personas
•Post:2 questions per article on WebCTdiscussion area
–Due by 12 noon on day of class

No comments: