Towards standard methods and metrics for evaluating IA

[Rashmi Sinha <rashmi AT rashmisinha.com>]

The current SIGIA thread on "Questioning common test scripting: 
http://www.info-arch.org/lists/sigia-l/0306/0159.html raises some relevant 
issues for this group.

Ash makes an important point that tests can have different goals (observe 
in natural settings, collect metrics, get users's subjective reactions), 
and the methodology depends on the goal. One method does not fit all goals. 
In some sense  "usability testing" has become an umbrella term that 
encompasses different methods, and different goals, leading to some confusion.

-rashmi