Hi Adam: >>#2-Regarding logging errors: >>I have difficulty accurately logging errors for the following reasons: -Successful completion of certain tasks can follow very different paths. -It's not always obvious which errors are "critical" (prevent successful completion) as opposed to "non-critical" (recoverable errors). So I guess my question is...how do you log errors? As long as the user completes the task, a deviation from the ideal path may be acceptable. Again, it depends on the acceptance criteria that you set. If the efficiency (of use) is not impacted, I wouldn't consider deviation. A error should be more catastrophic in nature. The two measures that we use to identify a catastrophic error are: 1. Errors unknown to the user 2. Errors that destroy users work Hope this helps. Shibu Daniel Usability Engineer PRT Group Inc.