I’ve been spending more time fooling around and doing exploratory testing, now that I’ve run 99% of the test cases for my current project multiple times. Exploratory testing can generally be thought of as testing that, while thorough and methodical, does not necessarily follow a test script or explicit test case. Usually a tester is given a scenario (like “test all the field limits in this form”) and then set loose. Sometimes this kind of testing leads the tester on other paths through the application under test, which is fine. I think this can be an effective way to test an application, especially if good notes are kept and any bugs and issues identified are carefully documented.
Sam Kalman has published some really great information about practical applications of this kind of testing on his blog, Climactic Ave. This post was really great, and I used his suggestions to create my own template for session-based test reports. I wish I’d seen Sam’s blog a few months ago!