This talk was given at the Rocky Mountain Ruby conference on 9/1/2011.
Selenium’s a wonderful tool for automating acceptance and functional tests; however, real-world implementations bring a lot of pain. I suffered all that pain, and more, as I piloted an effort that started out with Selenium IDE, moved through RC, and ended up with WebDriver. This talk covers things like setting up baseline data, creating backing test frameworks, dealing with brittle tests, and figuring out how to appropriately manage all those incredibly slow Selenium tests so that you actually get effective, useful testing in. Learn from my pain (and successes!) so that you don’t have to suffer it in your own projects!