Modern acceptance testing frameworks like Cucumber express tests in natural language, enabling organizations to establish their own Domain Specific Languages (DSLs). This powerful capability is a huge boon for communication: technical team members and non-technical business stakeholders can use the same vocabulary in expressing both requirements and tests. However, just creating a DSL does not ensure we will write good, clear tests. Indeed, it is all too easy to create acceptance tests that contain so many extraneous details that the real intentions behind the tests are obfuscated. In this session, Elisabeth will demonstrate how to edit technically correct but verbose Cucumber tests to increase clarity, reduce extraneous distracting details, and improve maintainability. Along the way we'll see how paying attention to the advice in Elements of Style applies to writing better automated acceptance tests. Strunk and White would be proud.