Automated testing is on the rise in the FOSS world, but there is still ample room for improvement when it comes to sufficiently comprehensive automated tests. The test suites of many FOSS projects leave a lot to be desired, a result that's often affected by useful FOSS practices that are taken too far. Identifying such practices, like placing excessive trust in code reviews because "given enough eyeballs, all bugs are shallow", or leaving tests for later in the spirit of "release often, release early", is a first step in having a discussion that will hopefully convince more projects to embrace automated testing, and improve the quality of FOSS overall.