The Road to Reliable Automated Tests
summary
Defines a reliable test as deterministic and gives concrete fixes: replace brittle copy‑paste suites (one example is 20 copy‑paste tests that took three days to patch) with small, focused tests and extracted helper methods; use custom webdriver waits instead of raw click/sendKeys and assert by waiting for outcomes; run tests in CI and locally often, debug with breakpoints and console traces, avoid catch‑all handlers, use CSS selectors, run static inspections in IntelliJ and publish a reusable waiter library on GitHub for common wait patterns.
AI-written summary · about
chapters (16)
- 0:00Start
- 8:12Creating a reliable test from
- 11:02Follow coding best
- 13:48Less code is be
- 16:37Don't add worka
- 21:03Write tests in sma Easy debugging . Lots of test runs
- 24:05Make sure to add
- 27:43Visually inspect the
- 29:08Check the test pass correctly
- 30:35Lookout for false
- 33:16Understand what
- 34:35Debugging you
- 38:21When colleagues debugging the code
- 39:20Run the tests on a
- 41:10You can fix tests at
- 41:45Less maintenand
related talks
42mReliable and Maintainable Webdriver Tests
45mFrom fragile to fail-safe: writing durable tests
47mThe Valley of Success
21mTest Kitchen: A Recipe for Good Tests
21mThe Essential Pillars of Automation Testing for Success
40mFinding Flakiness: The Real Reasons Your Tests Aren't Reliable
48mTesting is Confidence - A Developer's Perspective
41mDeterministic Solutions to Intermittent Failures
