Lessons Learned in Software Testing: A Context-Driven Approach Book Price in India, Specifications, Reviews & Offers. Buy online at Amazon.

Great talk, great insight
People try to automated testing by automation user behaviour. Tool pretend to do like human? No, it does not exist. The tool helps the tester to do a better, faster, stronger job. But it can’t replace humans. Washing machine, wash your clothes. Does it look dirty and know where do you need to clean more? Fantasy story “I can create a tool replace human”? Too ridiculous. The tool includes automated GUI, do like a user, open browser, click something, scroll a bit, capture, pass. You build a tool that takes time, effort, money -> it does not actually help your work + it is worthless. Automated programmer? It doesn’t exist, how can you think about an automated tester? This is really silly, such. The regression testing fantasy – I rerun all the tests to ensure that nothing has broken. A tool can catch all bugs in your product? This can only be true if your old tests cover everything completely with perfect oracles so that conceivable bugs are detected. Are you sure your tool can catch the all-important bugs? If it misses a bug, this will be an absolutely minor bug? A context-driven way. Call them a tool, not automation, you can not automated testing. And tools do fact-check, think of testing much more than output checking. Explore many ways to use tools to do testing. 30 years of experience in testing advice young people “don't think automation can do a magic work.”. Don’t follow Jame’s mistake(build a tool that can test your product). The CDT learning paradox. CDT means choosing the right tool for the situation. In the right situation, we need the right skill. To gain the right skills, we need to practice. Practice means doing something we’re not good at. To get good CDT, we must sometimes be doing bad work.
Thanks for recording and sharing this!
Thanks for doing the talk.