Our tooling is becoming increasingly smarter, whether it is due to machine learning, AI, or simply improved design. Whatever the cause, our way of working is becoming increasingly efficient and faster. My IDE is increasingly able to understand and even complete my code, even while I’m still typing – scary.

However, with all this tooling, a small problem can arise: the loss of mastery of basic skills. Because is it still necessary to master a basic skill if a tool takes over and performs it flawlessly?

And it is precisely about that flawless execution. What if things don’t go well? What if the tool is confronted with something unknown? Then we must dust off our old books and go back to the basics.

One of the skills, basic but very important for automation engineers, is building locators. Robust, unique locators. Or should I say: building locators WAS an important skillset for automation engineers. Browser plugins can now generate locators for us. Automation frameworks nowadays indicate what would be a better locator and discourage the use of CSS and XPath. Should we still think about that? Couldn’t we spend our time much better on other, more “complex” issues?

This question may no longer be relevant in a few year’s time, but now I dare to answer it: Yes!

Make sure you are always able to build a unique locator yourself. Understand the interplay between the DOM elements: which element is visible at a certain time and not at another, iFrames and the relationship between parents, siblings, and children. Understand why a dedicated test-id attribute is ALWAYS a good idea. I could go on like this for a while.

While it may become obsolete soon, it’s still a good thing to master for now: building locators!

And isn’t it just fun? 😉