Your team has just completed creating a deployment pipeline and moving to trunk based development, and you’ve discovered the next bottleneck: your application’s architecture! You want to develop that next great feature, but a bug fix has to go out and you don’t want to create complex branching and reduce feedback. Someone has suggested this “feature toggle” concept, but no one knows how to implement it. Your seasoned developers know that hiding code is more complicated than an if statement and insist toggles will never work. Your QA people reject the notion of toggles citing that there’s no way they can test all the toggle cases during a sprint. So what do you do? Watch me explain how it’s possible on a real, existing application!
In this session we’ll cover how to introduce feature toggles into your application to allow for new features to be hidden until they are ready for release, regardless of if they’ve been deployed. We’ll also show how you can introduce dependency injection and micro-services into an application to help isolate new development and cover best practices to make the most of these technologies as you iterate. Finally I’ll demonstrate how automated testing at the unit test and UI test layers can assist in ensuring your feature toggles are working as you expect without a lot of manual testing.