Voxxed Days Thessaloniki 2023 - Deciphering alerts: Make sense of what your software is saying

by LENA NYSTRÖM If you have used software, you have likely come across messages that made you pause and think “Did a human write this?” Information, errors, warnings and other alerts can be incredibly unhelpful and sometimes even misleading. But why is it so hard to tell us what happened and what we need to correct to do what we want to do? Well, the truth is there are a lot of perspectives to balance and they often clash. Messages tell a story. A story of choices, of compromise and of the world they live in. They leave a trail of breadcrumbs and following the clues might even show us the way to problems hiding underneath. From the often crisp and immediate messages in frontend validation to the multi-time translated notifications originating from a database or an integrated service, there are lots of hints to pick up on. In this sesson we will look into how different parts of the tech stack deal with validations, errors and other types of information to the user, how to guess which part of the stack they are from and how we can use that to do better testing. We will look at examples, look into their strengths and weaknesses and why it is so hard to design the perfect message. The goal is to level up our testing by better understanding what the system is actually trying to tell us.