Your app should crash more often.
Not because crashing is good. But because crashing is honest.
When your app hits an unexpected state and crashes, you know exactly what happened. Stack trace. Context. The full picture. You fix it.
When your app swallows the exception and keeps running, it is now operating in a state it never expected. You do not know it. Your users might not know it yet. But something is wrong, and there is nothing left to tell you where.
Most try…catch blocks I review in production code have no recovery logic. The developer caught the exception. They logged it (maybe). They moved on.
The app is now carrying invisible damage.
Catching an exception is not the same as handling one. Handling means you intercepted a specific, known failure and had a deliberate response. Catching everything with no plan is not handling. It is hiding.
Crash fast. Fix the root state. Build software that actually stays stable over time.
How many try…catch blocks in your last codebase had a real recovery path?