I agree. To me, its simpler to understand. Suppose x is a bool, reading the code, I say to myself "if not true..." or "if not false..." and my ape brain gets confused on what happens if its not true or not false.

Reading "if true == false" or "if false == false", it becomes much clearer what we're testing here and I understand it instantly.

It also helps readability if the ! is before a function name that doesn't follow the right naming convention for it. One of my pet peeves in C is "if (!strcmp(a, b))". "!strcmp" I read as "not string compare" and I would expect it to mean that the strings don't compare when it means the exact opposite. This is true of anything following the "0 means success, anything else is an error condition" error handling scheme. So I use "if (strcmp(a, b) == 0)" instead because the "==" makes look at what value it's being compared to specifically and I make fewer assumptions.

Even if the not operator in the language you're writing in happens to be the actual word 'not' ?

Nitpick (as in general you are right regarding java as well): I’m fairly sure they are the same for java in this instance. Both will convert Boolean to boolean, throwing an NPE if it was a null.

They're not equivalent in many languages (JS, C++/swift with operator overloads, if x is nullable etc. etc.).

While it may not matter in many circumstances, this is not the same as "if foo", because false is not nil.