Don't be such a fucking child. "Duuuuuurrrrr real men aren't afraid of undefined behavior!" Undefined behavior doesn't mean they don't know what could happen, or anything could happen, or they're trying to hide the truth from you. It means the truth is more trouble than it's worth. The C standards committee knows, and isn't afraid to reveal, exactly what will happen if you cause a buffer overrun: either you'll corrupt memory in unrelated use, you'll write to memory not in use at all and your changes could be overwritten at any time mid-execution without notice, you'll cause a segment violation, or, if you're running an interpreted or needlessly runtime-heavy implementation of C, the interpreter or runtime might bitch at you. Some mustache-twirling cartoon-villain-esque will to conceal this information isn't the reason they don't tell you. They don't tell you because it's not in their interest to make it their responsibility. It's undefined behavior not because there is *no* definition of what will happen, but because neither the language standard nor the compiler documentation (else it would be "implementation-defined") have any business authoritatively declaring what definition there *is*, nor would it be worth their trouble. To explain the result of a buffer overrun, the C standards committee would have to introduce all sorts of new abstractions into the standard that are of no use to the C programmer, and serve no purpose except to explain said result, which itself has no features of any use to the C programmer. That's a lot of useless cruft shoved into the standard just to say what amounts to "don't do this" when they could just say "don't do this" without explaining fuckall and all the cruft would be magicked away just like that. Imagine calling yourself a minimalist and getting so assblasted that you don't get to have a pointlessly bloated standards document. Grow the fuck up.