FA> I was only suggesting, that to me there is an inconsistency there.

No, no inconsistency. Very few C library functions (if any?) return purely
boolean information. Certainly you can /derive/ boolean information from
it, but the intention is to return as much information as possible. Most
library functions return a specific value to indicate a failure, but other
return codes are often useful to the caller. The fopen() function, for
example, returns a file descriptor, or NULL in an exception or failure
condition. If it returned boolean, you'd only know if the open succeded,
but you'd have no way of accessing the file thus opened.

 FA> After all if an operation was successful, in boolean terms it would be
 FA> TRUE, so functions returning 1 on success and 0 on failure would make
 FA> more sense.

Sometimes it is. Actually, in C++ I quite often use the bool type (which is
a built-in type in that language). When using classes, you can retain
additional state information within the class itself, so the boolean return
is often a convenient way of doing things. Even in C, I've often used FALSE
and TRUE return values where it makes sense.

--- MaltEd/2 1.0.b6