A recent Adobe Reader zero-day exploit is notable for being the first in the wild to fully escape Reader's sandboxing capabilities. Could you explain how this attack works? Does it cast doubt on sandboxing as an effective enterprise application hardening technique?