have been reluctant to reveal vulnerabilities that they have identified in their own
systems.
Going back to the framework in Figure 25.1, these findings give us some optimism that the shaded area at the upper left exists. That is, it is possible to detect
attacks that are crafted to be optimal against the well-known memory-based algorithms. It remains an open question to what extent these detection measures extend
downward and to the right, into regions where attacks differ from the optimal and
have correspondingly less impact, but still remain a source of profit for the attacker.