In Security or Security? The Real Risks Posed by New Wiretapping Technologies, Susan Landau focuses on the impacts of integrating surveillance systems into communications networks. Her specific thesis is that integrating surveillance capacities into communications networks does not necessarily or inherently make us more secure, but may introduce security vulnerabilities and thus make us less secure. This continues on threads that began to come together in the book she and Whitfield Diffie wrote, titled Privacy on the Line: The Politics of Wiretapping and Encryption, Updated and Expanded Edition.
Landau’s work is simultaneously technical and very easy to quickly read. This is the result of inspired prose and gifted editing. As a result, she doesn’t waver from working through the intricacies of DNSSEC, nor how encryption keys are exchanged or mobile surveillance conducted, and by the time the reader finishes the book they will have a good high-level understanding of how these technologies and systems (amongst many others!) work. On the policy side, she gracefully walks the reader through the encryption wars of the 1990s, as well as the politics of wiretapping more generally in the US. You don’t need to be a nerd to get the tech side of the book, nor do you need to be a policy wonk to understand the politics of American wiretapping.
Given that her policy analyses are based on deep technical understanding of the issues at hand, each of her recommendations carry a considerable amount of weight. As examples, after working through authentication systems and their deficits, she differentiates between three levels of online identification (machine-based, which relies on packets; human, which relies on application authentication; and digital, which depends on biometric identifiers). This differentiation lets her consider the kinds of threats and possibilities each identification-type provides. She rightly notes that the “real complication for attribution is that the type of attribution varies with the type of entity for which we are seeking attribution” (58). As such, totalizing identification systems are almost necessarily bound to fail and will endanger our overall security profiles by expanding the surface that attackers can target.