Issue No. 02 - March/April (2009 vol. 7)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MSP.2009.48
Shari Lawrence Pfleeger , RAND Corporation
I have a friend whose profession involves information technology, but she has no presence on the Web. A few years ago, her identity was stolen; the experience was so devastating that she now works hard to make sure that no one can find out anything about her online. At the same time, all of us have become increasingly dependent on the Web. Indeed, many businesses provide online information only, and then only if we "register" with the site in exchange for the information and services we need. So it's no surprise that our personal information—including data about our Web habits and searches—is the basis for many businesses' profit models. The result? Several times each day, we must decide (explicitly or implicitly) whether to trade our personal information for convenience, products, services, or information possessed by others.
Greg Conti addresses this problem in his new book, Googling Security. He clarifies up front that his catchy title uses "Google" as a verb, not just as a particular search engine. That is, the book warns readers about the proliferation of data capture and the degree to which we're tracked and traced on the Web. In 10 heavily illustrated chapters, Conti explains how our data are captured, retained, and shared (often for a price) with third parties. With screenshots, code snippets, and a plethora of citations, he describes how searches work, how they fit in the larger framework of communications, mapping, and advertising, and how they pose a threat not only by sharing information but also by possibly letting malware into our computers in the process.
Sometimes alarmist and often mired in more details than we might want to read, Conti nevertheless thoroughly lays out the various ways in which our personal data can end up where we might not want them to be. For example, he points us to Google's Web caching—once something is posted on the Web, it remains in the cache even if it's immediately taken down. Readers of S&P are already likely to understand the consequences of this exposure; for those who aren't, an extreme situation is depicted in the Peter Berry thriller, The Last Enemy, (available at www.pbs.org/wgbh/masterpiece/lastenemy), and the Consumer's Union has presented a balanced report about online risks (Consumer's Union, "Net Threats: Why Going Online Remains Risky," Consumer Reports, Sept. 2007). I strongly recommend that readers interested in data proliferation also read Robert O'Harrow's book, No Place to Hide(Free Press, 2005), to see what organizations do with the data they amass.
Conti devotes his penultimate chapter to ways of addressing the threat. He suggests methods for controlling cookies, diffusing or eliminating disclosures, encrypting important data, protecting a network address, and using policy as well as technical devices to minimize data capture and retention.
In the last chapter, Conti mentions the trade-off between these risks and the functionality and convenience offered by Web sites and services. It's here that Googling Security is disappointing. Once Conti raises the alarm, it would have been helpful to read not only about how to prevent some situations from occurring but also about how to tell if adding the functionality is worth the effort. Conti clearly knows such decision-making support is essential; for instance, he briefly mentions a recent Carnegie Mellon study about whether users are willing to pay extra for privacy. But the mention is brief and the study only a preliminary convenience survey, not a definitive assessment. The book whets our appetite but never serves the last, most satisfying course.
Shari Lawrence Pfleeger is a senior in-formation scientist at RAND Corporation. Contact her at firstname.lastname@example.org.