Who Needs Hackers? New York Times (09/12/07) P. H1; Schwartz, John
as it appeared in the September 12, 2007 edition of ACM TechNews.
Though conceding that computer hackers are a clear threat, experts maintain that some of the most serious and disruptive network problems can be traced to non-malevolent sources, most notably a network's complexity. "We don't need hackers to break the systems because they're falling apart by themselves," says SRI International principal scientist Peter G. Neumann. Nemertes Research's Andreas M. Antonopoulos says the transition from relatively simple computing architectures to massively distributed and connected networks has increased the difficulty of predicting, detecting, and correcting flaws. A problem as simple as a defective network card can have a cascading effect that leads to a network failure, such as the one that shut down computers for the U.S. Customs and Border Protection Agency and delayed flights at Los Angeles International Airport for hours last month. "Most of the problems we have day to day have nothing to do with malice," says Columbia University computer science professor Steven M. Bellovin. "Things break. Complex systems break in complex ways." He notes that it was a cascading series of failures that shut down the electrical grid in the Eastern United States and Canada in the summer of 2003. The integration and interdependence of multiple computer networks only makes system-wide vulnerability to a single weak link more likely, according to Veracode CEO Matt Moynahan. Johns Hopkins University professor Aviel D. Rubin says high-tech voting machines could be extremely susceptible to glitches, and he entertains the possibility that the emphasis on the hacker threat has eclipsed the threat of unintentional problems. One way to minimize non-malicious disruptions is to strengthen systems' capacity for recovery through backup protocols, while Neumann believes the best strategy is to design security and stability into computers from the very beginning. Peter G. Neumann moderates the ACM Risks Forum; http://www.risks.org/ Click Here to View Full Article
