Money Alone Not Enough to

Motivate Data Theft

The theft scenario explored in “The Illusion of Security” (Mar. 2008) by David Wright et al. was realistic, with one exception, though it didn’t detract from the article’s conclusions. Wright et al. wrote: “A third driver, not so dissimilar from the first, is that the data thieves are also impelled by the profit motive.” The thieves in this case were employees who disappeared one day and some time later turned up in Costa Rica.

PAUL WA TSON

I realize the scenario was abbreviated, with many details omitted, but after having interviewed more than 200 computer criminals in my own studies (see Crime by Computer, Scribners, 1976, and Fighting Computer Crime, Scribners, 1983), I conclude that few computer criminals are indeed motivated purely by profit. Employees become criminals during employment mostly to solve personal problems that may involve money, sabotage, or espionage. They are often motivated by debt, relationships gone bad with other employees or spouses, personal dissatisfaction, or an attempt to hide poor or unethical business decisions. If the thieves

cited in the article were motivated by profit alone, staying with the company and surreptitiously selling its products and services under the table and engaging in accounts-receivable or -payable fraud, their behavior would likely be less dangerous and obviate the need to flee. On the other hand, they may have been in touch from the start with a buyer interested in the whole database who initiated the theft. However, I have found that collusion between two perpetrators is rare and among three rarer still when IT is involved. It usually takes only a single person with the proper skills, knowledge, resources, authority, motives, and objectives.

DONN B. PARKER Los Altos, CA

 

COLLABORATIVE SOLUTIONS FOR DISPARATE PROBLEMS Peter Denning’s and Peter Yaholkovsky’s “The Profession of IT” column “Getting to ‘We’” (Apr. 2008) was especially interesting in terms of open-source development. Although both authority and competition have a place in vibrant open-source communities,

it is surprising how often collaborative solutions satisfy disparate needs. For example, both SMP scalability and real-time response are sometimes improved by changing the kernel’s synchronization design.

I cannot say whether the five stages of collaboration outlined in the column apply directly to open source but attest to the effectiveness of the third stage: “listen to and learn all perspectives.” I’m intrigued by how “declare” and “connect” might by undertaken by a new generation that has grown up with the Internet.

PAUL E. MCKENNEY

Beaverton, OR

DISK INCREASES SIZE OF MEMORY-LIMITED SEARCHES I agree wholeheartedly with Daniel Kunkle’s and Gene Cooperman’s “Viewpoint” “Solving Rubik’s Cube: Disk Is the New RAM” (Apr. 2008). In fact, my co-authors and I have been developing disk-based search algorithms and using them to solve combinatorial problems since 2002. For example, the well-known Towers of Hanoi problem