New Challenges for the New Year

From the Editors

ou’ve heard from me in this space before, but this is great interest that the RoboCup soc- cer competitions have generated. In my first column as IEEE Security & Privacy’s editor addition to possibly advancing the state of the art, such competitions can in chief. I feel both honored and privileged to have be highly educational and entertain- ing. A few years ago when I asked the opportunity to assume this responsibility. Google cofounder Sergey Brin what Y he had found most beneficial in his CARL E. George Cybenko, as both the driving force behind the magazine’s undergraduate career at the Univer- LANDWEHR sity of Maryland, he pointed to the Editor in Chief creation and its EIC for the first four I’m especially interested in hearing programming competitions in which years, is a hard act to follow. But be- from you about what you like and he had participated. cause he created such a strong base for don’t like in S&P. We can and do While at the US National Science the magazine, I’m hoping it won’t be a monitor the download rates for vari- Foundation, I spent some time trying difficult act to continue. You can ex- ous articles that we print, but those sta- to figure out how to structure a chal- pect the mix of articles, departments, tistics tell only part of the story. This is lenge or a competition that would and special issues on current topics to a volunteer effort, and we depend on help us move beyond our present continue. You’ll see some new names freely submitted contributions from stage of penetrate-and-patch security. in the masthead as we replace those the community. To ensure the quality A colleague with experience on both rotating off the editorial board; this is a of what we publish, we also depend on the defensive and offensive sides of normal process for all IEEE publica- a peer-review process, which requires software security told me why offense tions. We’ll continue to strive for fresh volunteer work. I want our volun- is easier: Most software comes bun- and interesting material to keep you at teers’ time to be well spent. dled, and something in the bundle is the forefront of technology and issues If you would like to help—partic- likely to have an exploitable flaw; in security and privacy. ularly if you have ideas for improve- once the flaw is exploited, today’s sys- These are changing times for ments—please get in touch with me tems have few internal barriers to print publications and for profes- by email ([email protected]), contain attacks. He attributed this sit- sional societies. The technology that but if you see me at a meeting, feel free uation, in part, to deficiencies in many of us helped develop is having to corner me. If you would like to computer science education. In his tremendous impact in publishing submit a contribution to one of the view, students rarely face the respon- and in information distribution gen- magazine’s departments, please con- sibility of developing a significant erally. As a trial, the IEEE has al- tact the department editor (their email piece of software and then integrating lowed us to offer subscriptions to addresses are at the top of each depart- it into a larger system. Without such nonmembers of the IEEE Com- ment’s opening page). Information experience, students are unlikely to puter Society at a greatly reduced on how to submit regular articles is grasp the need to carefully check price: US$29 per year. Together available on the magazine’s Web site input parameters and provide strong with our content, this offer has (www.computer.org/security/). internal barriers in software systems. helped us increase our subscription I thought it would be great to base in a time of generally declining come up with a competition that participation in professional soci- hifting gears a little, there’s an- could help students learn these lessons, eties. We’re also trying to take ad- S other activity in which I would but doing so isn’t easy. The US Na- vantage of new media. A prime like to engage you: grand challenges. tional Institute of Standards and Tech- example is Gary McGraw’s Silver Few people can have failed to notice nology has made very effective use of Bullet Security podcasts. Digested the interest that Darpa’s challenges for competitions to design new, open versions of these appear in the maga- autonomous vehicles have generated. source cryptographic algorithms, but zine (p. 9 of this issue). Similarly, I’ve been impressed by the continued on p. 4

Willis Ware: Data privacy pioneer

EEE Security & Privacy magazine awarded Willis Ware the Pioneer Award for outstanding contri- I butions to data privacy issues and his pioneering efforts in information security research and policy development. S&P editor in chief Carl Landwehr presented Ware with the award on 6 November 2006 at RAND in Santa Monica, California. S&P’s editorial board and RAND executives attended a reception immediately following the ceremony. Ware joined Princeton’s Institute for Advanced Study in 1946, contributing to the design of the first parallel and asynchronous digital computer with John von Neumann. In 1952, he joined RAND and remains a senior computer scientist emeritus. In addition to his work at RAND, the early 1970s saw Ware chair the Special Advisory Committee on Automated Personal Data Systems. The committee’s report, nicknamed the Ware Report, was the cornerstone of the Federal Privacy Act of 1974. The late US President Gerald Ford appointed Ware to the Privacy IEEE Security & Privacy editor in Protection Study Commission, which authored a report that remains the most definitive exami- chief Carl E. Landwehr presents nation of record-keeping practices in the private sector. Ware was the first chair of the Computer Willis Ware with the Pioneer System Security and Privacy Advisory Board (now called the Information Security and Privacy Award. Advisory Board), which advised the US government on computer technology’s societal impacts. Ware has a PhD in electrical engineering from Princeton University. He is a fellow of the IEEE, the ACM, and the American Association for the Advancement of Science, as well as a member of the National Academy of Engineering. In addition, Ware served as the first president of the American Federation of Information Processing Standards.

continued from p. 3 the primary competitions I’ve seen in Letters computer and network security have been capture-the-flag exercises or Dear Editor, • Accountability is to behavior as penetration tests of one sort or an- I read with interest the recent gravity is planetary orbits. other. These have their benefits, but theme issue of IEEE S&P (Data they don’t seem likely to lead to long- Surveillance). With the disclaimer To keep data in enough use to term technical progress in security. that data surveillance is precisely be valuable, but not so much use But just because I haven’t been what I do for a living and in mo- as to reach escape velocity, there is able to come up with the right idea dern societies if you do something no choice but to impose the right doesn’t mean you can’t. Here are my for a living then no one should amount of accountability. To not desiderata for a challenge problem in trust your opinion about it, I want gum up machinery nor keep computer and network security: to say something anyway, I’ll just people from getting their work confine myself to fact. done, that accountability cannot 1. It must be difficult enough, and These are facts: rely on the sentience or the coop- relevant enough, that accom- eration of whatever is involved. plishing it will lead to a measur- • Data has value; Accountability without cooperation able advance of some sort in • The fraction of corporate wealth means surveillance. security technology. that is data is growing; As such, the fork in the road we 2. It must be possible to impartially • Data is most valuable when it is face is either to surveil people or and repeatedly rank the results of most used. data. Yes, the books and records efforts by different competitors. produced are homeomorphs of 3. It must be interesting enough to These are facts except in a few each other, but if my choice is a attract widespread interest and special cases: datum or a person as the primary simple enough to explain to unit of observation, then I’ll take those not involved in the field. • Institutional data perimeters barely the datum; it’s the lesser of two evil exist; necessities. Your choice may differ, There’s much more to be said on this • In fact-changing environments, au- but not to decide is to decide. topic, but I would like to hear your thentication is diseconomic for —Daniel E. Geer, Jr., Sc.D. views. If you’ll contribute, I’ll summa- fine-grained data control; Verdasys rize the results in a future column.

4 IEEE SECURITY & PRIVACY ■ JANUARY/FEBRUARY 2007