TORONTO, May 28, 2004 — For all the benefits of a connected information society, the lightning speed at which vast amounts of data move across the Internet has made issues of privacy a primary concern for individuals, governments and organizations worldwide.
Caspar Bowden, Chief Privacy Strategist, Microsoft EMEA (Europe, Middle East and Africa)
According to Roger Dingledine, an independent researcher in the field of privacy and anonymity technologies, startling amounts of personal data — not only address-book information and account numbers, but purchasing habits and interests — can be accumulated and used for any purpose, the most pernicious of which is identity theft.
This week, many of the leading researchers in anonymity and privacy technology from around the world gathered in Toronto, Canada, for the third annual Workshop on Privacy Enhancing Technologies (PET), to discuss these issues.
“We’re turning into a networked world where every organization stores personal data and everybody is collecting more,” says Dingledine, who sits on this year’s PET Workshop program committee and was program chair for the past two years. “With so many pieces of data out there, you can correlate them into a surprisingly detailed profile. So it becomes valuable for people to attempt to control who ends up with their data and how it is used.”
Many governments have laws to prevent the inappropriate and unauthorized use of personal information. The European Union’s 1995 Data Protection Directive is one of the most comprehensive. E.U. residents are entitled to obtain access to all of the data collected about themselves — and how it has been used — from any public or private organisation.
Privacy and data protection is thus an issue of mounting concern for companies and governments, and the burden of regulatory compliance is considerable. Caspar Bowden, chief privacy strategist in the Technology Office for Microsoft Europe, Middle East and Africa, explains.
“When a person resident in the E.U. exercises the right to access their personal information from an organization, someone must collate and review the eligible material, a time-consuming and subjective task,” Bowden says. “The scope of information may be quite broad, including e-mail messages, archived data and structured information in databases, and the organization has to consider whether data which could impact the privacy of other persons should be omitted in the fulfilment of the request.”
All of this raises the question — what can technology do to help ensure that personal data is used only for authorised purposes and enhance people’s privacy as they use the Internet? A growing community of technology researchers is focusing on these issues.
“Everyone engaged in research, and indeed engaged as a practitioner in privacy, recognizes that the protection of privacy at the technical level is an extremely difficult problem,” Bowden says. “However, there’s a common belief that technical solutions can, in the future, play an important part in improving the protection of personal information.”
Microsoft, Other Industry Leaders, Join Privacy Researchers for Annual Meeting
Microsoft Europe, Middle East and Africa (EMEA) is a major sponsor of the PET Workshop, contributing funds for two purposes: the cash prize for the workshop’s Award for Outstanding Research in Privacy Enhancing Technologies, and stipends that subsidize travel costs for selected attendees.
The PET Workshop allows professors and graduate students in the privacy and anonymity field, as well as independent and corporate-affiliated researchers, to gather in one place to present and discuss the latest research, and share experiences of deploying new system designs.
“This is still an open area of research, and the PET Workshop conference is one of the leading forums for advancing knowledge, particularly of the computer science and economics of privacy,” Bowden says. “One of our objectives in this sponsorship is to encourage privacy research for its own sake, as well as to foster positive relationships in this community, which will undoubtedly shape the way privacy and data protection functionality develops in products over time.”
Bowden is on this year’s program committee, and in his role as chief privacy specialist for Microsoft in the EMEA region, he contributes an industry perspective on privacy issues as they relate to enterprise systems and software, particularly the problems businesses want to see addressed.
The PET Workshop grants an award for exemplary research in the field, which includes a
€
2,000 (US$2,400) prize. Members of the workshop’s program committee nominate research that has been published within the past year and determine which papers make an “outstanding contribution to the theory, design, implementation or deployment of privacy enhancing technology.” Typically, two winners are selected, representing different research areas. Microsoft will continue to fund the prize at least through 2005.
Microsoft’s contribution for attendee stipends is substantially greater, and helps offset registration for the attendees who might not be able to participate otherwise.
“A significant part of the success of the workshop is that we can bring 50 to 70 people together in one place,” Dingledine says. “Last year, we supported 25 people with stipend money from Microsoft, and this year it will be about the same. I’m very happy that Microsoft is interested in helping with this. The award money provides recognition to those who have made noteworthy contributions to the field, and give them an incentive to continue their research.”
PET Workshop 2004 Awards Announced
Last night (May 27), the winners of the 2004 Award for Outstanding Research in Privacy Enhancing Technologies were announced. The honor is shared by two entrants, each focused on a different area of privacy research.
One of these areas is preservation of anonymity in communications. Encryption is commonly used to protect the content of messages, but specialized techniques can also be used to prevent observation of the fact that communication is occurring at all. Studying the strengths and weaknesses of these techniques is important to establish solid foundations in the theory of privacy protection systems.
Matthias Bauer of the University of Erlangen in Nuremberg, Germany earned an award for his paper, which lays out a strategy for using normal online traffic — people browsing Web sites, for example — to protect sensitive communications from observation by blending them into the “cloud” of ordinary Internet transactions. This is a practical extension of technology that has existed for some years, but which suffered because of its reliance upon smaller groups of users and a limited volume of transactions among which to obscure communications.
“These techniques give protection in scenarios where a pattern of transactions might itself allow sensitive information to be inferred,” Bowden says. “As an analogy, an investment bank known to be preparing a takeover bid might inadvertently signal that an announcement is imminent with an unusually large number of pizza deliveries the night before.”
The other winning paper deals with a different realm of PET research: helping to rationalize privacy policies and their technical implementations across organizations. This work was conducted by a team from IBM’s research center in Zurich, which includes Michael Backes, Birgit Pfitzmann and Matthias Schunter. They tackled the problem of managing personal data in large, transnational organizations.
Different business units may have disparate privacy policies that are implemented in ways that are hard to reconcile. Furthermore, the legal requirements for handling personal data differ across jurisdictions. As a result, sharing customer data appropriately among operational divisions and across geographic borders is highly complex. The IBM team produced a “toolkit” for meshing privacy policies, including a vocabulary, rules and semantics.
“Transnational organizations are realizing that they need to come to grips with the different privacy regimes in different parts of the world,” Bowden says. “Integrating legal requirements with other aspects of corporate governance, and establishing effective policies, training and processes is a costly proposition. Enterprise customers that do business worldwide have told us they’d welcome technical solutions to such issues.”
From Research to Useful Products
Ultimately, advances in theoretical research must find their way into products that solve problems for people and organisations. Dingledine acknowledges that fostering interest and participation among the business community is a challenge.
“There are some companies that have built and are deploying tools for data privacy and anonymity, and we try to get representatives from those companies to participate,” Dingledine says. “But it’s an issue. How do you actually get these innovations out to the broader community? Not only does someone have to build the solutions, but people need to be educated about why they might want it.”
The E.U.’s Data Protection Directive now applies across 25 countries in Europe, so in theory it should result in substantial harmonization — and a common need for technical solutions — across the region. In practice, Bowden says, there are wide variations in the stringency both of enforcement and interpretation among different member states.
“Research into privacy protection technologies can clarify the alternative options for operationalizing privacy rules,” Bowden says. “A greater degree of convergence in national interpretations will lead to a more uniform view of how privacy can be designed into systems, and simplify compliance.”
Thus, an interdisciplinary effort is required, as well as wider dissemination of privacy research within the broader computer security research community. Indeed, one of the objectives of the PET Workshop is to draw more people into the work of providing more safety and individual control over how personal information is used and disseminated.
“The talks are technical,” Dingledine says, “but our ultimate motivation for doing this work is to help people.”
