Privacy and Property on the Net: Research Questions

[Eugen Leitl]

http://www.sciencemag.org/cgi/content/full/302/5651/1686

PERCEPTIONS OF SCIENCE ESSAY:
Privacy and Property on the Net: Research Questions
William Sims Bainbridge*

The Net, which consists of the Internet plus the parallel networks of
commercial and military data communication, raises serious issues of personal
information vulnerability and ownership of intellectual property.
Peer-to-peer sharing of music files over Internet, for example, challenges
the existing system of copyright. Counterterrorism "data mining" systems hunt
for signs of suspicious activity across vast databases of information about
people who do not wish to be spied upon. Research by social and information
scientists is needed to clarify many such issues.

To a sociologist, perhaps the most interesting quality of Internet and other
data transmission networks is their potential to alter power relationships
with respect to personal privacy and intellectual property. Both are
restrictions on the free flow of information. Government is implicated in
both, on the one hand by providing legal support for them and, on the other
hand, by potentially violating them in the pursuit of its own goals, such as
national security. Social scientists have only just begun framing research
projects to learn how the Internet is liberating information from traditional
restraints or to understand the likely human consequences.

Although both privacy and property are rooted deep in humanity's evolutionary
past, they are variables, and societal norms change. For example, until a
century and a half ago, U.S. census takers would post their completed
enumeration forms in the town square for anyone to read, but from 1850 until
1954 when Title 13 of the U.S. Code forbade publication of an individual's
records, a complex series of steps gradually increased confidentiality
protections (1). Today, the Census Bureau keeps the data confidential for 72
years. This change has largely been driven by the increasing government
collection of data about its citizens, to facilitate social services,
taxation, and management of the economy (2). By offering confidentiality,
government hopes the public will relinquish some of its traditional privacy.
Whether citizens benefit from government collection of data about them is
another matter. In Maryland, every prospective juror is asked what his or her
religion is, even though this information is not used in the jury selection
process. The religion data are kept confidential, but all information about
prospective jurors that is used in the selection is made public. Thus, we
have the bizarre situation of costly information being kept confidential
precisely because it is useless.

The idea that government should regulate intellectual property through
copyrights and patents is relatively recent in human history, and the precise
details of what intellectual property is protected for how long vary across
nations and occasionally change. There are two standard sociological
justifications for patents or copyrights: They reward creators for their
labor, and they encourage greater creativity. Both of these are empirical
claims that can be tested scientifically and could be false in some realms
(3, 4).

Consider music (5). Star performers existed before the 20th century, such as
Franz Liszt and Niccolo Paganini, but mass media produced a celebrity system
promoting a few stars whose music was not necessarily the best or most
diverse. Copyright provides protection for distribution companies and for a
few celebrities, thereby helping to support the industry as currently
defined, but it may actually harm the majority of performers. This is
comparable to Anatole France's famous irony, "The law, in its majestic
equality, forbids the rich as well as the poor to sleep under bridges." In
theory, copyright covers the creations of celebrities and obscurities
equally, but only major distribution companies have the resources to defend
their property rights in court. In a sense, this is quite fair, because
nobody wants to steal unpopular music, but by supporting the property rights
of celebrities, copyright strengthens them as a class in contrast to
anonymous musicians.

Internet music file sharing has become a significant factor in the social
lives of children, who download bootleg music tracks for their own use and to
give as gifts to friends. If we are to believe one recent poll done by a
marketing firm rather than social scientists, 48% of American Internet users
aged 12 to 17 had downloaded music files in the past month (6). In so doing,
they violate copyright laws, and criminologists would hypothesize they
thereby learn contempt for laws in general. A poll by the Pew Internet and
American Life Project found that two-thirds of an estimated 35 million
Americans who download music files do not care whether they are copyrighted
(7). Thus, on the level of families, ending copyright could be morally as
well as economically advantageous. On a much higher level, however, the
culture- exporting nations (notably the United States) could stand to lose,
although we cannot really predict the net balance of costs and benefits in
the absence of proper research. We do not presently have good cross-national
data on file sharing or a well-developed theoretical framework to guide
research on whether copyright protection supports cultural imperialism versus
enhancing the positions of diverse cultures in the global marketplace.

It will not be easy to test such hypotheses, and extensive economic research
has not conclusively answered the question of whether the patent system
really promotes innovation. We will need many careful, sharp-focus studies of
well-formed hypotheses in specific industries and sectors of life. For
example, observational and interview research can uncover the factors that
really promote cultural innovation among artists of various kinds and
determine the actual consequences for children of Internet peer-to-peer file
sharing.

The classic sociological theory of privacy asserts two main principles that
are relevant to current information debates (8, 9). First, private relations
between individuals are not merely a personal luxury but the very basis of
all societies. Thus, surveillance of citizens in the name of national
security undermines the very society it was intended to protect. Second,
privacy and its violation are inescapably issues of power, with power defined
as the ability to defend one's own privacy while being able to invade the
privacy of others. Thus, citizens' privacy is violated by government
surveillance, even if the information is kept confidential afterward by the
agencies.

Computer scientists have begun to develop systems that would defend people's
information privacy (10). For example, the World Wide Web consortium, which
is the chief forum for development of Web standards, has launched the
Platform for Privacy Preferences, which automatically manages personal
information when interacting with Web sites, following the explicit wishes of
the user (11).

Ethically informed research can be valuable for design and implementation of
information systems. Data mining, for example, can be used in
noncontroversial situations like intrusion detection, when you need to defend
your own data against attack from outside. But it can also be used
aggressively to sift through vast troves of data, pulled together via the Net
in a process called data fusion, in a way that violates the privacy of
law-abiding citizens while hunting for a few criminals or terrorists. In its
recent report, Information Technology for Counterterrorism, the National
Research Council blithely suggests that all airport security baggage x-ray
machines could send their pictures to a unified computer network that would
monitor the collective movements of terrorists boarding different airplanes
in different cities, incidentally spying on everybody else who flies (12).
Proper sociotechnical design could limit the harm. For example, an automatic
data-mining system could seek patterns of suspicious behavior, without
allowing any human being to see the data. Then, specially sworn court
officials could carefully examine the suspicious cases, before reporting to
law enforcement only those few that met statutory definitions of probable
cause.

What is the optimal design for such a multitiered confidentiality system?
Would there be unintended consequences? Would it be acceptable to the public?
Without good answers to such questions, information technology could fall
under the same cloud of public suspicion that hangs over nuclear power and
genetic engineering. Already, aggressive telemarketing has made people very
reluctant to answer scientific surveys administered by telephone (13).
Internet-based surveys have promise, but they typically lack the reliability
of traditional random-sample polling. Another challenge is that impartial
research on the real consequences of government security-motivated data
fusion and mining may not be possible within the government security regime
itself, because impartiality in science requires the research to be made
public.

Like other buzzwords, "data mining" is difficult to define precisely, and it
is practically indistinguishable from a wide range of statistical and pattern
recognition techniques used throughout the sciences. Thus, if it is brought
into disrepute by privacy violations, science could be harmed. Similarly,
many scientists use data that belong to other people, such as the social
survey data archived at the Inter-University Consortium for Political and
Social Research (14), so they would be affected by any change in the
intellectual property regime. Internet arose as a medium of scientific
communication and is an indispensable tool for scientists today. Thus,
research on Internet privacy and property issues could benefit science
itself, as well as society.

Only an anarchist would argue for complete government deregulation of
information, but any social scientist would find it interesting to
contemplate the implications of major changes in government enforcement of
intellectual property rights and government collection of data about its
citizens. This essay has briefly sketched theoretical issues that would
underlie research on information privacy and property in the age of Internet,
and policy-makers should be aware that current practices may reflect cultural
lag, the conflict-ridden situation when technological development has
rendered traditional norms obsolete (15).

References and Notes

   1. R. McCaa, S. Ruggles, Scand. Pop. Stud. 13, (2002).
   2. D. Bell, The Coming of Post-Industrial Society (Basic Books, New York,
1973).
   3. C. A. Ganz-Brown, J. World Intellect. Property 1, (1998).
   4. National Research Council, Computer Science and Telecommunications
Board, The Digital Dilemma: Intellectual Property in the Information Age
(National Academy Press, Washington, DC, 2000).
   5. W. S. Bainbridge, in Social Sciences for a Digital World, M. Renaud,
Ed. (Organization for Economic Co-Operation and Development, Paris, 2000).
   6. Ipsos-Insight, "Legal issues don't hinder American downloaders" (14
March 2003); [cited from www.ipsos-reid.com 5 May 2003]; available at
www/ipsos-insight.com.
   7. M. Madden, S. Lenhart, "Music downloading, file-sharing and copyright"
(July 2003) at www.pewtrusts.com.
   8. B. Schwartz, Am. J. Sociol. 73 (1968).
   9. W. S. Bainbridge, "Privacy," forthcoming in Encyclopedia of Community,
K. Christensen and D. Levinson, Eds. (Sage, Thousand Oaks, CA, 2003).
  10. National Research Council, Computer Science and Telecommunications
Board, Who Goes There?: Authentication Through the Lens of Privacy (National
Academy Press, Washington, DC, 2003).
  11. World Wide Web Consortium at www.w3c.org.
  12. National Research Council, Computer Science and Telecommunications
Board, Information Technology for Counterterrorism (National Academies Press,
Washington, DC, 2003).
  13. D. El Boghdady, Washington Post, 8 September 2002, p. H1.
  14. Inter-University Consortium for Political and Social Research at
www.icpsr.umich.edu.
  15. W. F. Ogburn, Social Change (Huebsch, New York, 1922).
  16. The views expressed in this essay do not necessarily represent the
views of the National Science Foundation or the United States.

The author is in the Division of Information and Intelligent Systems, the
Directorate for Computer and Information Science and Engineering, National
Science Foundation, 4201 Wilson Boulevard, Arlington, VA 22230, USA. E-mail:
william.bainbridge at verizon.net

-- Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
http://moleculardevices.org         http://nanomachines.net

[demime 0.97c removed an attachment of type application/pgp-signature]