Lawrence M. Hinman
In this paper, I will be arguing in support of several theses:
Importance. Search engines play an increasingly crucial role in public life. For those who are working in this area, the importance of search engines is almost self-evident, yet for the general public search engines are transparent; they see the sites to which search engines lead them, not the search engines themselves. Consider three areas in which search engines have assumed a growing importance.
Political Life. In the political world, the common citizen depends on access to knowledge, without which the kind of discourse that Habermas and others see as essential to democracy would be impossible. Search engines play a crucial role in providing or denying that access. The absence of Abu Ghraib photos from an image search on Google illustrates the way in which search engines can shape political perceptions. Google’s acquiescence to Chinese political demands, as well as its country-specific treatment of hate speech, further add to this complex picture.
The Virtual Constitution of Knowledge and Reality.Initially, they are the “gatekeepers” of the web, but their importance is in fact far more important than that. Through their influence, they become part of a feedback loop and, as such, increasingly become constitutive of knowledge itself. This is most obviously the case with the development of Google Scholar, which will shape the next generation’s perception of knowledge in extraordinarily powerful ways.
Economic Life. Economic success or failure often depends on how close to the top a site is ranked in search engines. While moral progress has been made by separating sponsored sites from non-sponsored ones, but many thorny issues remain, such as the apparently greater weight given commercial sites over noncommercial ones, even when the noncommercial sites provide far more resources than their commercial counterparts.
Accountability. Search engines are responsible to the public, but accountable only to their stockholders and to their advertisers, who are the paying customers. Despite their rapidly growing importance in public life and in the growth of knowledge, search engines are under very little if any public regulation. Some companies, such as Google, have been quite concerned about self-regulation, but it is far from clear that this is sufficient, given the magnitude of the role that search engines will play in society.
Comparable Cases. When we look at other industries that occupy a similar public role-television and airlines are two examples-we see that they are regularly subject to federal scrutiny.
Proposal. Both individual countries and international consortia need to move toward developing appropriate regulatory structures for search engines. This needs to be done in a way that is simultaneously sensitive to the growth incentives for the search engine companies themselves and to the needs and rights of the larger public.
Areas of concern. What, if anything, needs to be scrutinized and regulated in search engines? This paper discusses three principal areas.
Algorithms. While early search engines used primitive criteria such as the number of page visits or back links, contemporary search and ranking algorithms are far more sophisticated and hidden from view. It is no longer clear what it means to have a given site come up first in a specific site-it is not merely popularity, not merely the number of other sites that link to that one, but it is unclear precisely what it means for a site to come up first.
What kind of regulatory structure would be appropriate for this situation? Clearly search companies have both a right and a duty to keep the details of their algorithms confidential. Not only are those algorithms what gives one company an advantage over other companies, but publication of the detailed algorithms leaves search engines open to even greater manipulation by spammers than currently possible. In this section, we explore what model regulatory structures might look like, structures that protect legitimate needs for secrecy and yet provide sufficient transparency that the public knows and understands what the search results actually mean.
When in Rome, do as? Search engine technology presents interesting issues in regard to ethical relativism. To what extent should search engines adopt a maxim of “When in China, do as the Chinese do” or other similar guidelines? We are moving increasingly toward an era of global ethics, and in this section we will argue in favor of the necessity of global solutions to local problems. A consideration of this problem will also help elucidate the shortcomings of ethical relativism in the twenty-first century.
Economic Privacy. As our buying habits become subject to increasing tracking, we find that sellers are able to manipulate our choices all the more effectively. In this section, we explore ways to insure consumer privacy while maintaining the economic incentives necessary to the continued growth of the search industry.
Political Privacy. In an era in which political privacy rights have often been sacrificed in the name of the war against terrorism, we find that search engine records offer a potentially powerful way to track and control citizens. We must establish guidelines that limit the power of governments to access information about citizens. In the United States, the government’s Carnivore program failed, not for want to political support but because of inept programming, a problem that has plagued the FBI for years. However, might search engines not deliver comparable information about citizens-what searches they have made, what links they have chosen to explore, etc.? Once again, the challenge is to develop regulatory structures that preserve legitimate interests while still maximizing the political freedom of the ordinary citizen.
Introna, Lucas D. and Helen Nissenbaum (2000) “Shaping the Web: Why the Politics of Search Engines Matters”, The Information Society, Vol. 16, No.3, 1-17.
Machill, M., Welp, C., eds. Wegweiser im Netz: Qualität und Nutzung von Suchmaschinen. Bielefeld: Verlag Bertelsman Stiftung, 2003.
Machill, M., et al. “Navigating the Internet: A Study of German-Language Search Engines.” European Journal of Communication.2004; 19: 321-347.
Baye, M. R. and Morgan, J (2001). Information Gatekeepers on the Internet and the Competitiveness of Homogeneous Product Markets, American Economic Review 91(3): 454-474.
Brake, D. “Judging Search Engines.” LSE MSc Coursework. http://www.well.com/user/derb/papers/judgingsearchenginesabstract.html Full copy available from author on request.
Phua, V. 1998. “Towards a set of ethical rules for search engines.” MSc dissertation, LSE [I have not yet been able to locate this author or the dissertation.]
In addition, I will be drawing on:
Numerous articles from GoogleWatch.com and related websites
Direct interviews and correspondence with Google representatives.