i want you to want me: the effect of reputation systems in online dating sites
posted by:jennifer barrigar // 11:59 PM // February 13, 2007 // ID TRAIL MIX
This piece is abstracted from a longer paper that is currently seeking publication venue.
By now it is almost trite to point out that the scale and breadth of the internet opens up the possibility of reaching large numbers of people quickly and easily, facilitating social and commercial matching on a scale hitherto unimaginable. At the same time, however, the internet is fraught with ambiguity. Text communications are denuded of gesture, tone and the million nuances that inform our interpretation of meaning. Even in visual arenas such as You Tube, recent events show conclusively that the lines between vlogging, fiction and commerce are fluid and difficult to discern. 
Reputation systems have been developed as a technological means to harness the potential of the Internet by making trust possible in online environments. This technology is used on many well-known sites. eBay’s feedback system, for instance, allows both the buyer and seller in a transaction rate each other, and the cumulative ratings are available for perusal by any eBay user attempting to determine whether to enter into a transaction with a particular individual. Amazon also uses a variation of a reputation system, allowing users of the site to submit their reviews of materials. A reviewer may rise to the rank of “top reviewer” based on feedback of other users, while all users come to understand that a reviewer’s status is predictive of the helpfulness of her review. Slashdot.org has a similarly dynamic reputation system in place, where site users submit and review news items as well as actively reviewing the contributions of others. Users of the site are able to modify their settings to show only top-rated items, and top-rated authors acquire “karma points” which increase the weight of their reviews and ratings. In each of these systems, the “reputation” of an individual is established by meeting the needs/expectations of other users, whether for trustworthy buyer/seller behaviour, reliable reviews, or a good eye for interesting and newsworthy items.
The use of reputation systems in online dating is somewhat less intuitive than its use in other arenas, because the “product” being judged is less clear. On eBay, the performance of a particular contract is rated. Although there is not originating contract in the Amazon sense, ratings of a particular reviewer are based on how well her product has met the desires/needs of the user. Slashdot.org’s reputation rankings are similarly performance-based, with status incrementally built through accurately representing and satisfying the desires of users of the site. Michele White has noted how “Amazon’s personalization options seem to allow spectators, who are depicted as active users, to write into the system and program it according to their desires.”  In the recent introduction of reputation systems to online dating sites we see even more clearly the encoding of desire and consequent regulation of performance.
The Manifesto for the Reputation Society claims that “when, in colloquial language, we speak of a person’s ‘good reputation’ we are implicitly claiming that the person fulfills many of his or her local society’s expectations of good social behavior – typically including qualities like honesty, reliability, ‘good moral character’, and competence.” 
As Lees recognizes, while ‘reputation’ for a man invokes social and cultural qualities, for a womyn ‘reputation’ has always denoted sexual behaviour.  This particularly gendered implication of ‘reputation’ in the arenas of sexuality and dating is further exacerbated by the context of the online dating environment. Although both men and womyn use online dating sites, research indicates that compared to Internet users in general, online daters are more likely to be male.  In addition, all users of these sites are products of our inherently sexist culture, which necessarily informs their responses to the world and to each other. Sexism exerts a constituting force on our identity, as it is “continually endorsed and celebrated by the dominant culture. The mass media, the daily press, pornographic magazines and videos all reinforce the objectification of women’s bodies and celebrate a form of macho, aggressive masculinity.”  Accordingly, I would argue that the standards encoded into the online dating system are inherently gendered.
A negative reputation, then, is the result of failure to conform to the group standards of the dominant culture. When users of these sites fail to perform and present the gendered identities expected of them, this transgression is seen as a failure in them to uphold expected moral codes, and reputation is thus formed and assigned within the system. Accordingly, if “those who defy the dominant position will incur a form of disapproval that will lead them to be less trusted, liked, and respected in the future” ,  then s/he who seeks to avoid a bad reputation must necessarily come to both understand and perform the expectations of the dominant position.
Reputation is not simply about purchaser choice and assisting purchasers to make choices that will best satisfy their needs – indeed, it depends for its power on a resulting regulatory force. Looked at in its full social context, reputation functions as a form of surveillance and, “like surveillance, may induce people to police themselves.”  The normative effect of reputation systems in online dating environments leads to a situation where “the culturally constructed ways that women express their femininity (emotional, shy, weak and nurturant) and men express their masculinity (unemotional, aggressive, strong and potent) are deemed to be natural.”  As such, womyn subject to these expectations do not experience themselves as deviating from individual expectations, but rather as transgressing normative standards. Similarly, men who are “disappointed” in these transactions do not experience their expectations as problematic, but rather are encouraged by the reputation system to enforce conformity with expectations rather than re-consider the expectations.
This analysis suggests that reputation systems in online dating environments function as a form of self-regulating surveillance – they set the standards of expected gendered behaviour, they act to enforce adherence to those standards by stigmatizing those who fail to conform them, and they normativize those standards, resulting in internalization of the standards and self-policing of behaviour. Far from the transformative tool of cooperation that reputation systems purport to be, in this environment at least they act to perpetuate a particular gendered and sexualized inequality.
It might be suggested that this is an isolated and site-specific issue, relevant only to online dating. I note, however, that of late there have been suggestions that reputation systems move from their current site-specific assessment status to become anchored on the individual identity instead. This would create a mobility of reputation, where individuals could build an amalgamated reputation that would be accessible to any/all persons or organizations interested in entering into a relationship with a particular individual. Before we implement any kind of mobile reputation system (or even before we increase our reliance on existing reputation systems) we must recognize their regulating power and problematize what is being regulated in order to ensure that the enforcement of stereotyped norms of behaviour and performance does not become part of this matrix.
 For examples, see the recent “lonelygirl15” (http://www.nytimes.com/2006/09/13/technology/13lonely.html?ex=1315800000&en=7eae0c5f86be8939&ei=5090) and Sunsilk embedded ad (http://www.cbc.ca/arts/media/story/2007/02/04/bridezilla-campaign.html) controversies.
 Michele White, The Body and The Screen: Theories of Internet Spectatorship (Cambridge: MIT Press, 2006) at 24 [White 2006].
 Hassan Masum & Yi-Chang Zheng, “Manifesto for the Reputation Society” (2004) 9:7 First Monday, online: First Monday http://www.firstmonday.org/issues/issue9_7/masum/index.html at 4.
 Sue Lees, Ruling Passions: Sexual Violence, Reputation and the Law (UK: Open University Press, 1997) at 17.
 See for example Robert Brym & Rhonda Lenton, Love Online: A Report on Digital Dating in Canada, Toronto 6 February 2001; Canadians and Online Dating, Leger Marketing Report, 9 August 2004.
 Lees, supra note 2 at 48.
 Cass Sunstein, “Group Judgments: Statistical Means, Deliberation , and Information Markets” (2005) 80 N.Y.U.L. Rev. 962 at 986.
 Howard Rheingold, Smart MOBs: The Next Social Revolution (Cambridge: Perseus, 2003) at 126.
 Michele White, "On the Internet, Everybody Worries that You're a Dog: The Gender Expectations and Beauty Ideals of Online Personals and Text-Based Chat" in Mary Rose WIlliams & Phil Backlund, eds. Readings in Gender Communication (Wadworth, 2003) at 286.
Of course for any academic there is also the involuntary online rating system of ratemyprofessors.com. Not only can you rate your professor, you can comment on them (including on appearance) and you can indicate whether they are "hot" or not. It is interesting that the creators and users of the site feel that physical attractiveness is a relevant consideration in evaluating a professor.
At Dalhousie Law School, where I teach, there was a lengthy debate about whether the results of university administered teaching evaluations should be disclosed as a matter of course to students. Some argued that these evaluative materials were protected from such disclosure by the collective agreement and by FOIPOP legislation. Concerns were raised that irrelevant and harmful gendered and race-based considerations might be reflected in the comments by students, and that the public dissemination of the comments would amplify the impact of these comments in a context where the professor was powerless to respond. The concern was over the impact of such 'ratings' on reputation. Of course, it was pointed out that with ratemyprofessors.com being available to all, it might be preferable to have officially administered evaluations filled out by all students (whether they had a chip on their shoulder or not) made public. This would give a better picture than a handful of entries that might reflect a combination of sincere attempts at evaluation intermingled with reviews motivated by lust, spite, sexism, or a grudge over a poor grade. Nevertheless, the actual student evaluations at Dal remain confidential; while the ratemyprof data is open to the world.
As for our reputations among our peers -- it has been widely speculated that professors are busy amplifying the importance of their scholarly works by going online and requesting download after download from SSRN and other such databases of works. This ties into Patrick Derby's post on authenticity online. No doubt online reputations should be prey to authenticity concerns. Why be boring, disorganized and frumpy when ten good friends can make you scintillating, witty, insightful and incredibly, unbearably hot?
Posted by: Teresa Scassa at March 8, 2007 10:48 AM