This is a text-based, coded information for the social sorting of individuals rather than images, but now that images and image databases are so prevalent, this grows to incorporate the image (Gates, 2011, p. 112).
According to Ericson and Haggerty (1997), Gandy's idea of the panoptic sort (through the panoptic machine), is that probability and statistics are produced remotely by machines without direct contact with the people they are sorting, and this information doesn't make sense outside of the risk analysis system. It is no longer a bricks and mortar collection of information, and the results, since they are wide and generalized, produce different results. They quote Gandy as saying, "On the basis of the "information revolution" not just the prison or factory, but the social totality, comes to function as the hierarchical and disciplinary Panoptic machine." They also call this the "difference machine" which deselects people "according to institutional values and access to knowledge environments" (p. 96).
Lyon (2009) says the point of Gandy's panoptic sort was to "discriminate between different kinds of customers in order to offer them different treatment" (p. 106).
According to Barnard-Wills (2012), Gandy's "‘Panoptic Sort’ demonstrated the process of ‘social sorting’ – the sorting of people into categories on the basis of surveillance data and using these categories in social, political and economic decision-making – involved complex discriminatory technology" (p. 29).
References:
Barnard-Wills, D. (2012). Surveillance and identity: Discourse, subjectivity and the state. Burlington: Ashgate.
Gates, K A. (2011). Our biometric future: Facial recognition technology and the culture of surveillance. New York: New York UP.
Lyon, D. (2009). Identifying citizens: ID cards as surveillance. Malden: Polity.
Ericson, R. V., & Haggerty, K.D. (1997). The Risk Society. In R.V. Ericson & K.D. Haggerty (Eds.) Policing the risk society (pp. 81-130). U of Toronto: Toronto.