Concerning the undecidability of the digital image

Abstract

This paper considers the significance of metadata in relation to the image economy of the web. Social practices such as keywording, tagging, rating and viewing increasingly influence the modes of navigation and hence the utility of images in online environments. To a user faced with an avalanche of images, metadata promises to make photographs machine-readable in order to mobilize new knowledge, in a continuation of the archival paradigm. At the same time, metadata enables new topologies of the image, new temporalities and multiplicities which present a challenge to historical models of representation. As photography becomes an encoded discourse, we suggest that the turning away from the visual towards the mathematical and the algorithmic establishes undecidability as a key property of the networked image.

Notes

1 The production of meticulous and semantically unambiguous metadata is crucial to the Semantic Web (“Web 3.0”), Tim Berners-Lee's vision of “a web of data that can be processed directly and indirectly by machines”.

2 The dynamics of tagging in Flickr Groups is taken up by Dr LopLop in “Somebody Else's Cat”.

3 By way of example, the simple process of logging into Flickr or Facebook will trigger the retrieval of multiple data streams (photofeeds, status updates contingent to a user, time or tag) which is glued together on the fly to form a webpage.

4 Alphonso Lingis wrote at length on the noise in the message and on the message of the noise: “Is it not also false to suppose that only the meaning attached to words by a code, fixed or evolving, communicates? The rhythm, the tone, the periodicity, the stammerings and the silences communicate. … This noise is not analytically decomposable, as communication theory would have it, into a multiplicity of signals, information-bits, that are irrelevant or that conflict” (105).