Semantic Infrastructures and the Privatization of Meaning

Abstract

We explain how the Semantic Web operates and more specifically, how content producers are able to “discretize” information so that software such as search engines are then able to correlate it at Web level by generating ontologies. We compare the three types of syntax that can be used: the Resource Description Framework, microdata and microformats, together with the standardisation arenas and processes that are specific to each of these syntaxes. Through our analysis based on the operational theory of digital writing developed by the Universite Technologique de Compiegne, we will describe the role of the various “politics of meaning” in force on the Internet, each championed by stakeholders with different objectives and interests. We will decode the implications of the data description and processing scheme and, beyond that, the interactions between the person who describes and edits the information and the machine that processes and editorialises the data.