English

This article focuses on how to process the symbolic or the meaning of (digital) media data which the author considers as a potential cognitive resource. In his opinion, this data is to “undergo” more or less significant qualitative transformations in order to become relevant for a user or community of users. These qualitative transformations are performed through series of concrete operations such as the constitution and classification of relevant corpora of digital records, the description and indexing of records, the processing of digital data (segmentation, tagging, linking, montage, …), the (cultural, linguistic) versioning (commenting, translating, …) of given source records or again the (re-)publishing of digital records. These operations constitute what can be called the semiotic processing of (digital) media objects, corpora of (digital) media objects or again entire archives and libraries. They demonstrate practically and theoretically the well-known “from data to meta-data” or the “from (simple) information to (relevant) knowledge” problem – problem that obviously determines the effective use and also the future of digital archival knowledge.