An innovative approach to associative indexing is explored in this collaborative database tool, called nic-las in homage to the great late sociologist, developed by a Swiss team of programmers. Billed as a “nowledge integrating communication-based labeling and access system” (sic - acronym: nic-las) or a “software prototype of an autopoietic knowledge landscape for social systems,” it is basically a cooperative digital space for research groups, made up of textual components and java objects. Shielded and organized by a personalizable multi-user access portal, each team can decide to what extent their collaboration is visible also to outsiders, and to what extent their notes, citations, exchanges, and other documents are made available to search engines. Anonymous use is possible at least in principle, but experience has shown that the thirty or so research collaboratives currently using nic-las tend to express themselves in the idiosyncratic ways of a typical academic gathering, with concerns over attribution, credit, and accreditation still extant. New entries or modifications of existing entries are recognized and dynamically linked to relevant other notes in the system. An intriguing feature is that deleted elements end up, for a while, in a digital unconscious; they remain accessible to certain search operations, and can even return in unforeseen ways. The system distinguishes between a Freudian and a Deleuzian unconscious; while the former pushes some deleted objects back onto the documentation surface, the latter generates a random selection of deleted and undeleted objects in the form of new virtual index cards.