This paper presents the prototype of a platform aimed at curating simple open bricks for historical knowledge, shaped as short texts. These historical bricks, named “Pulses,” are meant to be based on no prefixed ontology, but collectively negotiated conventions and reformatting by bots. The Pulses curation platform that can be interpreted as hybridisation between some characteristics of Twitter and Wikipedia, organises a collective process of “smooth semantization” in which pulses are progressively rewritten to become more machine readable. The article reports on the principle and implementation of this system conducted in the last two years.