A system and method for characterizing, selecting, ordering and rendering discrete elements of digitized video content helpful to learn or teach various skills. Each of a plurality of observed or computer-generated instances of modeling of distinguishable skills are recorded as digitized assets. Microskills are identified and deconstructed in the abstract from one or more of the visual and audible recordings of skills modeling moments. Identifiers of microskills are associated by a human editor with recorded modeling instances and/or portions thereof. Modeling presentations are dynamically generated by a user asserting one or more microskill identifiers and a network-enabled selection, ordering and rendering of portions of modeling instances that are associated with the asserted microskill identifiers.