An e-lesson is comprised of a "body" and a "view". The body is the actual content of the e-lesson and the assumption is that it is an html document. The view is the metadata section of this html document, and it is made up of a set of words and phrases that are either explicit or implicit and are referred to as "terms". This paper proposes a methodology that is capable of discovering the physical and logical structure of a given e-lesson through the use of that elesson's view and resulting in a major change concerning the granularity of data retrieval by the elesson's retrieval system. The major change is that the proposed method allows for solely the requested specific segment of an e-lesson to be returned instead of the entire lesson. The methodology is based on a warehouse-oriented framework that has been suggested by the author in the past.