Sign languages are visual languages used in deaf communities. They are essentially tempo-spatial languages: signs are made of manual components, e.g., the hand movements, and non-manual components, e.g., facial expressions. The e-LIS project aims at the creation of the first web bidirectional dictionary for Italian sign language-verbal Italian. Whereas the lexicographic order is a standard and 'natural' way of ordering hence retrieving words in Italian dictionaries, there is nothing similar for Italian sign language dictionaries. Stokoe-based notations have been successfully employed for decomposing and ordering signs in paper dictionaries for Italian sign language; but consulting the dictionaries requires knowing the adopted Stokoe-based notation, which is not as easy-to-remember and well-known as Italian alphabet is. Users of a web dictionary cannot be expected to be expert of this. There the role of ontologies comes into play. The ontology presented in this paper analyses and relates the formational components of a sign; in some sense, the ontology allows us to 'enrich' the e-LIS dictionary with expert information concerning classes of sign components and, above all, their mutual relations. We conclude this paper with several open questions at the intersection of knowledge representation and reasoning, semantic web, sign and computational linguistics.