15 Years of triple graph grammars: Research challenges, new contributions, open problems

Abstract

Triple graph grammars (TGGs) have been invented 15 years ago as a formalism for the declarative specification of bidirectional graph-to-graph translations. In this paper we present a list of still open problems concerning the interpretation and the expressiveness of TGGs. We will comment on extensions proposed to improve the original approach and the drawbacks that arise thereof. Consequently a more precise formalization of compulsory properties of the translation of triple graph grammars into forward and backward graph translation functions is given. Regarding these properties an interpretation and implementation of negative application conditions is derived that does not destroy the benefits of the original approach. Additionally a new demand-driven forward/backward translation rule application strategy is proposed. It guarantees for the first time automatically a correct ordering of rule applications without imposing any additional requirements on the structure of the regarded graphs. © 2008 Springer-Verlag Berlin Heidelberg.