Nowadays, mobile devices host many applications that are directly downloaded and installed from a store. The existence of such a large amount of apps for a myriad of purposes imposes a huge overhead on users, who are in charge of selecting, installing, executing the appropriate apps and deleting them when no longer needed. These applications have mostly neglected to take into account the user's context by proposing static non evolving scenarios of use. In order to address these concerns, comes the idea of a single application constantly executed on the mobile device while automatically handling software components according to the users' needs. Long Life Application is new way to respond to the user's needs in a dynamic and intelligent way. It evolves at run time (include/exclude business functionalities, update interactions mode, migrate executions) according to the user's need. While he/she moves in his/her surroundings, the app understands the occurring events and builds contextually described situations. Thus, context awareness and contextual formulation are the main issues discussed in this paper.