In this paper we report the main ideas of an ongoing thread of research that aims at exploiting reversibility mechanisms to define programming abstractions for dependable distributed systems. In particular, we discuss the issues posed by concurrency in the definition of controlled forms of reversibility. We also discuss the need of introducing compensations to deal with irreversible actions and to avoid to repeat past errors. © 2013 Springer-Verlag Berlin Heidelberg.
Lanese, I., Mezzina, C. A., & Stefani, J. B. (2013). Controlled reversibility and compensations. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7581 LNCS, pp. 233–240). Springer Verlag. https://doi.org/10.1007/978-3-642-36315-3_19