The cross-disciplinary road to true computational science

Editor in Chief In the first Issue of JoCS Ed Seidel and Jeannette Wing discussed he notion of Computational Thinking. “Computational thinking rings a new dimension to traditional modeling and simulation asks: rather than just numerical models, symbolic models, in the orm of algorithms, abstract data types, abstract state machines, nd high-level computational languages can serve as the basis of odeling complex systems. Moreover, in turning these models, umeric and symbolic, into something scientists can manipulate nd reason about, one needs to develop software environments rom the computer system level to the application.” [1]. In my vision one of the major aspects of Computational Thinking s the level of abstraction we choose as the starting point of our escription of the model. Often the majority of the time I spend with y PhD students, when discussing a new phenomenon to model, s dedicated to finding that abstraction level. Aspects of multi-scale spatial and temporal), as well as the availability of reliable data lay a decisive role. Once that is done, the whole process of finding he right computational vehicle (PDE’s, Particle Based, etc.), and the ssociated mapping to the architecture, seems almost trivial. This odeling step is the most creative step in the process, it requires a eep insight in the phenomenon at hand and in the computational tructure of the problem. Here we find ourselves talking and thinking as biologists, engieers, economists, etc. while reformulating the domain knowledge nto computable structures. A true cross-disciplinary approach to cience indeed! In this third issue of JoCS we find nice examples of this creative spect of computational science. We are proud to open the issue ith a wonderful paper by Alex Vespignani and others. They show ow to integrate sociodemographic and population mobility inforation into a stochastic disease model to simulate the spread of orldwide epidemics. Noura Beji et al., couple particle swarm optimization and local earch algorithms. Their paper made it into JoCS because of the riginality and the general applicability of the hybrid algorithm.