Computer programming is experiencing a paradigm shift driven by the advent of powerful large language model–based tools for automatic source code generation. This transformation is also reshaping introductory programming courses worldwide, raising critical questions on how programming should be taught, learned, and assessed in the era of generative artificial intelligence.This article pursues two objectives. First, it reviews the main studies addressing this issue, emphasizing the reported advantages and limitations. Second, it proposes enhancements to teaching and learning methodologies by prioritizing code comprehension and execution over code writing or functionality. Specifically, the use of visual code representations and execution simulations is proposed as an effective approach for teaching, learning, and assessing programming, ensuring a deeper conceptual understanding among students.Finally, the article reports preliminary insights from students enrolled in an object-oriented programming course, providing contextual evidence to support the integration of visual simulations in Java (and other programming languages) within programming education.

This work presents the definition of a water risk index based on temperatures and precipitation during the years 1979 to 2019 in the Ñuble Region (Chile). Its calculation has been implemented in a software application that allows working with multidimensional data files in NetCDF format. In addition, our prototype allows to load and visualize this type of data in an easy way without the need of having an advanced knowledge in the management of geographic information systems. The result is an application that we would like to use in the future to improve water management and decision making in the region.