The main contribution of this work is to present a new software engineering model, the Adaptation Rule (AR) model for the description of adaptation rules in a service oriented environment that solves the previous issues, and its application to a hinting adaptive e-learning system. This new model has a strong focus on rules and enables the typical advantages of software engineering. In addition, the model enables the definition of atomic rules and the combination and reuse of atomic rules from different sources according to restrictions. The AR model provides a solution for several rule components working together that are flexible and extensible through the Web in line with the commented challenges previously cited in [4-5-6] and overcoming the rule design issues cited in [8].