Improvements can still be made in the development of Interactive Computing Systems (ICSs) aiming to ease their use. This is particularly true when trying to address users’ diversity. Most ICSs do not adjust themselves to the user nor consider user’s particularities. However, some provide solutions to address better specificities of expert and novice users. Others adjust themselves based on user’s interaction history, but this does not always lead to improvements in use. An aspect that prevents to address users’ diversity broadly is the fact that most of existing ICSs do not provide source code access. This means that only owners can introduce improvements on them.This paper proposes an approach (based on both affective computing and computer vision) to broadly improve design for diversity (without source code access) for both existing and to be developed ICSs. The results are twofold: (i) example of an initial set of design guidelines; (ii) opens the way to runtime Graphical User Interface (GUI) redefinition and adjustment based on both user’s features and emotions reducing therefore designers’ restrictions when addressing users’ diversity.