Code Generation for GUI vs WYSIWYG design of GUI

Advantages for Code Generation are:

• can generate for several target frameworks, perhaps even Desktop Application and Web Application
• can fix one problem centrally, and regenerate all screens with the fix
• unique look&feel across the application
• no problems with visual inheritance etc

Advantages for designing the GUI with a WYSIWYG editor are:

• the developer can see what he/she is doing
• easier connection of events and methods to the buttons and other controls

Option to use advantages of both:

• the WYSIWIG editor stores the layout in way that can be picked up by the generator, and then the generator can keep the manual settings
• both the generator and the WYSIWIG editor can use the same data format to define a screen, eg. Glade, or even the code to create the controls (eg InitializeComponent)

In the end the solution might look like this:

• screens that always look the same, are generated
• as much as possible, complex parts of a screen are wrapped into a user control, which can be edited by a WYSIWIG editor
• the window frames and menu and toolbars are generated