I'm in the process of converting parts of an application to use ASP.NET MVC from WebForms. The idea is that when possible I use MVC for new development. My problem is this:
Doing it this way means that my Models are not completely implemented in MVC. Let's say that my WebForms application has a robust Widget Management page. I'm adding new functionality to attach a color to a Widget. Click a widget, and it brings you to a new page to select a color, and that's all. One field. It seems silly to create a WidgetColorPreference Model with two properties: WidgetID and Color, for the sole purpose of validating and saving.
So, I have gone down the road of implementing a Service namespace with methods such as "SetWidgetColor", accepting a WidgetID and Color parameter. My controller action simply calls that service in my Models assembly.
My question: Is this a valid stopgap architecture?