I think the better strategy is to specialize in topics that are long living. For example, specializing into parallel programming, or virtualization, or APIs. Th

I think the better strategy is to specialize in topics that are long living. For example, specializing into parallel programming, or virtualization, or APIs. Those topics are less likely to come and go than a library, product or technology building on top of them. In your case for example you seem to be specialized in the Microsoft programming ecosystem, which is probably a safe bet.