Some very good points indeed.

As I see it, there are two compelling reasons that would drive one to use another tool today:

1. The need for better ActiveX Control support
2. The need to have scaleable components that work with MTS

On a longer term basis, if you need to be sure there is support for Windows DNA, then a move may be warranted.

Barring anything else, VFP is still a great tool to use to develop apps and I think will have life for some time to come.