>Regarding using .NET.
>All new applications are written in .NET.
>I wouldn't convert a VFP application to .NET for the sake of converting it, but with each change to a VFP application we ask the question "Can we do this change more efficiently if we convert the app to .NET?" and proceed accordingly.
>We recently converted a VFP6 application to VFP9 because that platform gave us what we wanted at a fraction of the cost of converting it to .NET.
>
>Finally- I think I that one responsibility to our clients is to stay healthy as a business.
>I think we do that best by maintaining a solid technical arsenal that we can offer existing and potential clients.


All excellent points. We've got a VFP product with a codebase of 20 years old. Granted a lot of code has been rewritten, but VFP gave us a hell of a ride and still does.
If I would have the chance to start over again with the existing tools of today, would I have written it in VFP? Definately not. I'd probably have choosen Windev, Lianja or Servoy. With .NET I'd probably would run into the same problem as with VFP: I still have to write my own framework and as I have learned over the last two decades: Frameworks never are bugfree or finished.

As for the point where we are now: a 20 year old VFP product, we are gradually moving towards a webbased solution. First we clean up the current codebase and make it true n-tier where biz_objects are seperated from the GUI. If that is done we can build the web applications on top of that.

But since there still is a lot of new development going one, even more than ever, we cannot afford to take two years to rewrite something without adding new functionality. So we have to walk and whistle.