>Have you ever migrated/ported a DOS to VFP enterprise-level application?

I am currently porting an FP 1.x application to VFP 6.0. I started getting into using the OOP facilities in earnest after some prodding from Jim Booth.

My app does not do much more than order entry, invoice and check creation, and production of various types of reports (all very simple from the perspective of what FP or VFP is capable of). In fact, my whole app would probably fall into the UI part of the upgraded application you propose (IOW, no backend to speak of). So I frankly cannot speak to whether or not your backend/frontend approach wouldn't be a better short term solution to your situation.

Part of that answer has to be how integrated the backend/frontend stuff is in your DOS version.

>
>If you have, did you later rewrite the application to be fully OOP and n-tiered in VFP 5.0 or VFP 6.0?

We talk about doing this...wistfully. I doubt if it will happen because
(a) what we have works...so why rewrite it?
(b) this is a small company that will not be recouping the 'R&D' cost of further development with increased operational efficiency or sales of the product being developed.

My situation may be different from yours because I came into my project knowing zip about database development. (I mean, totally zip no FP, no VFP, no VB, no DBASE ...no nuthin). The only reason this little venture was economical for my friend's company is that I agreed to pretty much work for peanuts because I knew I couldn't be anywhere nearly as efficient as a professional developer.

In retrospect, I think what has taken me about 4 month's work could have probably been accomplished by my present self in 6 weeks or less. So I'm pretty sure a whiz could have even done it faster.

If my example is any indication of how you should proceed, my recommendation would be to:

1) use your expertise and knowledge of your current app to write a really detailed spec for the new app

2) find a VFP guru who'll give you a fixed price to implement it. You don't pay him/her until it meets spec.

The downside is that you cannot prove apriori that this will cost less than shoehorning your 2.x app into VFP...nor will you be able to prove it aposteriori. Bosses hate this.

The upside is, after you have the spec, you should be able to get a very good estimate of the bottom-line for project cost. Bosses love this.

It sounds like you've already done some of the preliminary work to generate a spec for your app. Why not put it up for bids to a couple of private consultants and present your findings to your boss as a fait-acompli?

After you trigger the guru, you can take some company time to learn about VFP/OOP in an organized way (instead of the haphazard way you will be forced to learn it to fight each technical fire as it arises). Then you'll be ready to preside over maintenance and further incremental development of your new app with a professionally designed model to act as a guide.