Hi Rick. First thanks for the reply...

>
>I think that for certain types of applications (e-Commerce, Web backends) .NET will make a good platform once it ships and once it proves stable. However, I'm very doubtful of the Windows Form engine being fast enough and usable enough in its first incarnation. It's a clunker in its current state. That entire architecture is based on the VB form designer based on what ca. 1990 technology <s>... I'm sure that won't stop the VB crowd, he he...
>


This is one of the things that gets me. Right now the app is probably about 20-30% C++ and 70% VFP. But granted, I haven't had time to dig too deep yet, so those numbers could be off. To just discount FoxPro totally in the re-engineering app is bordering on foolish. From what I've seen so far, the app is primarily run on a desktop and in some cases, multi-user with the tables on the server. I haven't heard anything as of yet that is planned for the new version that can't be handled with VFP 6 or VFP 7.


<snip>
>COM will not die anytime soon. I suspect a very large number of developers will not make the switch to .NET due to the complexity. COM works well in many situations and many applications don't even need COM especially desktop apps. I can guarantee you that desktop apps will run faster using VFP or VB6 than they will under .NET at least in this first rev of .Net.
>


Again, until I get more information on what the average number of users per site is, I'm not going to be sure on this, but I suspect that converting to a total .NET, SQLServer,n-tier, rework may be a bit overkill. At least for some of the customers. I'm not saying the decisions that are being made are wrong, not at all, just that, have they thought everything through or are they just running scared from FoxPro because of what they've been told.


<snip>

>COM underlies alot of Windwos and will continue to do so in the future regardless of .NET. But whereas COM has been made easier with VB/VFP and other high level languages, I htink COM will slide back into the operating system layer at the C++ level... At the same time that means that anything that uses COM today (VFP/VB6) will continue to work into the future and you don't have to worry about that not working all of a sudden. The question becomes to a degree what features to you need and do you really need .NET and what it provides right now?
>


I think that the features of the main concern here is the development interface and maintainability. As I think I mentioned in one of my other replies, the code is mainly legacy FoxPro/VFP 3 being maintained in VFP 6 with C++ routines doing some of the main crunching. Since none of the advantages or functionality of VFP 5/6 have been really implemented, they truely don't realize what the Fox is capable of. The old code is like a bowl of "spagetti", but I've seen that before. It's the old way: lots of public variables, everything in prgs, etc. That's how things evolved and it was "if it ain't broke..." philosophy for good or bad. The product seems solid and they have a lot of customers using it. Unfortunately, from all gatherings so far, they have already decided that the Fox version is not going to be redesigned. We just have to maintain what is there and do our best to clean it up until the reengineering is completed. What is going to be gained from going to .NET? I have no idea right now. Do they need to? Probably not, but I don't know. Are they leaving Fox behind just because of bad information? Yep. I would bet on it.

Who knows, maybe they'll do a mock up and testing in .NET and won't like what they see. Maybe they will come back to the VFP'ers and give us a bit more freedom to improve what they've got. Maybe I'll win the lottery. Time will tell...


I'm going to go install VFP 7 now... see ya all tomorrow :)


- Brian