>One of our infrastructure on a client site includes a .NET Web Service which uses VFP and WWC for the data tier. Today, a top IT in regards to the management of that infrastructure says that this solution is not scalable. They said to us that a global .NET solution is scalable. How do we react to that?

Michel,

I think that you should concentrate to not defending the VFP, but rather attacking the architecture long time associated with VFP, which is client-server architecture with fat/GUI client and file-server access to DBFs. Then show that VFP middle-tier components could be a great player in web-based distrubuted system running against the true transactional databases (like SQLServer). The VFP database itself is a no-go in any serious enterpise projects, but the VFP-based components are great.

Try to push the idea of language-nuetral development, where components written in various languages communicate with each other using HTTP (not DCOM, not .NET remoting, not CORBA, etc.), doing their business in the middle tier, generating universal HTML-based presentation and running against major transactional databases like SQL Server or Oracle (bot not VFP database).

This works for me, people don't make a big deal about VFP in this architecture - they use it together with C#, PHP, VB.NET in the same project and nobody cares about the "language" aspect.

The "language" aspect was important 10 years ago, when if started with, lets say VB, you should continue with VB to maintain your huge EXE file. But not anymore - with universal presentation (HTML), universal server-to-server communication (HTTP) and universal database (the transactional one, please) - you really have more freedom for language to choose and the cost of [language choice] mistake is minimal this way. If you educate your clients about this approach, I am sure they would be more open minded about the VFP.

Igor.