Ken,

Thanks for responding. That did give me some ammo. However, I do have a follow-up question.

What sparked this was Kamel Patel's article in the UT magazine about being able to query ADO.NET datatables. Kamel presented a methodology of a VFP COM object that converts a datatable from XML to a cursor, then runs a query, then converts the resulting data back to XML.

The .NET reporting application we're trying to write does a TON of data-munging/grouping of raw result sets from SQL Server.

One of my co-workers, who more or less is our resident expert on server technology, etc., raises huge red flags anytime he hears about COM. In this instance, his reaction is 'you'll have to deal with all the scalability management issues, you'll have to use MTS, it isn't managed code, I don't want to have to manage all these COM objects and how often they might be hit, etc.'

Granted, there is a different method I could use (creating both the raw data AND grouped data as one result set in SQL Server, with a one-byte code for whether a row in the result set is detailed or grouped)....but it just seems cleaner to use Kamel's method.

Unfortunately, I'm more of programmer/developer than systems integrator/architecture-type person, so I don't have all the answers to his qeustions. Any further feedback you can give might help.

Thanks...
Kevin