Still working with tables. But conversion to SQL-Server (MSDE for now) has already begun.
Why?

Martín picked up on part of where I was going, so maybe now you can see why I asked. Suffice it to say you've evaluated pros and cons, and the direction you chose reflects that. I respect that.

The other part is this: you said you didn't want to waste your time on unfinished business. You also made references to product maturity. It seems you have one standard (pragmatic) of weighing pros and cons for tools you've used, and other standard of acceptance for those you haven't. That's where I see things differently.

Consider this...FoxPro has had an implementation of SQL for years - yet it took VFP9 to come close to supporting the SQL-92 standard. Until V9, many queries found in Joe Celko's books would not run in VFP. When I worked in environments where VFP was the back-end, I was discouraged that I couldn't use different techniques - but I found workarounds, didn't refer to the product as "unfinished business", and moved on.

It doesn't take stones to refer to .NET as "unfinished business", ".NOT", ".NYET", or any of the silly posts by the UT member who has the term "marquee buffet speakers" on the brain 24-7. It demonstrates something very different: a level of intellectual dishonesty, and a lack of the very analytical approach our industry demands of us.

Sometimes the best way for one to truly sound off like they've got a pair is to bear down and use some elbow grease to get the job done. If you wait for the "perfect product", you'll be waiting forever. The term "unfinished business" is a pretty weighty term. I can honestly say I'm a much better developer than I was 2 years ago, and not just because of natural evolution - it's also because of exposure to a new technology. .NET is far from perfect - but it can do many things, and it IS, in the words of college basketball icon Dick Vitale, a PTP (Prime Time Player).

Kevin