>>Yes, those techniques still work. I'm just saying that by using those techniques I didn't have a problem related to specifying results in my VFP7 groups.
>
>
>Gotcha.
>
>You know, I made a lot of noise about this when I started working in SQL Server, but I quickly adapted, and these days, it drives me a little crazy to see a statement in VFP that *doesn't* comply with the "include all fields" rule. it just seems sloppy to me to let VFP grab a value seemingly at random (at least max() makes it clear that you want *some* value and are willing to tell the compiler what it should be, even if, to you, it's somewhat arbitrary.
>
>Of course, that doesn't help with legacy code....

Oh - I agree whole-heartedly! That's why I sauntered off to the bookstore to find a good TSql book and rummaged through it to figure out how to properly do those types of queries. I was gratified to find WITH CUBE and WITH ROLLUP - but then when I got home and looked for those statements in the new VFP SQL syntax I found I was hosed!