>>The people who seem to be screaming the loudest are VFP developers.

Those who've never experienced close data integration can't be expected to "scream" for it. Sort of an inverse "you don't know what you've got till it's gone".

But what do you think yourself? Aren't you a dotNET fan with VFP background? Did you ever munge data locally in VFP? Don't you see any benefit from better data management in dotNET- I mean you personally, not other dotNET people you speak with?

BTW, my read of Anders' reply is that he came at it from a curious angle. His position that dotNET won't "specifically" borrow from VFP seems to boil down to VFP's exclusion from the CLR, that 4GLs were never widely accepted because they couldn't do some unstated stuff that programmers want, and to a definition of data as being from a database which he says is not necessarily true in C# but apparently is for VFP.

Exclusion from the CLR is a truism- of course C# isn't going to move away from the CLR. That can go unstated. Even if it is true that 4GLs were never widely accepted, that doesn't mean there aren't valuable components worth cherry-picking - in fact, in his response Anders seems to suggest that 4GL data techniques do have value. And the issue of whether data comes from XML or a traditional database is a complete red herring IMHO since VFP's strength comes from close binding of language and data in persistent autospanning internal cursors, wherever the data came from.

What does matter in the context of the interviewer's question is that we've been told by credible MS people right here in UT that some of the concepts we took for granted in FP, such as persistent autospanning local datasets, are a priority for dotNET. Anders himself said there were plans for better data management in C# and that VFP has already been where he is wanting to go. I don't really care whether that is politicized as "borrowing from VFP" but clearly others are keen to avoid any such suggestion... but who cares as long as it happens, right?!