>After reading about all the bugs/problems with VFP8, I think that ms has made things difficult for everyone, including themselves, by adding features to vfp8 that have a complicated architecture, and thus some are unreliable from our perspective. Too many gotcha's. I would personally prefer more RISC-like enhancements to the product in terms of its programming language so I could do more low-level things (such as use _real_ data structures, better API integration, callback support, etc), and thus get at Windows core capabilities more easily. Does anybody else feel this way?

I must say that I don't feel that way at all.

While I do feel that the bugs exposed so far are discouraging, I'm reminded that they still number far less than we had with either VFP6 or VFP7, and certainly with far fewer C0000005 baddies!

Software is software, and it **may** be that the VFP Team is understaffed for the complexity of the product now, but that in no way means that it would be sensible to dumb-down the product.

The idea with computers is that they do more and more work for you, not less and less. Primarily for the end-user, but also for the developers/maintainers/testers too.

And I'd worry that allowing more and more "low level" stuff like you mention would actually aggravate the complexity factor - the community is well known for mis-interpreting (thus misusing?) features as they now stand.
Also, we have some things (SYS functions) that mainly only say 'designed for use by "advanced programmers" only' already and I'd sure like to know what their intended use is before we add more of them into the mix.