>I have the impression that the 'error' in question is a catch-all for several errors, not all of them memory address related.
This has been stated by MS. I have a feeling that the error comes from the VFP teams effort at 'internalizing' many errors that used to result in a Dr. Watson. (If you watch in VFP 5.0, the error code always given with a Dr. Watson is- C0000005.)
Remember Calvin Hsia showing off some new VFP stuff at the Goodies night at DevCon? He stated that many errors that used to cause GPF were now 'caught' by VFP and translated into a normal error message. I have yet to see an error message likethe one that Calvin demonstrated with his SYS() function that caused a GPF. Instead, we are seeing C0000005. A good effort gone bad.
It is a bad thing for MS popularity as well, giving all of these errors one hideous name and hiding the cause from users- it makes it look like there is one big flaw with VFP 6 and its name is C0000005 ("why can't they just fix the bad part of the program?"), when in fact this is not the case.
Just my .02
Erik Moore
Clientelligence