>SNIP
>>
>>Jim, perhaps you can clarify the scope of your wish. However, I still don't believe VFP should have any say in the physical layer, even at the local level.
>
>It is NOT 'VFP having a say in the physical layer' at all, local or remote.
>
>It is simply the OS (not any "physical layer") not returning to VFP until it confirms that the FLUSH has indeed completed. The OS will do whatever it does with the physical layer and when it and the HD have finished their 'talk' then the OS can return to VFP to let it continue processing.
>

Let's just say you and I are defining "disk" differently. I view it as a disk subsystem...controller, cache on the hard drive, and the physical magnetic media are all part of that. If I understand you correctly, you want to know when the surface has been properly magnetized. I think you also want to be able to force that to happen.

>As I said earlier, this is a basic tenet of computing and it should indeed happen on every I/O. Since there is a Microsoft document that states flatly that the FLUSH command does NOT operate this way in the more modern OS' (note, this mens that it DID in older ones) then I want the option to have it go back to doing so.
>
< SNIP >
>Does this clarify it? If not, let me ask you this...
>
>Would you find it acceptable if random records in your networked application just disappeared, or they over-wrote other existing records randomly? Of course not. So, then, how do you think that the system(s) involved protect you from those things?
>

Again, I expect the disk subsystem, as a whole, to work correctly.

Jim, thanks for making me think about these things. If I have time, I'll look at your other thread.