David,

Compared to for instance many changes in VFP8, I think the change I and other want, is minimal and only a good one. What about the SQL change which made it necessary to include the SET ENGINEBEHAVIOR ON/OFF command? And the bindevent/_screen.resize/_screen hook issue where the behavior was changed in such a way that the actual example in help for VFP7 caused a C5 error in VFP8?

As far as I can remember, the issue we discuss here, is the only situation were VFP does not raise an error were the situation often, but not always, should logically raise an error. A new SET OPENDIALOG ON/OFF command and a minimal change in the not default behavior would make everone happy.

I have attended 9 devcons in USA and in Prague the last 9 years, and all the new versions of VFP have been presented on many of these devcons. For many of the enhancements and new features there was load cheers, although many of the new stuff was not necessary if we follow your reasoning. With all due respect, I think your attitude is counterproductive.

I really have trouble seeing this differently.

>Tore,
>
>Since it is a defined behavior of the language, you can believe that there are some poeople out there that do use it. It doesn't really matter whether or not you or I think it's a good interface.
>
>>Come on, do you honestly believe anyone would use anything else than a getfile() or at least a question mark to do this? I think your example shows that you haven't really understood the problem. Or you simply are not willing to understand it. Or you are just too arro.. (no, I will not lower myself to personal attacks, I have too much respect for you for that.)