>>YMMV, but to me reality (Public VFP sourcecode) is this mdot is providing a false sense of safety.
>
>The causes for the whole mess can be laid on Ashton-Tate, who wanted to simplify syntax for the end user, and since they used the clumsy -> as a separator between the alias and field, they wanted to help by allowing the omission of the alias for fields in the current workarea. So you could have a variable and a field with the same name, which is then ambiguous in expressions. To solve that, they simply said "we look at field first". Which was kind of easy to do when you had only two workareas, or even ten later, still easy.
>
>Then dr Dave wanted to be fully dBase compatible, at least the first few years, so he kept that. (Worse, he kept all those places where you can use a string literal without quotes. Then when you wanted to actually use a variable in those places they had to invent name expressions, which we, IIRC, still have to explain from time to time to those who are still confused with it. But that would be a separate rant.)
>
>There were ways to fix that - make the alias mandatory, for example. Or require a special prefix or suffix for unaliased fields (like some dialects of basic have a $ for string variables). But the wild success of Fox made this impossible - any such change would crash thousands of apps out there, and with each new version it was growing and made more impossible. The mdot was a praiseworthy option added so that it breaks nothing, and helps speed up things when needed - but it still doesn't change the ambiguity created by Ashton-Tate in the early eighties. It's still possible to shoot your foot.

Hi dragan,

Yep I totally agree with this.