Rick and Mark,

Here you display some of the reasons that I made this proposal!

And I must say I've never been comfortable with the concept of issuing a command, having it *appear* to have executed it function, only to find out (indirectly) that it did *NOTHING*.

I still *suggest* that if release released as commanded, then it would be much much **EASIER** to find the "bug" that some later reference is being made in the code.

Cheers,

Jim N

>>If you want a good idea try it in C++ by calling the destructor directly
>>from within a class method <s>...
>
>Actually, calling a destructor explicitly won't do any real harm (depending on what the destructor actually does, of course). It's just code. It doesn't release memory (unless that's what you wrote inside it).
>
>However, calling
>
> delete this;
>
>can have effects (you can execute code after it, but should not access any properties or call any virtual methods). Nevertheless, it's the correct solution in some cases.
>
>>>Yeah, but you're not talking about external release!
>
>No, we're talking about a Release() method that simply does RELEASE THIS.
>
>It's too bad Steven Black isn't around to comment on this. He's the first guy I saw to suggest it, after extensive testing (and hair-pulling).