Re: Hyphenated Op-Codes

James,

> I guess what I'm wondering is, what really distinguishes a BIF from just a
> normal function?  And is that difference so important and wonderful that
> it should be kept?

At least three things:

 1) The % prefix keeps from having a name conflict with your own functions
 2) You don't need to have built-in functions in a binding directory
 3) IBM supplied routines get the advantage of knowing numeric argument types

That is, we have operational descriptors for character fields but not
numerics, so we can't easily implement things like %editc() although
the compiler knows variable declarations.

It is the % prefix which solves the same issue as the proposed hyphen
in future opcodes.

Doug