Stack compatability

Given a high level programming language abstraction of stack machine instruction set, stack compatibility means that the operations passed to functions and the values they return correspond to values directly placed on top of the stack in the same order as they were put in. In the JVM user defined functions can only return a single value so stack compatibility and the principle of consistency dictate that multiple valued functions should not be used by the programmer, so operations like dup, dup2, dup_x1, dup2_x1, dup_x2, and dup2_x2 should be handled by the compiler and not the programmer. Since these multivalued functions are the stack manipulation operations, this means that stack manipulation should be left to the compiler and there should only be a higher level expression based language, whose compiler automatically handles stack maintenance. In this way, operations will all be consistent with user defined methods.

To make things consistent, the this argument passed to a function through the stack can be a parameter rather then a special keyword. This detail is forgotten by some high level programmers that use Java without using the bytecode. A form of this sort (.method this a b c) actually maintains stack consistency because this is passed onto the stack first, so it maintains stack compatibility with the bytecode produced. Stack compatibility will be make compilation between the high level language and the underlying stack machine rather seamless, allowing the programmer full access to underlying system.