my opinion below this line. correct me if some of my supposed facts are wrong
__________________________

in response to Ugh's post:

1. Pure dumb clicks-- definitely ok imo. as stated, possible with existing hardware/approved software. i'd like to use this to summon 15 trees on the boomkins i'm leveling.

2. pre-programmed click offsets-- also ok imo. doesn't use any knowledge of the game interface at all, 1 button/click = 1 predefined action. basically the same as key cloning... you perform 1 action on your end, and the click is sent to each window. the same action on your end can have different results in the game. building on the "you pull 1 handle to move 5 levers" analogy that the blue gave... it's *basically* the same thing-- you pull 1 handle (click your mouse) and 5 levers are moved. in this case, those 5 levers are each in a very large array of levers, and you choose before-hand which lever from each array you want to move.
counter-argument for myself-- you can change key bindings in-game, you can't change "mouse position bindings".

3. Intelligent click locations based on in-game UI elements-- i can't really think of a reason why I would want to use this, but i'll comment anyway. i don't really have a problem with this. you still have a player behind the screen calling the shots, pulling levers. you pull one, 5 levers move. The big difference and the reason why it would be in the grey area is because in this case, you have a helper... you give him a description of the lever you want pulled in each array links the correct levers (instantly - he's a very fast helper). You are still only pulling one lever and only one lever is being pulled in each array, you know what each lever does... but you have a little helper who is running around linking the levers together. blizzard doesn't like little helpers :P


Here is the original quote from Vyn's post:
Think of a single key-press as a lever. You pull the lever, and something happens.
Think of multi-boxing as simply attaching 5 levers to a single handle. You're still only pulling one lever, it just affects more than one something.
Now, think of automation as a lever attached to a set of gears and pulleys. You pull the lever, and a whole slew of bits and bobs start working, gears whirring, pulleys spinning. You might pull a lever, but it sets a process in motion that would be impossible with an ordinary pull of the lever if those gears and pulleys were not in place. Automation can apply to a single character just as much as it could with multiple characters.
The point is that the 'something' that occurs spools out without direct human involvement aside from the initial pull of the lever. That is automation. Even if it's only a single extra step.
In multi-boxing, every action taken by those characters has its source in a human command. Each individual action. Thus, it is not automation