Just keep it a secret until released. Since code and mechanics are behind the scene, consider the possibilty of it being pirated, slightly modified and released. What are the odds you would be able to detect your mechanics in someone else's game, and the odds of you actually playing that game.

Even though you may take all the legal steps of protecting your mechanics, the government is not watching out for you. You would have to be the one who finds it in use then hire a legal team to fight it.

Then there is the "What are the odds" of 2 or more people coming up with the idea and who started working on it first.

The only way you can stop your idea from being used is to keep it a secret.

The unfortunate truth is that games, like most things, are 1% inspiration and 99% perspiration. The people who can build the idea are much more valuable than the person who comes up with ideas, so there's no value in the idea unless you can build it into a game yourself first.

If you're not capable of building it into a game by yourself, then there's not much point in keeping it secret, as it has no value. If you're in that situation you may as well think about how you can get some value out of it, such as using it to increase your reputation as a designer of innovating game mechanics, etc...

This is relevant: http://www.lostgarden.com/2005/08/why-you-should-share-your-game-designs.html

Regarding that article, I invented an original aiming control scheme for FPS games 10 years ago, which I still enjoy using more than the usual mouselook systems that haven't changed from Quake 1 to Call-of-Duty 9... and no one has ever copied it. Maybe I'm just biased and my system isn't actually that good, or perhaps gamers don't like change, or perhaps no game designer ever played my mod to steal it, but for whatever reason, I've never seen this original mechanic copied. So if, 10 years ago, I had simply blogged about my awesome FPS aiming, or I'd not published an implementation of it at all, the world would be pretty much the same place as it is now anyway, except that as is, I've got the experience of having shared my work

Maybe by discussing your idea before you release your game, you'll refine it into an even better idea

And hey, if EA comes out with "[First-Person RPG] 2013" in 6 months and has totally ripped off your idea, at least you can link to that discussion, and say "I was doing FPRPG controls like that before it was mainstream"

Maybe someone else already independently came up with the same idea, and are working on a game that uses it right now -- if you start the discussion about this idea now, then you get to claim that this other person just copied you, whereas if they release first, they'll accuse you of copying them!

Maybe the idea already exists in some obscure game from the 90's that no one played

If you want to make an idea public, but still restrict it from being used, you will need tens of thousands of dollars to draft the idea as a patent (via a reputable and experienced patent lawyer) and register it with all the world's patent offices. Even then, this doesn't stop other people from using the idea -- once they've used the idea, you then have to take them to court for patent infringement, at which point your patent will be tested (and possibly invalidated!), at a huge cost to you and/or the target. You'll also be labelled as a "patent troll" (someone who patents ideas, but doesn't build them, and then tries to profit from the work of others who do actually build things) and universally hated.

Also, it's fairly easy for someone to make some small modifications to your idea to work around the wording of a patent, so that they're not infringing (i.e. so that you lose your lawsuit against them).

As for the odds of me being able to tell if they copied it, I would know almost instantly if they cloned it.

So someone somewhere in the world posts a game (using some aspect of your unique concept) some obscure place on the internet. How are you going to know about it "almost instantly"?

Also for your aiming, I may be wrong as to how yours in done. But james bond golden eye for the N64 let you aim your gun independently of your view... it was a pain in the butt and you only needed it in multiplayer if someone used the really short dude.

Similar but very different In the same way, I'd bet when you release your idea, you'll also be told about old games that contain a similar scheme

In Goldeneye 64 you could hold R to switch to a mode where the thumbstick moved the crosshair instead of moving the view.

In Operation Flashpoint (the original, not the shameful Codemasters thievery) there was a 'deadzone' system, where the mouse only moved the crosshair until it hit the edge of an invisible box in the centre of the screen, at which point the camera would actually start to rotate.

MoveIn's aiming was more similar to OFP's system. The mouse directly moved the gun at all times in the exactly same manner as every Quake derivative game, however you had no control directly over the camera/view. Instead of drawing the crosshair in the centre of the screen, a ray was traced out of the gun and the crosshair placed where the gun was pointing. The player's view itself was on a spring that would follow the gun's aiming direction. The speed that the view would rotate to catch up with the gun was dependent on how far the crosshair was away from the centre of the screen. If you were moving your aim between different targets in the middleish of your screen, then the view would remain mostly static like an arcade shooter (virtua cop, house of the dead, goldeneye when holding R, etc) but when making larger movements the view would behave more like a typical FPS, with a continuum in-between.

[edit]And yes, many regard Oddjob as a cheat character in Goldeneye due to the difficulty in aiming at him!