The Watchmaker: my fav argument

[John 6IX Breezy]

Obviously we can be mistaken about these qualities. (Affordances?) I might perceive a gap in the wall as a place to walk through, when in fact it's a large window with clean glass, and I end up bumping my nose. So the ____-ability of something would be falsifiable. 

Yes, that is the basic premise. I do want to add, since you emphasized use, that an object's use is a reflection of my ability to use it. And it is this ability to do something that I'm focusing on for the purposes of testability. If I can grasp something, it is graspable. If I can design something, it is designable. Design is very much a product, the result of activity, which can be emulated.

So what we perceive as designable will be predetermined by the metaphysical assumptions we have about the world. There's a danger of begging the question. 

Yes, but under my specific definition (which could be nearsighted or wrong) we can test it out. It is less about what we perceive and more about what we can do. We can discover what is designable by simply designing it. It is this attempt at replication that constrains designability, just like graspability is constrained by our inability to grasp a wall.

There could be randomness built in, in which some things are allowed to evolve randomly, while others are made intentionally. 

I agree; but with randomness specifically it can still be simulated. Scientists often create computer programs in which randomness and evolution are allowed to occur, so that they can measure the end result of competition. I've seen similar simulations of the cosmos where gravity is given to pixels on a desktop, and are set in motion. The end result is similar to orbits in solar systems and galaxies.

Or God could design the system and the natural laws, and then put them in motion so they evolve in a pre-determined way. Would that be considered designed or not designed?

I think yes, it would be considered designed, but two questions need to be answered first to determine that:

1. Is assimilation sufficient to infer designability? With events like randomness and the butterfly effect, it might be impossible to recreate the exact outcome; but if we could set similar laws in motion and end up with a hundred similar universes, is that similarity sufficient to infer designability?

2. And specifically, is simulation a valid form of testing? This is important because we cannot create matter, but we can simulate it. Is that enough?