We can identify cracks, talk about them, point to them. Cracks spread, travel, propagate. There is even a field of physics, fracture mechanics, that studies the underlying laws of their motion. There are cracks; cracks are. But cracks, we can very well say, are not existing things.
We can say similar things of holes, of gaps, of empty spaces, of blanks, of silences. There is an entirely understandable sense in which all of these are 'nothings'. But they also in some sense are. These paradoxical nothings you find in the real world are what we might call quasi-entitative non-entities, although, to be sure, we could also call them, more vaguely and loosely, 'things that in a way really aren't but in a way definitely are'.
The scholastics distinguished between thing-beings and reason-beings, and it is clear enough from the sense in which we call these things 'nothing', quasi-entative non-entities have to be reason-beings rather than thing-beings. A reason-being (ens rationis) is being that is only found in and of itself as an object for thought.
This is not necessarily to say that they are purely imaginary or made-up, which is a confusion that always arises when people talk about beings of reason. Reason-beings can really be; they just aren't real things in themselves. The hole currently in my ceiling as I wait for it eventually to get repaired is in my ceiling. You can see it. You can put your hand through it. You can fill it. If you don't take it into account, you will not understand certain things about the ceiling. But the hole is there only as negative space, only in the sense that the really existing ceiling stops and has non-meeting borders from several directions. The being it has as a hole consists in the fact that we think of it on the model of a real thing, that we take it as an object of our minds. A hole is in the thing insofar as the thing is (or could be) an object of our minds.
It's possible that thinking in terms of holes, cracks, etc., at least as such, is due to limitations of our minds. God knows things fully as they are, in and of themselves, but in navigating the world, we find that we can't know thing-beings, like ceilings or windows or stones, both directly and adequately. Much of what we know about real things we have to know by thinking of them on the model of other real things. And much of what we know about real things requires having a way of going beyond the limits of what we directly know about them.
Traditionally, the classes of reason-beings were negations and rational relations, although sometimes people preferred to divide the first class into bare negations and privations, and although you can even find some philosophers, like Thomas Compton Carleton, who argued for the (mostly unpopular and difficult to defend) opinion that bare negations were actually thing-beings. I think we have to be a little careful with this -- we should really think of these as being-by-negating and being-by-relating, and (as I will go on to suggest) take them to be combinable.
All quasi-entitative non-entities are 'quasi-entitative' because we think of them in terms of thing-beings. Take a hole in the wall. We can think of this as a hole in the wall in two ways. We can think of it as fillable. When we do this, we are thinking of the hole in terms of thing-beings (usually but not necessarily exclusively the normal components of the wall) that could be there but are not. If I walk into the room and find to my surprise that there is a hole in the wall, it makes sense that I would think of it as a not-there part-of-a-wall. Or we can think of it as traversable. For instance, we could move through it. If we do this, we are thinking of the hole in terms of its relation to thing-beings in its context -- the wall that makes its borders, the things that can move through it. In fact, we all have deliberately placed holes in our walls -- windows to let in light and perhaps air, doorways to let physical things in and out. Passageways are there, but a passageway is an emptiness understood as related to its context in a certain way.
I would suggest that all quasi-entitative non-entities have this feature: they are negative reason-beings that are also relative reason-beings. (I think you can go in the reverse direction and have primarily relative reason-beings that are also negative reason-beings; distinctions and partitions, I think, are such things.) They are not mere non-entity. Walking into a room and discovering a hole in the wall is not like walking into a room and discovering a lack of unicorns. There just aren't any unicorns, but there is a hole in the wall. I can't do anything directly with a lack of unicorns; I can do something to a hole, like put my hand through it. So holes are negative, but not merely negative; it's actually essential to their being holes that they also be understood as relative. The same is true, of course, of cracks, blanks, gaps. It may be a bit less obviously true of empty spaces and silences, but I don't think it's difficult to argue for them, either. It's probably also true of shadows, which as someone somewhere has said are like holes in light. In general, I think particular non-things will always be in this same family; mere negating leaves you with something indefinite, so you need relating to get a particular negation-of-thing. Thus evils, not evil as such but particular evils, are in this family: they are privations related to a context so that we can reason about those contexts more easily, in the way that we would do if they were particular examples of substances, qualities, or stuffs.