I'm trying to pick layers apart and I'm getting a little confused about how to apply them. Consider the following two rulings:

Humility + Opalescence, trying to determine what Humility is now.
You end up with
layer 4 (type): Opalescence turns humility into a creature
Layer 6 (ability): Humility removes its own ability (Because it is now a creature)
Layer 7b (p/t setting): Humility makes itself 1/1, Opalescence makes humility 4/4, end result depends on timestamp order.

Blood Moon + Urborg, Tomb of Yawgmoth.
You end up with:
Layer 4 (type): Blood moon turns it into a mountain (therefore removing its abilities), Urborg wants to turn everything into a Swamp.
Urborg can't turn everything into a Swamp if it doesn't have its ability, therefore there is dependency on blood moon.
Apply blood moon first, nothing is turned into a swamp.

Now, I agree that these are both correct per the CR, as that's how dependency works (same layer, one effect among other things affecting the existence of another). It's just conceptually odd because now in some cases, we do look at what has happened earlier in the layer process and sometimes we don't.

Here is what I mean: Humility removes it's own ability in layer 6, but we still appy it in layer 7b, because ‘that’s how layers work'. We first make a list of all the effects in all the layers, and then simply go through them, regardless of whether the ability that generated the effect is still there when we arrive at a particular layer.
Urborg's ability gets removed as we start layer 4 (as a side effect of the type change), so that later in layer 4 we don't apply it anymore, because that's what the dependency rules tell us to do. In this scenario, an effect actually gets crossed off the list as we go through the layers.

So sometimes abilities are removed and we do still apply them later, and sometimes abilities are removed and we don't apply them later.

Now, I remember the mess layers were before the current system so I'm not about to propose a change, but can someone give me some kind of explanation about why this makes sense, because I'm having a hard time wrapping my mind around it.

Edited Anniek Van der Peijl (Nov. 9, 2014 12:46:36 PM)

To make a rough analogy, it's impossible to turn off half of a single light bulb. In the same way, it's impossible to turn off half of a single effect. If the game has started to apply an effect, the rest of that effect will continue to apply, even if the ability that generates the effect has disappeared in an earlier layer.

In the first scenario, Humility's effect starts to apply in Layer 6, and we continue to apply the rest of it in later layers even though Humility has removed its own ability in the meantime, because the light bulb was turned on, which means all of it was lit, not just part of it. In the second scenario, however, Urborg, Tomb of Yawgmoth's ability is removed before it can ever start applying, so none of it applies–the light bulb was never turned on in the first place.

A full explanation of why the layering rules are set up to work this way is beyond the scope of this forum; it involves a significant amount of rules philosophy and history to put things into context, and a place where you can't ask follow-up questions or request clarification isn't the best place to have an in-depth discussion anyway. I suggest re-posting in the Rules Theory and Templating board on the Wizards Community forums if you'd like to have that discussion; I'd be happy to go into more detail there.