Spoiler test of depth

The spoiler test of depth states that if something can be "spoiled" by spoilers, then it isn't deep, or in other words, that truly deep things cannot be spoiled by giving away some secret.

The spoiler test is distinct from the related question of whether one would like to hear spoilers before consuming a story.

I notice that in math, I often want to be spoiled -- I want to understand something, and I want people to tell me the answers. That doesn't mean that I won't eventually try to re-derive everything from scratch.

I started having thoughts along these lines because Jonathan Blow often mentions spoilers (especially when talking about The Witness), and how he doesn't want to spoil the game for people. I think I have a disagreement or a personality difference with him. The disagreement isn't whether enjoyment from playing the game will be diminished by reading analyses about the game and watching let's plays -- I think we both agree that doing that does diminish the enjoyment you get. Rather, the disagreement is about whether a game whose enjoyment can be diminished in such a way is even worth playing in the first place.

Why should spoilers have anything to do with depth? I'm not quite sure. But I think it has something to do with "replay value", and how truly great things can be consumed over and over again (sometimes delivering greater enjoyment on later iterations). If an idea is deep, you can't really comprehend it in the first attempt; you need to keep coming back to it over and over again, to "attack" it from different angles and see it in a new light (say if you come back to it after having learned more about some other topic).