Jeff Caldwell's recounting of the "interior doesn't exist" interpretation of black holes was fascinating, but I perceive the following problems: 1. There also are horizons in an expanding de Sitter universe (anything on the other side of the horizon can never come back to us, since universe "expanding faster than light"). Also, for a uniformly accelerating observer, there is a horizon (stuff on other side "can never catch up" to us, if we keep accelerating at same rate forever). But unlike for a black hole, these two kinds of horizons are observer-dependent, and the lattermost kind stops existing if you stop accelerating. But like for a black hole these horizons also emit Hawking radiation and feature "uniform surface gravity" and "infinite red shift." So, trying to say the stuff on other side of these horizons "doesn't exist" does not work. So then you'd have a funny situation of treating black hole horizons and these horizons differently, despite their similarities. 2. Caldwell's whole recounting seems in deep trouble if employ quantum gravity instead of GR. Jump into a black hole? Caldwell's way, your mass & charge & momentum & angular momentum merge with the boundary surface, but other than that, you stop existing. But that is a NON-UNITARY, and NON-INVERTIBLE many-to-one transformation, which is totally incompatible with the foundations of quantum mechanics. So, to accept this black hole interpretation, you've got to totally redo QM somehow. My conclusion: we must accept that black holes have interiors. Or store a crapload of information on their surfaces.