Now, it's entirely possible that a complete theory of quantum gravity comes out and upends our understanding of what happens everywhere in space, including at event horizons, and perhaps it's the case that black holes are never formed. But to date, a fully consistent theory of quantum gravity has yet to be created. But just saying "assume black holes don't form, and all contradictions go away" by itself doesn't really help, because that just says get rid of GR and/or QFT, but doesn't say what to replace them with, beyond "something that doesn't produce black holes". Which, sure, would be great, but the devil is in the details. The more important thing to most people in the field is solving the UV divergence problem anyway, which will answer in detail questions about Planck level effects. Whatever comes out of that will (hopefully) unfold naturally into the answer to the information paradox too.
As far as the final question, nobody knows. That said, care is needed when describing events from different perspectives. The principle of relativity isn't that all observers agree on all events. It's that physics is the same in all frames of reference. Which is nuanced: what does one mean by "the physics"? In essence, it's a layman's statement of Noether's Theorem, relating symmetries and conservation laws. So, depending on quantum gravity's symmetries, different observers could see wildly different sequences of events, but they would agree that the conserved properties of the theory were indeed conserved. But until such a theory exists, it's anybody's guess as to what those symmetries actually are.
(>>highlight<< mine)
You're not contradicting my argument: I'm saying that there is only a "before", and never an "after". Saying that light emitted before crossing a horizon is visible in a finite time is not incompatible with what I'm saying.
Keep in mind that Hawking's model is known to be oversimplified as well! In his simple evaporation model, the information encoded in the infalling matter is lost, violating QM information conservation.
If you simply assume that no infalling matter ever crosses any horizon, that it all just "blows up" very slowly from the perspective of an outside observer, then there is no QM information paradox, no inconsistent observations, etc...
If you disagree, please cite a recent paper.
Here's a nice thought experiment for you: What happens at the last moment of an evaporating black hole's life? How does the horizon "disappear"? Whatever you imagine happens, now update your mental model for a relativistic observer. What do they see? I.e.: Does the increased apparent mass delay the observed disappearance of the horizon!? If so, how can this be? How can two observers disagree on the presence or absence of this horizon? Now consider what would occur at this juncture if there was only smooth curvature, no horizon, and no singularity. Would this enable the last moments of a BH's life to be consistently modelled for all observers?