But that 20m year old light is in our past light cone; we can say it is in our past. If it is not in our light cone, we cannot communicate any temporal info so we cannot say if it’s now, past or future in reference to us as observer.
Pluto is around five light hours away from us.
If an astronaut goes to Pluto and films herself doing activity A and then broadcasts that back to Earth, it would take around five hours for the broadcast to reach us.
So you’re saying that it is not correct to say her recorded activity happened five hours ago?
This becomes painfully apparent when dealing with black holes. Person A falling into a black hole passes through the event horizon normally and reaches the singularity in finite time. However, Person B observing from outside never sees Person A pass through the event horizon. From Person B's perspective, it doesn't make sense to ask what happens simultaneously with Person A reaching the singularity - it takes infinite time just to reach the event horizon. Even asking the question becomes nonsense.
But from the Earth’s reference frame wasn’t the Pluto activity I mentioned earlier simultaneous with what happened 5 hours ago on Earth?
So couldn’t you say from Earth’s reference frame this black hole’s activity happened simultaneously with events 20 million years ago?
I think the difference in the scenario you proposed is that we can conceive of a practical situation in which someone on Pluto sends a message and waits for a reply. We can then say that they are going to have to wait at least 10 hours to receive a reply. This has meaning.
But when the distances become too great the urge is to resist believing that things have an existence beyond the interactions of particles. There is no practical reality to that object beyond our ability to observe it when the light reaches us.
It's obviously fair to say they happened at the same time in this scenario.
Someone sends an email or speaks on a voice call on the other side of world and a receiver still gets that note at the speed of light, there is slight latency. It's still completely reasonable to measure that latency as we can effectuate improvements to our machines and networks when it increases. The same as if someone wrote a letter in the 1860s and the horses were sick, causing a delay. Simply because information was received by us, the common understanding is that the notification was sent previously, because that is how cause and effect work.
No. Special relativity says that simultaneity is relative for things moving at different velocities. The classic example is a fast-moving train and a platform disagreeing on simultaneity, even when they're right next to each other. Distance, and light travel time, has nothing to do with it; that's a separate (and mostly less interesting) problem.
Relativity absolutely lets you calculate what was happening simultaneously with a given past event within your own reference frame (i.e. other things moving at roughly the same velocity you are). In fact, you can think of that process, figuring out a consistent version of events after you've gathered all the data, as being exactly what it means to "observe" something in relativity. This is as opposed to "seeing it", which is what happens the exact moment the photons hit your eyes/sensors.