Brian Greene wrote
a NY Times op-ed about the implications of the accelerating expansion of the universe. As he says,
Because of this, when future astronomers look to the sky, they will no longer witness the past. The past will have drifted beyond the cliffs of space. Observations will reveal nothing but an endless stretch of inky black stillness.
If astronomers in the far future have records handed down from our era, attesting to an expanding cosmos filled with galaxies, they will face a peculiar choice: Should they believe “primitive” knowledge that speaks of a cosmos very much at odds with what anyone has seen for billions and billions of years? Or should they focus on their own observations and valiantly seek explanations for an island universe containing a small cluster of galaxies floating within an unchanging sea of darkness - a conception of the cosmos that we know definitively to be wrong?
I'd thought before of the implications for scientists from future races that never had the chance to see other galaxies at all. Before reading this I'd always just assumed that we humans would be okay (assuming we survived that long) because we'd been lucky enough to see the truth. But now I worry that he has a point: assuming we survive that long, how much will those future scientists really trust our observations, so at odds with what they can see for themselves?
That's related to one of the philosophical realizations I've had about studying things like string theory and cosmology: Some true things are simply impossible for us to observe or measure, and many more are just not the sort of thing one can predict. It's hard to be comfortable with that: I'd like to imagine that I live in a universe that's not just understandable but verifiably so, but there is no guarantee that our universe will cooperate with that desire. The thought that it might cooperate for a while and then stop is especially uncomfortable: what might we miss if we don't develop the technology to observe it fast enough?