To deal with the galaxies moving together it's helpful to look at buoyancy. If you've got something floating in the water and you move it up or down (relative to the water's surface) then let go does it just stay where you left it? No- gravity yanks it back into position until the downward force is matched by the upward force.
For the atoms in dense things like rocks and people the gravity and chemical bonds and such all keep them in the same relative positions, and any deviation is quickly corrected since there's very little space expansion going on within that area.
For much larger and more spread out things like galaxies the gravity cannot act so quickly.Furthermore you have to think about how the galaxies on the other side of them are impacting the gravity. Generally galaxies are all about the same distance apart (in very loose terms,) but here and there you get some asymmetric situations where the gravity has more of an impact in a particular direction.
So going back to the buoyancy we can ask which is greater: the net gravity pulling us and Andromeda together or the expansion between them pushing them apart. Now unlike buoyancy there's not really some stable middle position where these will equal each other- if expansion were the stronger push at any particular moment then we would be moving apart and as we got further away the tug of gravity would diminish. With it being the other way around though we move ever closer together and that gravitational tug gets stronger all the time.
But we can again ask about how compare to each other when we think about the accelerating expansion increasing in speed. If the gravity was only just a tiny tiny bit bigger than the expansion then you can imagine it might take a very very long time for the galaxies to completely close the gap between each other. If the expansion speeds up fast enough you could imagine that we might get halfway together before the expansion took over and proceeded to fling us away from each other- this is about the way your dad is stating it- but in reality the expansion accelerates very very slowly compared to the gravity bringing us together. The galaxies will merge before the expansion becomes so great that it could keep us forever separate.
And so we've now essentially talked our way through a calculus problem- comparing different changing rates to each other. I'm not going to try and crunch the actual numbers but I'm sure you can see and probably even convey how they matter compared to each other. If you want to throw your dad a bone you can chime in that with an accelerating expansion with the atoms torn away from each other the galaxies will technically be torn back away from each other, but this is all assuming that the expansion accelerates at a uniform rate and will still be growing at that oh so very distant time.
The speed at which the speed changes can itself change and frankly it's difficult to have much certainty that this must be constant when we don't even really know why the expansion is speeding up.
Here is a question that is bugging me and I am sure it is because my mind can only wrap itself around 3 dimensions. They say that there isn't a center to the universe but if the universe is flat how is that possible. If the universe doesn't loop back on itself, "like the dots on an expanding balloon example" how is there not a set of edges to the universe?
Think of the balloon, as it was deflated every dot was the center cause it was one clump. As it started expanding every dot started moving away from every dot so it's impossible to say which is the center. If you did know the complete shape of the universe (which you never will be able to do because things move away faster than the speed of light) you could find the center of the universe by geometry. At least that's my understanding.
That's not so much the case at all. Uhm, letsee-
Alright, imagine that light only moves 1 meter per second and that space grows at .1 meters per meter per second- basically just small enough numbers that it's easy to deal with them. If something is one meter away from you then you won't quite be able to see it after 1 second- the object will be .1 meters further away from you and the photons of light will have had to cover just a pinch of that expanding space (sort of like how constant compounding makes interest rates not quite the flat value they're measured as.)
I'm actually going to cheat with the numbers and not do constant changes here though to keep it simple enough that everyone can definitely do it in their heads.
So for 10 meters of travel you have an extra meter after 1 second- but wait, the light only traveled 1 meter in that time. The light still needs to go 10 meters to get to you but after another second it's still going to be 10 meters away.
In this slow light fast expansion space anything more than 10 meters from you is never going to get light to you. 10 meters is a kind of event horizon where you can never receive any light. So then what does this look like to an observer? Anything just a smidge closer than 10 meters from them is zipping away at some breakneck speed while things a few millimeters from them don't seem to be moving much at all. Everything inside that 10 meter sphere is what we would call the observable universe.
It doesn't matter one iota where that observer is at- for any spot there's an event horizon ten meters away and the further away something is that faster it seems to be moving away from them. None of these things actually have much velocity to them though- there's no real inertia making them physically move through the space. No, the space itself is expanding while they sit pretty much still.
But then there's the element of time. It would take an infinitely long time for light from something 10 meters away to reach an observer so something just the tiniest bit less than 10 meters away might take years to finally close the gap- there is a longer delay between events taking place and the observer actually seeing them, again, regardless of where they stand.
Okay, enough with that small universe, let's look at our own. Where is that much-more-than-10 meter event horizon? Well actually with space not expanding so fast, light moving much faster, and the constant interplay of these (instead of my phony 1 second intervals as per above,) we've got a little barrier that's closer than that event horizon. Turns out that instead of space being clear and mostly empty like we're used to there was a point shortly after expansion started where it was all just an annoying soup that didn't allow light to travel uninterrupted. My fellow cosmological thinkers are probably just about screaming CMBR right now and yes, that's what we see when we look out as far as we can. There's a nice sphere where we're looking at light from so far back in time that we metaphorically smack into that soup with our skyward gaze. Many many years down the road we will reach a point where light from that era has so much ground to cover that the expansion just kind of swallows it up without us ever being able to see it again, and at that point we will have a clear view out to that very distant event horizon.
Ok, I've prattled on about that more than long enough- we still need to talk about the center of the universe. Well the center of the observable universe is always wherever the observer is. That's simple. Anything we can justly treat like an edge is always the same distance in all directions. But that deep thinker inside gets a bit nauseous at that weak answer since we've already established that the sphere for the observable universe is the same no matter where you stand. The good old anthropic principle, that there's nothing special about where we are right now compared to other places, demands that we look past our own not-so-unique place in existence and piece together the bigger picture. So then we wonder what is the center where all the particles in the universe originally sprang forth. Sprang? Well I'm going with it.
So how do we determine what edges there could be beyond what we can ever observe? Well like I said with gravity before there are things on all sides of our galaxy tugging on it. If we had some physical edge beyond which there was no matter galaxies around that spot would only really have the gravity of other galaxies pulling them back to the center (anything above or below, left or right would mostly cancel out, leaving only the center or edge directions.) You'd expect the galaxies there to pile up quite a bit more than those near the middle of the universe- and so with those many merged galaxies out around the edge you'd see about a layer in that those galaxies were pulled more strongly to the edge and so forth. You should basically be able to see some sphere shaped patterns echoing all the way back to the center of the universe (since every observable universe for a particular galaxy is centered on that galaxy they could interact like this just fine without breaking the speed of light limit for particles and forces and such.)
Yet we don't see anything of the sort. Our entire observable universe seems to be extremely uniform everywhere (after you correct for the time it took light to reach us- or only looking at equally distant galaxies,) without any pattern like you would get from an edge to the universe. To completely mask such a thing from us the true size of the universe must either be absolutely unfathomably large, curve back around on itself, or be truly infinite. Personally I don't buy the first option and every time we've checked the curvature of space we've found it to be so very flat that we cannot detect any non-local curvature so I've got just one option left- and it's not all that difficult to accept when you think of the expansion of the universe as the space between things growing instead of those things moving.
