Concerning Eliéra

They were never wholly separated. Be it bold explorers and outward traders crossing the Dominions-Kaidoná land bridge (or very lost spelunkers and bold downward traders going the short way around) and bringing back lore and exotic goods, or sailors with very good ships and no fear of death sailing through the Edgestorm, there was always some contact. (Later, while airships would never have been a good idea, a well-built aeronef and a pilot with big, shiny brass ones could also make it through the storm; in the modern era, semi-ballistics find it easier to go over the top.)

So no-one was truly surprised by the existence of eldraeic civilization on the Underside. (Or Upperside, depending on your perspective.) It had always been hovering off at the edge of things, like a New World or a Terra Australis that was a very interesting place, but alas, one limited in use for now by how much of a pain in the ass it was to get to.

And then a certain physicist did a really comprehensive survey of Eliéra’s gravitational field, noticed those dimples, and suddenly - with, admittedly, the aid of an entirely new generation of colossal boring machines and roughly 146 years of continuous labor on the first one - it was right next door.

(Which was, I hasten to say, more than enough time for news to travel to the other end the old-fashioned way. The foreign policy folks were busy .)

Very, very early in pre-Imperial times. You don’t have to spend much time looking at the sky to note that those things in it are round and never change shape; you don’t need much in the way of tools - a crude lens and a nice clear night might do, and even then only if you ignore the moons - to observe that if the shape stays the same and the features change regularly, you’re looking at a sphere.

And primitive folks, of course, assumed the world to be flat for much the same reason that we did, except with the slightly greater justification that there literally wasn’t a horizon, just a distance-haze. Plus, in some place, the rather noticeable edge. (It was proved by an ancient Eratosthenes back in the day, but it was more in the nature of a “behold, what we all know actually is true”.

(Spoiler alert: it isn’t, because it’s actually slightly convex, which is necessary to make it look flat thanks to atmospheric refraction.)

Heliocentrism, incidentally, was also doomed to an early death because binary systems do help give you something of an early tutorial in celestial mechanics (“How many epicycles?”), and seasons and deep seasons on a flat world (without, therefore, axial tilt) drop some heavy hints about ellipses.

Not a whole lot in the way of social consequences to all these, mostly because they happened sufficiently early no-one remembers them if there were.