I know it's all a joke.
It's all about what credibility you can give to scientific observations that you havent observed yourself. (ex. How can you know for sure that the world is a globe if you haven't been in space, yourself, to observe it?)
According to Occam's Razor, logic used by scientists to determine causality:
We are to admit no more causes of natural things than such as are both true and sufficient to explain their appearances.
or,
The simplest explanation is usually the best.
This works great to explain the everyday observance that the horizon out there on the distance seems flat all around but in order to justify easily observable natural phenomenon that doesnt fit in with the FE theory, you cannot rely on the Razor any more.
Take for example the FE theory's explanation of Lunar Eclipses:
The moon isn't a spotlight; it glows with light from the sun, reflected off the Earth. Different parts of the Earth are more reflective than others (the seas, the polar cap, the ice wall, for example). Sometimes, the position of the sun (which is a spotlight) means that only very low-reflective or non-reflective parts of the Earth's surface are illuminated, so the moon is abnormally dark. This could potentially explain lunar phases as well.
Much more Easily explained by spheres obscuring light radiating from other spheres.
Other phenomena yet to be reasonably explained by FE Theory:
the "optical illusion" causing sunrises and sunsets.
the Coriolis force
How the atmosphere stays on the Earth
So, in the end, what began as basically a thought experiment about observable evidence has turned into an abberation of pseudoscience.