Once again.
The shape of the horizon is immaterial. The horizon is nothing more than the furthest one can see in any direction. It will appear around the observer in the shape of a circle. Therefore it will be curved. As long as a horizon exists, it has to be curved. This is basic logic and geometry here.
non sequitur. Roundy, you're not making any sense, pal.
Why is it a non sequitur? The FE has a reason for why we see a horizon. Why should it necessarily not be curved? You can see a curve in the horizon from a bridge. It defines the boundary around us, making it a circle, making it curved.
Roundy, you really need to work on stepping through your logic one step at a time. Let me show you how.
The standard Cartesian coordinate system in three dimensions uses three axes, x, y, and z. The main vertex of the system is x=0, y=0, z=0, or more shortly (0,0,0). Let's have you as the observer at (0,0,0) facing the x-axis's positive vector. The x-y plane lies tangent with the Earth. Assuming a viewing distance of 6 miles, the horizon would be the circle described by the equations: {x
2+y
2=36 miles
2, z=0}. As you look from (0,0,0) you can only see a straight line. You're not high enough to make out the curvature. Even if you could, the horizon would curve towards you in the x-y plane. There is no way on a flat Earth to see the horizon with z < 0, that would be underground. RE predicts the z<0 curvature that we see. FE does not. RE is more predictive than FE. This builds our confidence in RE.
Your basic mistake is saying that a circle in the x-y plane results in a curvature with z<0.
I hope that helps.