Side-view curvature discussion moved to https://www.metabunk.org/a-side-view-of-the-curvature-of-the-earth-at-lake-pontchartrain.t9268/ note: Soundly provides new proof that curvature can be seen in "side view".

I noticed that one or two people wondered if refraction was causing the curve of the towers to be exaggerated closer to the horizon. Well, I don't think so. If you look at the picture, there are as many towers in the last 20% of the line as there are in the 80% closest to the observer. That's why the towers further out appear to curve more. The curve over the first 80% of the line is the same curve as the one represented by the last 20%, which is compressed into a quarter of the space.

You could not be more wrong. There are countless designs where CURVES, are made up of straight lines, and your perspective on those lines create the curve..

But here we are talking about a single straight line. Not multiple lines. It’s impossible to make the straight edge on the red level look anything other than straight with perspective (unless you use a non-rectilinear lense) and certainly impossible to make it curve over and behind the horizon, regardless of lens.

New member here. The use of these camera angles to compress what we are looking at into a smaller space to show existing warps or bends really works. It shows existing curves that would not otherwise be seen. I've seen this used for instance on runways, where you can clearly see how warped a runway is by looking at it from these compressed angles. Even to the point where you can see a plane landing gear disappear behind such curves or warps on the runway. It really works. Here is an example. Source: https://youtu.be/7P9OAng32F0?t=55s [Mod: Edits for clarity]