Synthetic vision is a coming thing. Which amounts to a camera in a turret that may be using IR or visible light or both. Which will be overlaid on a cartoon version of what should be seen out the window. A la the scene generated and displayed in a training simulator or high end video game.
This can permit lower safer approaches in bad visibility with less expensive ground equipment and aircraft avionics than current practice. The aerodynamic and structural hit from real windows is tolerable for subsonic aircraft, so I expect EVS to be in addition to, not instead of, cockpit windows for subsonic aircraft UFN. See for more:
But … the aerodynamic and structural hit from real windows gets to be a problem for supersonics, and especially for low-boom supersonics. The way to make a low-boom supersonic airplane is to make it extremely long and thin. Given the size of the engines, wings, and payload compartment, the best way to make it thin(ner) is to put a skinny stupid-long beak on the front. Sort of like the current style in men’s dress shoes but turned up to eleventy.
And with a beak that long it’s impractical to have a drooping nose a la Concorde or XB-70. In that case the pilot’s view out the front will be 100% through the camera(s). Logically akin to Lindbergh’s periscope back on the Spirit of St. Louis, but vastly more capable.
This experimental airplane is an example:
This proposed low-boom executive jet SST flies at a fairly low Mach. And can just barely afford windows. Based on the nose shape plus likely AOA on landing I would expect the pilots to be landing mostly via what they see through the camera(s), not out the windows. Much like taxiing a taildragger, the view straight ahead is full of airplane. Which is much better when it’s your airplane filling your windscreen rather than some other one.
The next gen of Aerion-like airplanes that’re faster will probably lack cockpit windows. Assuming such are ever built.