Wikipedia’s article on the British Interplanetary Society and their proposed moon rocket has one rather remarkable cite tossed away virtually in passing:
As it was considered that the cabin would have to rotate, BIS member Ralph A. Smith, who helped re-establish the society post-WW2, invented the first instrument for space travel: the Coelostat, a navigation mechanism that would ingeniously cancel out the rotating view.
Presumably you’d have some sort of counter-rotating mechanism to match the ship’s spin, but then what? You might keep the same field of view in sight, but for the life of me I can’t picture any arrangement where the field of view wouldn’t rotate around its center at the same rate as the ship. Is there some “ingeniously” way of completely de-spinning the view? Would the viewer simply be stationary and let the cabin rotate around him while observing?
In astronomy, a coelostath is a device that compensates for the rotation of the Earth when you observe extraterrestrial objects with a telescope. Normally you’d use an equatorial mount, which places the telescope on two rotational axes, one of which is parallel to the Earth’s axis of rotation. As the Earth spins, you rotate the telescope around this axis to keep the target object in sight. When the telescope is too big to be mounted that way, you’d use a coelostath, which is essentially an auxiliary mirror that is mounted equatorially and that reflects the light from the target to the actual telescope. I suppose the coelostath mentioned in that Wiki article would be similar.
Picture a periscope: You’re looking forward, and then there’s a mirror that reflects your field of view upwards, and then there’s another mirror at the top that reflects your field of view horizontal, instead. Except instead of pointing the top in the same direction you’re looking, it looks backwards. The two mirrors will invert your view through this device, so that you’ll see the ground on top and the sky on bottom. This is the same view you’d get if you turned around by doing a headstand.
Now take this same device, and rotate it, so the “vertical” tube isn’t vertical any more, but off to your right side. Now, when you look through it, the horizon will be right-side-up, and you’ll get the same view as if you turned around in the normal way (keeping your head up).
If your view went from upside down to rightside up as you rotated the device, then if you were looking through the device as you rotated it, your view must also be rotating (at twice the speed that the device itself is rotating).
There have been plenty of optical devices invented over the years that can provide rotation of the image. There wasn’t any need to invent another.
The simplest is probably a Dove Prism, which is basically a truncated equilateral prism. It’s commonly used for this purpose. It introduces astigmatism (though not so you’d notice in casual use), so it’s best suited for very distant objects. Which makes it perfect for rotating a view of stars.