this is indeed very cool, but its not really a substitute for a full fledged "camera" object because it doesn't take into account things like parallax and 3d effects (box, distortion) which are based on the scroll position of an INDIVIDUAL viewpoint. these things would have to be re rendered for each camera.
of course with lucid s method you could solve this by moving the scroll to that position, pasting everything, ect for each viewpoint, but i dont think paralax and 3d things can be pasted properly (besides 3dbox) and im not sure the effects like that would update before frame drawing in the event phase. meaning things would be alot more complicated than simply having a nice object to do it for you, which is kinda what construct is supposed to be about, elegance and simplicity in game creation. so that you don't need to always fumble around and do multiple steps for things which should only take 1. another thing which would be cool to have in a camera object is for it to be able to rotate and move along XYZ, but that might be asking too much