we can also state today that it is not possible to prove the fifth from the other four. this is a consequence of godel.
but, are the other four postulates truly obvious? i think the first, third and fourth are pretty clear. the second one causes me some pause.
2. Any straight line segment can be extended indefinitely in a straight line.
later commentators tried to redefine the concept of a "straight line" to something inherent to the surface. and it's important to point out that indefinite does not mean infinite. but, i think this assumes parallelism in the plane in the first place.
perhaps a better approach is to define the space you're working in in the first place. mathematicians do this regularly when they discuss algebraic structures, so it's kind of weird that they don't when they discuss geometry. this is in fact the necessary adjustment that's come out of the acceptance of non-euclidean geometries, and how mathematicians approach things in practice, it just strangely hasn't been formalized. to be clear: mathematicians don't pretend a geometry applies to reality any more, they just treat them like abstractions and then let physicists deal with the applications. once you've set the actual characteristics of a plane (defined by intersecting right angles), you actually only need three postulates: 1, 3 and 4. two and five follow rather trivially, by the nature of the surface. further, the non-euclidean geometries become extensions of the surface.
it's easy to accuse me of missing the point, but i'm not - i'm actually getting the point, which is that there isn't a universal geometry. the geometry is specific to the surface. it really ought to be formalized that way, by setting down the characteristics of the surface first and then setting axioms as to what you can draw on it.