@everything comes in twos - offtopic and long
A spaceship could totally descend through the atmosphere like a skydiver and not experience the burning-up, but it would have to slow down first. The man in a balloon is 'orbiting' the earth geosynchronously, by virtue of being coupled with the atmosphere, and so he has no net velocity in the tangential plane (neglecting wind currents). He's held aloft by his balloon and when he lets go he falls straight down, as viewed from someone down on the planet.
An orbiting spacecraft isn't held aloft by a positive bouyancy and so it isn't really 'aloft' in the same way: it's actually falling (at least, being accelerated towards the centre of the planet) but also it's travelling at right angles to the path it would fall so fast that, in the time it takes to fall a small distance towards the planet it's moved past the planet by such an amount (think of right-angled triangles) that it's no further away from the planet's centre, but is a few millidegrees round the orbital path and travelling with the same speed in a direction at right angles to the line between it and the centre of the planet, which is the identical situation to above but for a small rotation, so the same thing happens again and you move a few more millidegrees. It's all one smooth motion, and as you're out of the atmosphere (mostly) there's very little friction to slow you down and you can maintain an orbit with no energy input for a long time (the moon does not have engines, as far as I know, although that would be very cool).
The height of the orbit determines the linear velocity of the object, which is very high (>>thousands of mph). This is why geosynchronous orbit (its linear velocity as it travels round the earth is such that it covers one rotation in 24 hours, in the same direction as the planet rotates and therefore staying above the same spot) is a certain orbital band (i.e. between this height and this height). BTW outside the atmosphere you aren't influenced by the planet's rotation and you can orbit the planet travelling the opposite way, or from pole to pole, but inside the atmosphere you get dragged around with the planet.
To lower the orbit relative to the planet you just decrease your linear velocity, and gradually you enter thicker and thicker atmosphere, which exerts more and more friction. If you want to land, you have to slow right down from thousands of mph to zero (relative to the rotating earth now so you have to account for that too but it's a smaller effect). If you're moving too fast then contact with the atmosphere generates huge friction (the force is linearly proportional to velocity and in the opposite direction) and you burn up. This is what happens to shooting stars and whatnot, and why the shuttle has the heat resistant pads.
So the skydiver starts at height z with no horizontal velocity and no vertical velocity, and as he falls he accelerates (-d2z/dt2 - negative and second order with respect to time). His parachute exerts a drag force (dz/dt - positive and first order with respect to time) which reduces the effect of the acceleration. He's still accelerating though, and the faster he goes the greater the friction force until it counteracts the force exerted by gravity. At this point he is no longer accelerating, but still travelling at speed. The size of the parachute determines this eventual speed, aka terminal velocity. You aim to have this low enough to be able to survive impact at that speed, so it's far too low for the frictional forces to burn you up (although if you're skydiving from high altitude you'd have to wait until you got into atmosphere to deploy your parachute, at which point you might be travelling faster than terminal velocity and would need to slow down to it, so it might be close).
The shuttle on the other hand is also at height z and with no vertical velocity but a lot of horizontal velocity which is enough to make it burn up in contact with the atmosphere, and it loses some of it by decelerating with its engine until it can take the atmospheric friction, at which point it sheds the rest of its velocity that way. If it does it right it will find itself flying through the sky like a plane, and it then stays aloft by aerodynamics and lands like a plane.
The difference between the two is the amount of energy they initially posess. They may be in the same place in the sky (for a moment at least) but they have very different inertias and it's the shedding of this inertia which generates the heat, as it's done by friction. The shuttle could slow down using engines instead but it would have to use twice as much fuel then, which would mean taking even more fuel in order to get it up there in the first place.
Orbital mechanics are quite complicated, cause you've got two reference frames (one being the rotating planet, the other being the nonrotating surface of the planet's volume), two coordinate systems (rectangular works at small distances like manchester to london but spherical works better for orbits), and a gradual boundary between them, where teh atmosphere thins to next to nothing, and two very different inertia levels, the difference between them being determined by the gravitational pull of the planet and thus ultimately by its mass (escape velocity is a measure of the difficulty of getting off a planet of a certain size that comes from this difference in energy levels). To calculate orbits and burn times you need to factor all this in - I suppose it must be like sailing, where the world is divided into two half-spaces, each with a different vector field of instantaneous force vectors (wind currents and sea currents). The boundary between the two fields is more defined than in orbital mechanics though.
I would get my coat but I don't need one as my hull is impervious to atmospheric effects.