Eric Hedman has some ideas. I agree with Clark Lindsey--he is far too optimistic about the prospects for scramjets providing a solution to the problem:

Scramjet propulsion theoretically has the advantage of increasing a vehicle’s mass fraction for payload by reducing the oxidizer load it has to carry. The big question is, can this theoretical advantage be turned into a practical advantage with a solid, sustained research and development effort? I think it’s time we find out.

The answer to his big question is...probably not. While the military may have applications for scramjets, I still think that it would be a foolish priority for technologies to reduce launch costs.

How long is a rocket in the 'Scram' regime anyway?

That is, how many seconds is a lifter normally over a minimum necessary speed (Mach 3?) and yet still close enough to the Earth to have a sufficient partial pressure of oxygen?

If you're flying a standard rocket trajectory, you'd only be within the operational envelop of a scramjet for perhaps 2-3 minutes. However, that's not how many people envision using a scramjet. For example, you'd take off horizontally using turbojet or turbofan engines like a conventional plane. These engines might possibly be surplus SR-71 engines that were designed to go Mach 3.2 or so. This would most likely be a two stage to orbit vehicle with the jet powered portion being the "mothership." You could top off from a tanker shortly after takeoff like the SR-71s did. Once you're ready, you light off the afterburners and accelerate to a speed fast enough to light the scramjets (high Mach 3 range or faster). Shut down the turbojets and accelerate to the upper speed of your scramjets. Some suggest limiting that to about Mach 7 so you can use hydrocarbon fuels and the heating issues aren't so bad. Once you reach that speed, you launch the smaller vehicle that would use rocket power to fly into orbit. The mothership would slow down and return for landing.

One of the hardest part of all this is dealing with the heat. Rockets climb through the region of maximum heating very quickly so it isn't much of an issue. The vehicle I described would get very hot because it would going Mach 3+ for quite some time. IIRC, parts of the SR-71 reached over 1000 degrees Fahrenheit when over Mach 3. This vehicle would be going much faster and I suspect it would take quite some time to accelerate from Mach 3 to Mach 7.

There's also the little problem with R&D costs. If NASA developed it, the mothership would likely cost many billion dollars and the smaller space plane billions more. Even if you could greatly reduce the amount of maintenance between missions, you'd have to fly a lot of flights to lower the cost to orbit below that of expendables when you factor in R&D costs, production costs, fixed costs (e.g. facilities, training, etc) and operational costs. A vehicle like that would probably be best suited for carrying a small crew to LEO. It certainly wouldn't be a good heavy lifter.

I've been told by folks who worked on it, that NASP was supposed to fly with slush hydrogen fuel and condense oxygen from the atmosphere using the thermal sink of the hydrogen (LACE scramjet). It would fly at high mach numbers only long enough to get in a position where it could use the hydrogen/oxygen rocket to boost it into a permanent orbit, thereby lowering the thermal load.

There was a note in Av week years ago about engines without turbo machinery, presumably meaning pulse detonation engines. If you could combine a PDE/ramjet/scramjet into one package, it would considerably lower your weight to orbit, at least compared to lugging along jet engines for no other purpose than getting off the ground and landing.

There's also been some papers about laser air spikes which may reduce the heat load into a hypersonic body and magnetohydrodynamic heat reduction near stagnation points where you expect a plasma at high mach numbers.

Which is why I'd like to see something like the X-43 flying fairly often, to either prove or disprove technologies which could be relevant to an aerospace plane.

I've liked something Henry Spencer had to say on airbreathing spaceplanes (primairily Skylon) a few years back:

http://tinyurl.com/yrhuv3