Call the Air Force, the NRO, and Burt Rutan and demand they stop building a vehicle whose possible existence you've disproved.
None of them are building a horizontally launched vehicle that don't use wings. And the Air Force and the NRO aren't doing anything horizontal.
Of course they're not. Notice the thread title you're in. "Envisioning the world of 2100."
The reason they're not is that currently rocket engines are extremely
expensive. If you asked someone about lifting something like the SLS horizontally, they'd look at using a dozen RS-25's at probably $120 million a piece, adding about $1.5 billion to the launch, and then throwing them away
. You could use a million
man hours of pad support and still come out cheaper than adding one engine.
But of course the whole reason you can throw crazy amounts of labor into a launch is that the launch itself is extremely expensive. The SLS is going to throw away four or five RS-25's per launch already. These costs also keep spaceflight from being either routine or affordable.
The horizontal launch becomes preferable only when motor costs drop (and reliability goes up), and SpaceX is making a lot of progress along those lines even while using largely conventional construction. As we start moving to re-usable systems, not just rebuildable or refurbishable systems like the Shuttle, the key to low cost operations is to massively reduce the amount of man-hours needed to operate the system. Once you've got low-cost rocket engines that are as reliable as any Pratt and Whitney or Rolls Royce on a Boeing, you can add extra engines to a rocket to save man-hours.
The performance penalty is quite small. I've just run some numbers on about 30 different rocket stages, adding 33% to the dry tank weight and providing the extra engine weight to accelerate the structure, payload, and fuel at 1.3 G's (generally doubling the engine mass). That's upping the structural coefficient of most stages by 39 to 49%, which isn't a whole lot considering that structural coefficients of existing stages vary by a factor of thee or more. Almost all of the stages retain 91 to 96% of their original delta-V, or for the same delta-V deliver 75 to 96% of the original payload. If half the cost of the rocket still remains the engine, then such a rocket would cost about 150% as much as the vertical version, and the cost per pound to LEO is 54% to 100% higher (which is among the reasons why we don't do this yet
But the horizontal stage is much easier to reuse, more controllable for re-entry and landing, never needs a VAB or a giant launch tower, and if the engines are truly reliable, it won't need much ground maintenance, either. Its support costs start looking much more like an airliner's than the billion dollar launches we currently employ. Using twice the engines for fifty flights is 25 times cheaper than throwing away one engine every flight. When it comes down to it, eventually the extra manpower in the vertical support (giant crawlers, Guinness Book buldings, $100 million dollar towers) will lose out to a company that has figured out how to launch a rocket over and over with a ground crew of forty instead of a thousand, able to maintain vastly higher flight rates as the designs evolve.
Keep in mind that engineers and aerospace systems analysis comes up with solutions optimized (hopefully) for the current cost
environment. When relative costs shift, so should the designs.
(I highly recommend MIT's graduate level course on aerospace system design that was entirely devoted to the Space Shuttle). It's far more fascinating than Star Trek.