When considering an engineering project, there are three potential barriers to succeeding:
Political/Economic. These are things like funding and NIMBYism. A store that wouldn't mind having your train station across the street may be upset if it's so close it requires them to find another location. Such considerations need either to be bought off or taken by eminent domain. Enough such problems may make it impossible to implement your project with the funding you have, but are solvable problems, at least in theory.
Debugging. It's rare that the first design of anything very complicated succeeds. Some debugging or redesign is nearly always necessary. Oversights, mistakes, shortcomings, misunderstandings, etc., always occur. The engineering or construction time to fix it costs money and if there's not enough available, the project may fail, but if the problem is one of debugging, it is a solvable problem.
Breakthrough required. Some ideas are simply contradictions of the laws of physics. For example, a nineteenth century scientist likely thought that traveling to the stars was just a matter of going faster and applying more energy. But in the early 20th century it was discovered that the speed of light is an absolute limit and that the energy required to get even close to it was impractical. To get to the nearest star will take more than a human lifetime. Unless some breakthrough, like Star Trek's "Space Warp" technology is discovered. It's conceivable but nobody has even the first idea of how to do it.
So, for example, Hyperloop (transit in an evacuated tube near the speed of sound) does not require any breakthroughs to succeed, although there may be some. All the technology required is fully understood, in at least a general way. There may be some friction from politics and engineering, but with enough determination, they are all solvable.
A space elevator (a cable anchored to a geosynchronous satellite) is mostly straightforward engineering, except for one thing: the cable needs to be strong and light enough that it doesn't break under 22,000 miles of its own weight. No fiber presently known can do this. There are some breakthroughs on the horizon that might succeed (e.g. carbon nanotubes), but this is impossible until there is an actual breakthrough.
