Look, rockets are cool. Really cool. They look good. Very videogenic. Films (real or virtual) set in space are usually cool just because they involve rockets.

They can get us to Mars. I admire Elon Musk, SpaceX, Blue Origin, etc. I think their efforts are extremely impressive, and the use of software to achieve very fine-grained control of the rocket engine is something that's genuinely new and interesting.

But... rocket engines can't get us to another solar system. The maths doesn't work.

So: Here's my thesis:

A workable interstellar drive will emerge from strange new experiments in physics, not from gradual improvements in current space travel technology.

Now, usually you learn best by doing, through taking actions and adapting yourself to the results of those actions, changing the way in which you interpret the world, not just intellectually but at a gut/hind-brain level, all the way down through your biological structure. (See The Path of Blue and Orange for a related meditation). So, if there is a space industry, and various people are working in the field, trying to solve problems, mostly failing, sometimes succeeding, then it's much more likely that someone will say "hey, what about trying X?", and X will turn out to be really cool. When many people are focused for a long time on a particular type of problem, the odds are decent that one of them will see something interesting.

Therefore, the existence of the rocket-engine-based space industry increases the chance that someone will discover a new type of engine.

Relatedly: Coal-mining companies in the 1700s had to pump water out of the coal mines in order to get at the coal. Eventually, someone developed a coal-powered steam engine to do the pumping. Then, someone else realised that such an engine could be used to move a train along rails...

So, I am not suggesting that we should stop working on rockets. I am saying that we should be much more ambitious. Officially so, i.e. "We are definitely trying to make a warp drive", rather than just vaguely hoping one emerges from new experiments.

Obviously, it may actually be impossible to develop an interstellar engine that we can use. The universe and its physics may just not allow it, and that's that. [0] Fin. But: If you believe that it's impossible, you guarantee that the effort will not be made. I prefer to deliberately believe that it's possible, and then to organise my activity accordingly. That's the only way you get a chance of success.

I sat and thought for a while recently, about what I would do if I no longer had to focus entirely on making money and/or protecting money (which today basically means working on Bitcoin-related code and politics - see The intrinsic value of Bitcoin). I realised that I would spend my available time working on a warp drive.

Mars is cool, but I would like to visit Alpha Centauri.

So, to business...

In physics, if there's some physical domain, and you take a parameter and shift it steadily, eventually you reach some boundary condition, an edge case. At that point, the domain undergoes a phase transition, and the conditions within the domain may now be very different.

Example: When water (a domain) gets cold enough (a parameter hits a boundary condition), it undergoes a phase transition and becomes ice. When a river freezes, you can now drive a car across it, whereas before, you couldn't. A new pathway in the world has opened up.

The car should ideally be outfitted with appropriate equipment to handle the domain (e.g. winter tires with more grooves), so the analogy to some sort of spaceship is rather good.

So: To break out from existing categories, to find new ones that may have new utility, look for edge cases, places where the rules break.

The conceptual toolset is:
- Where are the edge cases in current-day physics ?
- What is the next experiment that could show something interesting ?
- How could such an experiment be constructed ?
- Once interesting behaviour of some physical object has been demonstrated experimentally, can this behaviour be controlled ? Can it be scaled up ? Can new tools be developed to do this ?

My gut feeling is that quantum physics is the place to look first. It's a domain in which really weird behaviour is regularly found (e.g. quantum tunnelling, quantum teleportation), which violates the "normal laws of physics" i.e. the laws that usually apply to human activity.

Interestingly, there is one situation I know of in which quantum effects become visible to the naked eye: Liquid helium flowing uphill.

Perhaps there's a way to control and increase some quantum effect so that we can use it on our scale. I.e. a quantum gate, a portal through which objects / people / spaceships can travel.

For local tests, Earth is fine. But if we ever manage to open a portal to Alpha Centauri, it would probably be best to run the final experiments in space, in case something weird came back through from the destination. So let's do that one in orbit, and get the stargate up there using standard rocket engines.