Here is another future technology where it is not yet clear how things will work out – the hyperloop concept. Several companies are working on developing what is called a hyperloop – a closed tube with reduced air pressure that will operate like a rail system for high-speed travel. The idea is almost a staple of science fiction, but could one day be a reality.

Most of the press for hyperloop has focused on Elon Musk’s project, because he is Musk. I like that Musk is not afraid to pursue bold new technology, and not afraid to fail. He will throw a lot of money, and dedicated focus, at a problem and has produce some stunning results, such as SpaceX’s rockets able to land after use. But he doesn’t have any magical ability to get around real-world limitations, just the money to hire great engineers and sick them on a specific problem.

One of those problems is developing the hyperloop. The idea has some solid merit. Having a dedicated path for traffic (you know, like a train or subway) has advantages, especially if that pathway is isolated from pedestrian or competing traffic. The idea of using a tube with reduced air pressure, in order to minimize air resistance to improve travel efficiency, is also a good one. I remember reading about this decades ago, so the idea itself is not new to Musk. The closed tube would also protect the path from weather.

From there things get progressively tricky. Is the tube going to be above ground or below ground? Neither option is great, and a system may need to use a combination. Above ground is far cheaper, but then you have a tube cutting across the landscape. You could elevate it so as not to interfere with animals and other traffic, but that adds some expense and creates another point of maintenance and possible failure. Such a structure would also be hugely vulnerable to terrorism.

An underground tunnel would be optimal – out of sight, away from harm and the elements, and the path can run under cities, roads, rivers, and other obstacles. However, an underground system would be much more expensive, prohibitively so.

This, of course, is the exact same problem any similar mass transit system has. These are major infrastructure investments, and which lucky cities get to be on the path? Just working out the rights to pass through many different pieces of land can get horrifically complicated. But if we made this investment, what would be the potential pay off? This is where the hype comes in. Optimistic projections are that some hyperloop systems could have cruising speeds around 500 mph. There is time to speed up and slow down, but still, that’s a good clip.

Different companies also imagine different types of systems. One approach is to have dedicated pods, like a subway, that carries many people. This is the system being developed by Virgin Hyperloop One. They claim they have hit a maximum speed of 240 mph on their test track. They still have a way to go, but even if that is the top speed, it’s good enough for reasonably efficient travel at medium distance. For long distance you would still want to fly.

Musk is developing a different approach – one in which a regular car drives into the loop, is placed on a sled that then magnetically levitates down the track at high speed. At the end of the trip the car and passengers just drive away. This would function more like a sophisticated road tunnel than a mass transit system. Musk’s system, however, has hit a snag. In early testing, even at relatively low speeds like 50 mph, the rid was apparently very bumpy. Musk promised to sort that out, and he has. But the solution was to simply pave over the sled tracks, so now cars just drive on regular pavement, but in a low pressure tunnel. Musk says this system, “Is simple and just works.” Sure – but then what is new? In that system you simply have a dedicated road, and the only real innovation is the reduced-pressure tunnel. I doubt that is enough to justify the massive investment.

I think it’s fair to say that by paving over the rails on the test track Musk is admitting that his system has essentially failed. Again – I don’t blame the guy for failing. Go ahead and fail all you want, as long as it’s honest and you occasionally produce something useful. Also I should mention that Musk wrote a white paper on the hyperloop concept in 2013 which he essentially made open source. Hyperloop One started in 2014 as a response. So I give Musk credit – he wants the technology to be developed, even if he is not the one to do it.

Hyperloop One is attracting investors still, but the companies is apparently have some setbacks and has a long way to go before having a product that can be implemented.

Let’s say that we work the technology out. Hyperloop one or someone else develops what is essentially a subway system in a depressurized sealed tube, allowing for much faster speeds – something in the 500-600 mph range. That is in the range of commercial jets. So for long distance travel jets would still be the better option. The only infrastructure you need is an airport at either end. Their safety record is also excellent. They are also highly versatile – you can add and change routs to accommodate demand without having to build new infrastructure. If you want faster, then it is probably better to invest in supersonic jets.

Could a ticket on a hyperloop car be cheaper, or could the experience be more efficient? Hard to say, but given the infrastructure investment I doubt they will be cheaper anytime soon.

The niche may be not to compete with jets, but to compete with trains or buses. A hyperloop system between cities with high populations and high traffic between them and at medium distance – New York to DC or Boston, or LA to San Francisco or Vegas. I think that is the idea – this would essentially be a high-speed train system. High speed trains, maglev and non-maglev, are already deployed around the world, with speeds in the 200-267 mph range. You may notice none of the fastest trains are in the US. So – if we cannot work out the rights and investments to have a maglev in the US, why will we be able to have a hyperloop?

It is perhaps telling that one of the major investors for Hyperloop One is in Dubai. If a hyperloop system is put into place, it probably won’t be in the US.

I think the bottom line is that this technology can probably work, but it still is not certain if it will be worth it. Right now we have a hyperloop train system with a max speed of 240 mph. We can compare this to a maglev train, the fastest of which is 267 mph. The question, therefore, is – will the hyperloop train achieve speeds significantly faster than a good maglev system, fast enough to justify the far higher infrastructure costs? I have to say I am pessimistic about this. It probably will take achieving speeds in the 500-600 mph range or more, so until that happens, it just won’t be worth it.