[txcwpalpha]

>And I'm pointing out that there's no magic in the tunnel construction. By the time you have added those safety redundancies you end up with something very close to a subway system, just with smaller carriages running on batteries.

What safety redundancies? You still haven't pointed out any required safety redundancies that would raise the cost of the tunnel any significant amount. Some RFID tags and low voltage wiring and fiber for a couple of wireless access points is still negligible cost compared to a full blown track-and-third-rail system. Your entire argument rests on "what if the cars aren't good at self-driving", but again, we already have vehicles with these technologies on the market that drive thousands of miles a day without issue. Tesla cars drive on autopilot on completely dark, unlit roads all the time without issue. They park themselves in tight parking spots every day without issue. They avoid collisions with other vehicles at high speeds every day without issue.

I'm no fan of Tesla, and I'll be the first person telling you that they fucked up bigtime handling the crash a few months ago, but that still doesn't change the fact that they have existing autopilot systems that are more than capable of handling this 'Dugout Loop' system.

As for your skepticism that the tunnels will be cheaper, go do some research on how expensive it is to build out subway tunnels and the components of those costs. The Dugout Loop system eschews significant amounts of the major cost components.

If you're skeptical because we're talking about Musk and his penchant for cost overruns and missed deadlines, I'd understand. But if you're skeptical because for some reason you don't think a barebones concrete tunnel (even with some additional wiring and networking components) is cheaper than a full blown subway tunnel and station, that's just because you're being stubborn.