Yeah, and as always, it comes down to energy -- there's an established way to turn CO2 and H20 into Methane and Oxygen, given enough energy for input. And SpaceX has put some research into methane-fueled rockets already. So some people are theorizing that the play is to plunk an unmanned nuclear reactor and electrolysis equipment on the surface of mars, let it crank out a trickle of fuel for a couple years, and we've got the infrastructure to send a there-and-back mission to mars with on-site refueling. No big deal, right?