I feel like a good first step would be setting up an orbital base/space station that orbits mars. It might be be a little easier than habitation On the planet itself, and more risk for the astronauts involved. However, if we set that up using knowledge gleaned from the ISS, I think its a viable path to take.
The attractive thing about landing on dirt is that the raw materials mean that a closed life support system isn't needed. There are relatively accessible supplies.
Space stations anywhere, be that LEO, lunar orbit, Mars orbit, deep space, or floating in Venus' atmosphere, lack that option.
As I told the robotics people at Stanford who were talking about something similar on the moon in the 1980s, "how soon can you do it in Arizona?" The machines sent to Mars so far are very small and wimpy. Because they're semi-teleoperated, and have very limited solar power, they're really slow. If you want to do any high-power activity on Mars, you'll probably have to send a nuclear reactor.
