How about, before all that, we try and inhabit, say, deserts on Earth first? Make that psychologically and economically feasible first. To my knowledge, we are yet to successfully demonstrate that we can maintain a viable, materially-closed ecosystem, without having to worry about extra things we'd encounter on Mars like pressure and temperature differentials and excess radiation.
I am all for Mars, 100%. The sooner the better. But it's not like Earth doesn't have its own inhospitable climates that are vastly easier to conquer than Mars, and we haven't even managed that (but then, perhaps there's just not enough motivation?).
We already comfortably inhabit a desert, we call it Los Angleles :-) Perhaps a better example is the underground city of Australia, Coober Pedy. Non self-sustaining examples exist in the Antarctic as well.
The point being, Earth has hard to live in places, but not a lot of motivation for living in those places. Mars as a sort of sex appeal to it "Mars!" but its a valid question what happens once people figure out that its just a place that is really really far away from Walmart.
I think the point is not that it is technologically infeasible to live in deserts, but that doing it years with only what you can fit in a capsule the size of a U-Haul or dig out of the ground (with a drill you brought with you in the U-Haul) is extremely difficult.
And, that no one has demonstrated how that can be turned into a positive, long-term revenue stream, even in principle.
Camping in the Sierras as I have done over the years has lead to seeing a lot of mining towns. And a mining town in the 19th century was a modest analog of a Mars colony in that they made most of their own tools out of locally available material, lived off food that could be sourced locally, and had few support structures outside of their own ingenuity.
The argument that presumably robotic missions would pre-supply a landing site with materials that could not be locally sourced, tools for processing locally sourced materials into useful form (including but not limited to air/water). Puts what is left being there for the hell of it.
Failed earth experiments like the Biosphere ones depended primarily on self containment (good for when you are surrounded by vacuum), successful experiments like the Russian 'mobile home from hell' [1], and some things in the middle like 'earth ships'[2] actually point to feasibility.
What is lacking is the certainty that you will die if this doesn't work. And that, is pretty hard to simulate on Earth because there is always an "option" of some sort for rescue.
The Mars Society has already been doing this with a habitat simulation in the Canadian arctic[1]. They're also talking about doing a reality TV show[2] and you can try to sign up as a volunteer crew member[3].
I recall a pretty good article shared here a while ago about how Mars makes no sense as a target compared to Mercury. You can put people on the dark/light edge of Mercury and find a pretty good temperature. A tiny solar panel can generate gobs of power. There are minerals worth mining on Mercury.
Because Mars has no magnetosphere all habitable structures will have to be well underground to protect against radiation. It's really not more habitable compared to other solid bodies as you might expect.
"Because it has almost no atmosphere to retain heat, Mercury's surface experiences the greatest temperature variation of all the planets, ranging from 100 K (−173 °C; −280 °F) at night to 700 K (427 °C; 800 °F) during the day at some equatorial regions." — From Wikipedia.
That doesn't really sound like pleasent temperature fluctuations or something that is easy to work with?
I am all for Mars, 100%. The sooner the better. But it's not like Earth doesn't have its own inhospitable climates that are vastly easier to conquer than Mars, and we haven't even managed that (but then, perhaps there's just not enough motivation?).