The first human interplanetary mission will, more likely than not, go to the Red Planet. Apart from the Moon it is the easiest major body in the Solar System for us to reach. Its surface bears evidence of rivers and seas, and liquid water may still occasionally flow there. The similarity of its landscapes to the American Southwest is an illusion, but it remains a candidate for life, and its chief rivals - Europa and Titan - are far more distant and hard to reach.
A single image (of an artifact, for example) could of course change all this. But for now at least, Mars is the likely first place to go.
I am not concerned with pure stunt missions, whether by some multi-billionaire or an emergent great power. For this discussion I presume that a human mission to Mars is intended to explore Mars. Yes, other motivations will surely be in the background, involving the usual suspects. But the more serious the exploratory intent, the more likely that going to Mars will lead to further developments rather than end up as a costly one-off.
This does not mean that every individual mission must 'do science.' The engineering demands of safe human interplanetary travel justify what amount to shakedown missions, say, a manned flyby that does nothing but demonstrate our ability to perform the transfer mission. But the program as a whole should be aimed at advancing our knowledge of Mars.
The model here is not Apollo but the robotic deep space program, which in spite of some embarrassing failures (feet/second != meters/second), has in all been a profound success.
A first implication is that a human mission will be undertaken only when robotic missions reach diminishing returns. It is no doubt true that a geologist with a hammer could learn more about Mars in a day than our rovers have learned over years. But the cost of sending that geologist would be very much greater than the next few rovers. And human missions are subject to some constraints, such as landing where it is safe to land.
My instinct in thinking about a Mars program is to proceed cautiously and incrementally - unmanned tests of the vehicle, then a human flyby, then orbit without landing, and only then a full-on landing. I am not sure that each of these stage is strictly necessary, and there are alternate possibilities, such as a visit to a NEO, that would similarly exercise spacecraft and procedures.
And some of these missions - notably, orbiting without landing - might be able to 'do science' on an important and relevant scale. In particular, a crew on Mars orbit can operate surface vehicles and manipulators remotely, but without significant light lag - like steering a minisub from the surface, not steering a rover from Pasadena.
This is as good a place as any to note that 'going to Mars' combines at least two very distinct space missions: the deep space journey to Mars orbit and back, lasting at least months and perhaps most of two years; and landing on Mars, working there for weeks, and lifting back off to Mars orbit. This in itself is reason to test the Mars orbital capability before the surface landing.
Mars Direct argues in effect for a very different functional division - in effect a one way trip to the surface of Mars, then use of a second pre-positioned (and pre-fueled) spacecraft for the trip back. This feels cut-rate and precarious to me, not least because Zubrin is so much like a real estate promoter, not quite trustworthy on principle.
To be sure, there's no reason this architecture couldn't be tested in an unmanned mission, but I also have doubts about packing the hab elements into form factors suited to Mars landing and liftoff. It seems like awfully cramped crew quarters for such a long mission, or else a much larger cabin than you need to carry the crew and some rock samples from the surface to Mars orbit.
In any case, as longtime readers know, I have bias in favor of fast human travel using high ISP propulsion - most likely solar electric, though perhaps nuclear electric - for getting to Mars orbit and back. Faster travel means less exposure to radiation and weightlessness, the main health hazards of prolonged spaceflight.
Solar electric has now been successfully flight tested by the Dawn mission. A fast (~3 month) trip to Mars requires a much higher drive power/mass ratio, which may not be attainable. But solar electric space propulsion is far from technical maturity, and it may well be able to approach the 1 kg/kW standard for fast travel. .
Electric drive pretty much precludes the Mars Direct approach anyway, since an electric spacecraft is ill-suited to aerobraking, and doesn't need it. The surface components of the mission can be sent separately - a Hohmann transfer is fine, since the crew isn't aboard until the Mars landing.
The image, from a European proposal for Mars exploration, is recycled from an earlier post on interplanetary exploration.