Actually, Kennedy's speech was in 1961, not 1962, but your point remains that a definite timeline that's within the foreseeable future is desireable. However, there is a problem if all we do is declare the intention to accomplish X by date Y - all we do is build a minimal solution to accomplish X. Further, "Go Fever" kicks in resulting in the kinds of shortcuts and shoddy work that killed the Apollo 1 crew.

As impressive as the Saturn V was, it was essentially the minimal vehicle necessary to land 2 men on the moon with enough supplies to stay for about 3 days. It would've had to be upgraded to increase the lift capacity had they wanted to do a more ambitious mission. Likewise, the Lunar Module, Command Module, and Service Module would've required upgrades to handle longer missions.

I'm in favor of going to Mars within a reasonable timeframe, say 20 years. Based on decades of flight and space test experience, that time would be sufficient to adequately test the systems before launching people on a years long mission. There's a reason why companies like Boeing don't carry paying passengers on a new plane's first flight (or even the 100th flight). Decades of blood-soaked experience have shown that incremental testing is the way to go. Ignoring that testing experience to meet an arbitrary deadline or to rush things results in dead people.

There's a reason why NASA created the Gemini program (they'd originally planned on going straight from Mercury to Apollo). Gemini allowed them to develop the technology and experience necessary for maneuvering in space, rendevous and docking, longer duration missions, and EVAs. All of those skills were vital for Apollo.

Even the Apollo program used incremental testing, with each mission building on the last. Apollo 7 tested the CM and SM in Earth orbit. Apollo 8 sent those components to lunar orbit, allowing test of the Deep Space Network and translunar navigation. Apollo 9 tested the LM in Earth orbit, and Apollo 10 took it to lunar orbit as a dress rehersal for Apollo 11. Apollo 12 increased landing accuracy and stayed on the moon longer, Apollo 14 stayed longer and added the "golf cart" to carry tools. Apollos 15-17 each carried the Lunar Rover and stayed on the moon for longer periods. Those were the most fruitful missions from a scientific standput.

I very much opposed to another "flags and footprints" mission even if the destination is Mars. To me, that would be a waste of resources. Properly structured, the program can develop the capability to perform several space missions, not just going to Mars. Quite a few of the systems necessary for a Mars mission could be tested on other missions, such as further exploration of the moon and/or visiting an NEO. By the time we're ready to go to Mars, the systems would've been proven and the crews had the opportunity to gain a lot of exploration experience. This would not only increase the chances of success, it would allow the infrastructure to do many things in space, not just going to Mars.

If 20 years seems like a long time, I speak from professional experience that it really isn't when you're talking about space systems. I'm working in a military organization that hopes to influence how the government does space in 2014. The National Security Space Architect office's National Security Space Plan is looking to 2025. Unless you're willing to throw massive amounts of money at the program, developing even relatively simple space projects in the government domain is a years long process. Even now, it's too late to affect the FY 2006 budget cycle (at least for the DoD), so the soonest we can hope to start a new project is the FY 2008 cycle.