Will either SpaceX or Rocketplane-Kistler be able to fly their COTS spaceship before the end of 2010? That's a good question, but the reality is that NASA will need the services that only COTS can offer. The Shuttle's lift capability will be lost when that system is retired in 2010, and Orion won't be ready for ISS missions until 2014. Only COTS is able to fill that 4-year gap, but the current schedule projections do not look promising for that to happen.
I believe that the RpK K-1 vehicle is a long-term answer to the COTS challenge, but one that should not be rushed. It's the first fully-reusable, orbital spacecraft that has a realistic chance of success, and we should allow it to develop according to its own schedule (consistent with the funds that RpK can raise to complete it.)
SpaceX's Dragon capsule, on the other hand, is a lower risk development that recalls the heroic age when men strapped themselves on top of Atlas and Titan II rockets. Still, it's a challenge for a small company for SpaceX that has already committed itself to building the Falcon I, V, and IX rockets.
My suggestion to Space X is to adopt a spiral-development philosophy for the Dragon, and to take drastic measures to acclerate the schedule. Specifically, I think that the "Spiral 1" version of Dragon should be designed for launch on the Atlas V rather than the Falcon IX. My logic is that SpaceX will not be able to ready both the capsule and the rocket for flight in 2010. By putting all of its resources behind Dragon and delaying the Falcon IX, SpaceX might just have a shot at meeting the 2010 schedule target. If Dragon succeeds, SpaceX can focus on renewed Falcon IX development and a "Spiral 2" Dragon that will fly on the Falcon IX.
Atlas V has some important advantages over Falcon IX. Most importantly, Atlas V already exists as flight-qualified hardware, and it has a good (albeit short) flight history. Also, Lockheed Martin has expressed its willingness to work with anybody who proposes a capsule to fly on the Atlas V. Presumably, this would include a competitor in the launch vehicle field like SpaceX.
While SpaceX has reasons to favor Falcon IX due to their projected costs, I feel that this may be a bit misguided. Within the launch vehicle industry, there are very few who believe that SpaceX's rocket engineering
will lead to cost savings. The final Falcon IX will probably cost almost as much to build as an Atlas V 401. The real savings come from lean management, automated processing that reduces the size of the rocket's standing army, and a high flight rate that results in economies of scale (like spreading the standing army and management costs over many more launches, as well as mass-production of components in the rockets.)
There will be consequences if SpaceX delays Falcon IX development (which is my suggestion, not SpaceX's.) Falcon IX already has a list of potential customers which includes Bigelow, MacDonald-Dettweiler, NASA, and "US Government" (probably NRO.) NASA and "US Government" will probably be able to move to alternate launchers without too much problem, but MacDonald-Dettweiler and Bigelow will be pretty annoyed, to put it mildly. If NASA's COTS development takes precedence over Falcon IX development, SpaceX should use NASA to help persuade the commercial customers to move to other launchers. SpaceX will also have to take bold steps to bring potential customers back after Falcon IX development is complete.
SpaceX has a lot on its plate, and I believe that the company will have to examine its priorities. Is it really a priority to "reinvent the wheel" and come up with a competitor to the government-subsidized Atlas V? Or does SpaceX want to make the most of NASA's unprecendented COTS program and use the opportunity to develop the first private-sector orbital manned spacecraft? The window of opportunity for COTS is small, and if SpaceX can't get Dragon to fly between 2010 and 2014, NASA will give the space station resupply mission to the business-as-usual Orion capsule and its absurd Stick launcher.