NASA dropped the methane engine requirement but didn't say that the competing companies couldn't still propose it. From a near-term schedule and technical risk perspective, I don't have a real problem with that. They're trying to get the CEV in action by 2012 but they aren't talking about going to the moon for several years after that (and Mars even later).

If they mandate methane engines for the first iteration CEV intended for flights to the ISS, they're putting that technology on the critical path. Rightly or wrongly, NASA might consider that too risky from a technical perspective for the 2012 timeline. However, that isn't an irreversible decision for later vehicles intended for the moon and Mars. Assuming lunar ice exists and is harvestable (something we don't know yet), then having methane engines makes tremendous sense for moon missions but only after we have a way to harvest and process the ice into methane and LOX. I doubt that capability would be ready by the first landing mission, or even the 10th.

There is some historical precedent for a two-phase CEV production but it isn't necessarily a good one. When it was started, there were two versions of the Apollo capsule, a Block 1 and Block 2 design. The Block 1 design was intended for the first flights to be followed by the more capable Block 2 design for the lunar flights. However, the Block 1 Apollo capsule was a can of worms with reports of very poor quality control that led to the Apollo 1 fire that killed Grissom, White, and Chaffee.

You don't necessarily have to build in every capability that you ever anticipate needing in time for the first mission. Instead, an iterative (spiral) development approach like what is done for software and military aircraft might be a better strategy. Using fighter planes as an example, you have two extremes with the F-16 and the F-22. The F-16 is a classic example of iterative development with the early models having basic functionality that was added to over time to create the much more capable planes we have today. From the initial request for proposals to Initial Operational Capability (IOC) for the F-16 was something like 5 or 6 years - very fast by modern standards. By contrast, it seems the Air Force went for an "everything but the kitchen sink" approach with the development of the F-22. The first planes now entering service has been in development for over 20 years at tremendous cost.

Despite the Apollo mismanagement, it might not be a bad idea to consider a Block 1 and Block 2 CEV (perhaps Block 3 for Mars missions). The Block 1 vehicle would have the initial capability for flying Earth orbiting missions. Later, the Block 2 vehicle would have the enhancements need for the lunar missions like an upgraded heat shield, better navigation, enhanced radiation shielding, etc. By then, having a methane/LOX SM and LM ascent engines would make a lot of sense. The Block 3 CEV would be enhanced even more for the Mars missions.