There has already been a lot of discussion of what's good, bad, and bogus about the plan. Although the Apollo-capsule-like form has been defended as being "convergent" (it was a good solution in 1969, it's still a good solution), it's still hard to believe this is the best NASA could come up with if its goals were to do the best job possible in building up the infrastructure for future space development, rather than the best job possible consistent with keeping the same level of employment at the agency and at its traditional contractors (with due consideration to getting dollars into key Congressional districts).
That's part of the argument made by Greg Burch in an excellent post on his blog ("The Trouble with Orion," September 2). As a partial counterpoint to this, here is a new Technology Review interview (9/11/06) with a Lockmart business development manager for the Orion project. She says the things you would expect, including the following that seems to echo Greg's point about NASA's fear of a "staffing valley" between the Shuttle program and its successors:
When it gets right down to it, NASA is signing up for a relationship with an industrial partner that's going to last a couple of decades. They wanted to know that it would be a happy marriage, where the spirit of partnership was in real evidence. During Phase I [when NASA paid several bidders to develop designs for Orion], we took the initiative to make sure our project office was co-located in Houston, which made it easy for them to participate in all of our control board meetings and other important events over and above the typical bimonthly reviews. We've got a significant workforce at the Michoud Assembly Facility in New Orleans [where the Shuttle's external tanks are put together]; we made a decision early on to do final assembly and checkout at Kennedy Space Center; we're going to be doing engine testing at Stennis Space Center in Mississippi [NASA's primary rocket propulsion test site]. I think NASA has appreciated that.