Frank Sietzen suggests that another reason to keep the Constellation program unchanged is to keep from undermining NASA's "rocket-making reputation" : Reinventing NASA's Rocket Reputation - OnSpace/Aviation Week.

After the enormous shortcomings of the Shuttle program compared to its original goals (e.g. weekly flights) and the failures of X-33, X-34, X-38, SLI, OSP and after all the problems with Ares I, I didn't know that NASA had anything more to lose when it came to its rocket making reputation. This reminds me that in the early period of Mike Griffin's tenure, he would mention the need to nurture NASA's in-house rocket making skills as one of the reasons for not turning launcher development over to the private sector. (Besides those safe, simple, soon reasons.) So for ~$40 billion dollars, we got a great training program for NASA engineers.

The Ares I/V/Orion projects are certainly ideal for NASA as an institution. They will keep its centers busy and lots of NASA employees well paid for a couple of decades at least. However, when it comes to making progress towards becoming sparefaring, these projects are a staggering waste of money and opportunity. They will take nearly forever to develop and when they finally fly they will be stupendously expensive to operate. While there are some ancillary systems of interest in development by Constellation, particularly with respect to lunar hardware, the launchers provide no technological path to lower cost transport from earth to LEO.

Instead of focusing on what it does worst, i.e. rocket-making, NASA should concentrate on what it does best: in-space operations and assembly. In the Hubble repair and upgrade missions and the construction of the ISS, NASA has displayed spectacular skill, knowledge, and adaptability. Such capabilities are essential to genuine spacefaring and they match perfectly with the currently available medium sized launchers. Very elaborate and ambitious space systems can be built from medium sized modules. Propellant depots, which will have a tremendously positive impact on in-space operations, can be supplied by such vehicles. This approach leads to high launch rates, which will bring down costs and raise reliability and safety.

The loss of NASA's in-space assembly and operations skills would be comparable to the loss in capability that the agency suffered when it gave up Saturn and the lunar hardware at the end of the Apollo program. I hope the Augustine committee doesn't let this happen.