Dan Schrimpsher pointed out in comments to my monculture post one of the interesting ideas that has gained a little too much traction lately at NASA–the importance of not having cargo and crew on the same vehicle. Apparently, most people now believe that one of the big reasons why Shuttle is so expensive is that it tries to be both a delivery truck and a passenger van. While there is some real grain of truth in that idea, I think it is missing a subtle point that I’ll try to explain below.
The technical problem I see with the shuttle isn’t that it combines crew and cargo per se. The technical problem that I see with the Shuttle is that it tries to be an earth-to-orbit crew transport, while simultaneously trying to be a heavy lift cargo vehicle while simultaneously trying to be a science lab, while simultaneously trying to be a space motel, not that it has both crew and cargo onboard. Most other manned vehicles to date have also carried some cargo. Soyuz carries several hundred pounds of cargo, Apollo did too. I’m not sure about Gemini, but imagine it had the capability. The ability to bring some cargo with crew is important, a Toyota without a trunk would be a relatively useless vehicle.
In fact, to me, a cargo delivery ship is begging to have at least two or three crew members on board to make the whole thing simpler–automated deployment of satellites without any people there to help is a major expense in the design of satellites for example. Cargo jets do fairly well mixing cargo and a little bit of crew, while passenger jets do well with lots of passengers and a little bit of cargo. I think it’s when you try to combine large amounts of crew, huge amounts of cargo, and long-term habitation facilities into a vehicle that it starts getting bloated.
Another part of this belief in a separation between cargo and crew is the belief that a crewed vehicle needs a higher safety level than a vehicle that is just for hauling cargo. I’m not sure how true this really is. How much happier would the people underwriting satellite launch insurance be if they knew that even in a bad situation the satellite will probably land intact and just be launched again? Guess what, when insurers are happier, eventually they start charging lower premiums. Who really would want to fly their expensive satellite on a vehicle whose designers wouldn’t trust flying themselves on? There isn’t much of alternative now, but that speaks more to the immaturity and lack of alternatives in the current launch industry than anything else. Whatever you’re flying, the ability to do an intact abort with the full cargo and crew complement (and the vehicle too if possible) is really important to actually making a profit in this field.
There are some additional nuances to be mentioned, but I think that I’ve said enough to make my point here.