We have had two catastrophic failures in a little over 100 missions, suggesting a catastrophic failure rate probability of 1:50, or 2% per mission. This is a much higher risk than has previously been acknowledged by anyone in the space program, and even by many of its critics. (Richard Feynman speculated that the risk might be as low as 1:100.)
Is it an acceptable risk? Everyone understands that manned space flight is inherently dangerous, and that it is probably impossible to drive the risks down to zero. But the real question is whether or not there are feasible, less risky alternatives.
There has been considerable research and discussion regarding smaller reusable "space planes" designed for human payloads only. Such vehicles, being so much smaller and lighter, would need much smaller rockets to lift them into orbit; perhaps the inherently dangerous combination of solid fuel rockets strapped to a huge liquid fuel tank could be eliminated altogether. A smaller space plane might still have to be lined by tiles, but there would need to be fewer of them, and thus fewer points of vulnerability. If a smaller space plane was our primary manned vehicle, the system could be designed so that each mission would include at least the capability to rendezvous with the ISS. This would at least allow a close visual inspection of the tiles by ISS personnel before a re-entry was attempted. Should the space plane be damaged, they could use docking hardware (which would be standard equipment on all space planes) to dock with the ISS and use the ISS as their "lifeboat" until a rescue or repair mission could get to them.
As for bulky and heavy cargo, this should be boosted into orbit on unmanned rockets. Given this tried, true, and reasonably economical option, I have trouble seeing any good reason to put astronauts lives at risk just to haul cargo into orbit. There may indeed be things that need to be done with cargo once it is in orbit: satelites need to be deployed, lab experiments need to be tended to, etc. We also need to have a repair capability in space. Thus, a space plane needs to be equipped with a remote-control arm, and also an airlock so that astronauts can perform EVAs. A lighter spacecraft with more fuel on board might actually be able to climb into higher orbits than is usually done with the shuttles, thus making it more useful for a wider range of missions.
Would it be more costly than the shuttle? I am not so sure about that. Sending bulk cargo into orbit on unmanned boosters is clearly more economical than using the shuttle. A smaller space plane would require smaller, fewer rockets, yielding some small savings there. It is possible that a smaller, simpler space plane system might result in a quicker turnaround time between missions, thus requiring a smaller fleet of vehicles. A smaller, less complex spacecraft should cost a little less to build.
I am forced to conclude that the shuttle is by no means the best possible solution. Less risky, and possibly less costly alternatives should have been researched, developed, and implemented years ago. Time is past due to seriously look into them now.
When I'm cynical,
I see our space program as
our generation's
version of things like
Europe's gothic cathedrals
or the ancient world's
famous pyramids --
that is, they are giant scale,
awe-inspiring
projects that busy
millions of people, funnel
endless streams of cash
through entrenched rulers,
and only must accomplish
just "being awesome."
(This view would explain
why big, heavy and flashy
things like the shuttles
beat out logical
systems such as you suggest,
separating freight
from people lifting.
The political pluses
of a "flashy show,"
labor intensive
system with few (or no) goals
makes me think we won't
see a rational
space program in our life time.
Not in our country
where political
considerations define
all social business.)