The safe return of Discovery and her seven-person crew from NASA’s first space shuttle mission since the Columbia disaster is as much an occasion for review and reflection as it is for celebration.

In many, if not most of its tangible aspects, the STS-114 mission was a resounding success. All key mission objectives — demonstrating on-orbit shuttle repair and inspection techniques, replacing balky attitude-control hardware on the international space station and resupplying that facility — were achieved.

Cameras installed on the ground and on the orbiter to monitor launches and inspect for the types of debris damage that doomed Columbia performed spectacularly, providing dramatic new perspectives of the space shuttle in flight. In fact, the images led to an unplanned spacewalk in which astronauts performed the first-ever on-orbit repair to the heat-resistant shielding on the underbelly of a space shuttle.

But the mission also raised a number of troubling questions. They start, of course, with the foam-shedding incidents, and end up on the more fundamental issue of the NASA culture, which was cited by the Columbia Accident Investigation Board (CAIB) as a major underlying contributor to the disaster.

The Discovery mission also gives NASA officials some things to think about as they edge the shuttle fleet toward retirement and begin work on a replacement vehicle that would ferry astronauts to and from the space station starting next decade.

There is little to be said specifically about Discovery’s foam-shedding incident that hasn’t already been said, but some things nonetheless bear repeating. Suffice to say that NASA engineers, despite two and one half years of intensive study, still have not come to fully understand the phenomenon, let alone correct the problem.

The largest piece of foam broke off Discovery’s external tank in an area that had undergone an impromptu repair prior to the launch. While this raises the possibility that the incident was an isolated event, NASA officials were not prepared to draw that conclusion during a press conference Aug. 11.

In fact, these officials suggested that the repair alone probably could not explain the shedding. Further, the pre-launch repair does nothing to explain why other dangerously large foam chunks shook loose from elsewhere on Discovery’s external tank during the liftoff.

NASA has now acknowledged what most observers concluded soon after the Discovery foam-shedding came to light — that the Space Shuttle Atlantis will not make its September launch window. How long the fleet will remain grounded is anybody’s guess, but it probably is fair to assume the shuttle’s manifest will shrink because of this and other delays that likely await the program.

NASA’s international space station partners, nervous even before Discovery’s flight about getting their hardware launched , now have even more reason to worry. The European Space Agency has gone so far as to begin studying scenarios in which its Columbus module does not launch, a possibility that would have major political ramifications and affect other programs such as the Automated Transfer Vehicle space station cargo carrier .

Questions also remain about NASA’s culture, and whether it has changed to the degree that the CAIB report said was needed to reduce the chances of catastrophe in the future.

The CAIB focused heavily on communication — or rather the lack thereof — between NASA technicians and managers as a factor that caused agency officials to miss potential opportunities to learn of the danger Columbia’s crew was facing. There is no evidence whatsoever that NASA officials ignored or dismissed potential safety concerns leading up to or during Discovery’s flight — in fact the opposite is true.

But the CAIB report’s assertion that NASA suffered “blind spots” in its safety culture resonates today given the agency’s inability to judge its understanding of the foam-shedding problem. It is possible that NASA in this case fell into the potentially fatal trap of overconfidence.

The CAIB report also cited the political, budgetary and schedule pressures faced by NASA’s shuttle program. And while it is clear that NASA was given as much time and money as it felt it needed to return the space shuttle to flight, the years leading up to the orbiter fleet’s 2010 retirement are almost certain to ratchet up these pressures, not alleviate them.

Cultures do not change overnight, however. And in any case, NASA’s biggest problem stems not so much from its culture as with the space shuttle itself. It may well be a technical marvel, but over the years the orbiter also has proven to be temperamental, unwieldy, enigmatic and, yes, dangerous. This is due in large part to the design decisions and compromises necessitated by NASA’s misguided desire during the early 1970s to build a vehicle that would almost literally be all things to all people.

The good news is NASA seems intent on avoiding these mistakes on the shuttle’s replacement and other hardware it will need to fulfill its presidential mandate to return astronauts to the Moon by or before 2020. The current plan for the Crew Exploration Vehicle, as well as the separate rockets that will be used to launch it and heavy loads of Moon-bound hardware, is to focus solely on NASA requirements. NASA also has correctly decided to separate out the crew and cargo missions, thereby putting astronauts at risk only when they are in fact the cargo.

For the time being, however, NASA appears stuck with the space shuttle. Barring another accident, retiring the orbiter before 2010 does not seem feasible unless NASA is prepared to betray its international partners on the space station and effectively throw away the tens of billions of dollars U.S. taxpayers have invested in that program. Eliminating the foam-shedding problem once and for all is the best way for NASA to prove that it is ready to take on the challenges of the post-shuttle era.