Just a guess as to the final recommendations (in progress)

1) Unless there are very good reasons, no shuttle flights should be made except in station support, or else in orbits where Abort to Station is a option. Tile inspection should take place on either/both approach or departure as part of standard station operation.

2) Emergency overflight rights should be established with Russia and China, or if they already exist, confirmed. Modified entry angle, attitude, and dogleg re-entries need to be examined and pre-programmed for onboard use, and pilot instruction enhanced to include these cases.
If the vehicle can not be saved, a goal here should be to at least get the shuttle down to speeds safe enough for the crew to bail out.

3) A safe bail out indicator needs to be installed in the crew compartment hatch. In this context, perhaps the costs and benefits of reduced crew size and ejection seat installation should be examined once again, and/or the use of stowable side-ways ejection seats for the mid-deck.

4) On board emergency CO2 scrubbers, candles, or other systems need to be enhanced to increase shuttle life time on orbit to the ammount of time necessary for a rescue shuttle to be dispatched in case of failure.

5) Lift off debris cameras were not considered an essential part of flight operations, and they were not given redundancy. Clearly the camera failure contributed to the faulty estimate of the object which struck Columbia as foam and not ice. In other words, because of a failure to ensure redundancy of launch camera systems at a nominal cost, a $3.5 billion spacecraft and 7 crew were lost.

Duplicate cameras need to be installed at all launch camera locations.

6) The line on the external fuel tank needs to be moved to prevent ice accumulation in a position where it can detach and strike the orbiter.

7) The panel with the sensor indicator lights needs to be redone, if it has not been redone already. In this case, key sensors had gone off nominal, and the pilot had not noticed, but instead his attention was called to the failures by controllers. The indicator lights need to be displayed by sensor location on the shuttle, and in a more central position, perhaps in several displays broken down by fault light system function.

(If LEDs can be rated to a space environment, the replacement of light bulbs with brighter LEDs should be looked at, for power conservation reasons if no other.)

8) The failure to note fault sensor lights seems to indicate that the shuttle commander and pilot were not adequately informed about the possibility of tile damage. I don't know if there's any one method or several methods to improve listening and communication skills in manned operations, including station, but clearly improvements need to be made.

9) Shuttle hydraulic systems need to be isolated/replaced so that single failure does not lead to total hydraulic system failure.

for all you playing along at home, that's my view right now...

now for today's news
ep