Continuing this sad tale, we saw the loss of a launch because of faulty O-ring design caused by small, but significant, warpage from the weight of the vehicle resting on its side during the O-ring installation.
That's not what caused the O-ring failure, and the vehicle was attached vertically in the VAB, well after the SRBs were fully assembled and mated to the tank. The temperature at launch was below freezing, and about 25 degrees lower than any previous launch. The O-rings lost most of their flexibility due to the cold and failed to seal the joint as a result.
Sort of both and then some more.
The O-rings did fail due to loss of flexibility, but root problem was that the design of the joint was defective.
The joint design parameter was that the O-rings should not come into use - the joint should close and keep the gases from the O-rings. After all, they are "rubber" and rubber does not live long when exposed to hot high pressure gas.
The O-rings were to be backup, and in a perfect design backup doesn't come into play. There was also putty in the gap that should prevent the combustion gases from reaching the O-rings, but maybe failed as well.
When the joint was designed, it was believed that combustion pressure would cause the joint to close and form a tighter seal, but static pressure-testing showed that pressure caused the joint to rotate open and expose the O-ring to combustion gases. O-ring erosion was seen from the very beginning of the flights due to this flaw, (not all flights)
Low temperatures exacerbated the problem, the O-ring worked so long as the gap in the joint was NOT too narrow. Too narrow a gap and the O-rings can't move freely as the SRB and joints are flexing, and with the cold making the O-rings stiff, they failed to move and close in the place where the gap was most narrow.
As rickyslashdot pointed out, the motors were out of round due to shipping constraints, and it by coincidence happened that the place most narrow pointed at the hydrogen external tank. Also, when the main engines are lit before takeoff, the shuttle tilts forward and bends the solid rocket boosters. the assembly then bends backwards and oscillates at 3 cps. Oddly enough, this is when the SRB experiences the greatest mechanical stress. The puff of smoke seen on the pad at ignition happen at 3cps, so there's another factor in stressing the joints and requiring the O-rings to be flexible.
Thiokol began redesigning the joint in 1985, but didn't get it into production in time to prevent the Challenger disaster.
Here's a link to the commission's report. It has much detail and pictures, and fills in the gaps in my over-simplified exposition.
Here's the conclusion from the report:
"In view of the findings, the Commission concluded that the cause of the Challenger accident was the failure of the pressure seal in the aft field joint of the right Solid Rocket Motor. The failure was due to a faulty design unacceptably sensitive to a number of factors. These factors were the effects of temperature, physical dimensions, the character of materials, the effects of reusability, processing, and the reaction of the joint to dynamic loading. "