Short of asking the computer to record a "burning smell", then tag it and transmit to mx, the question remains in general what brought this a/c down in a "seemingly" familiar way to other incidents? I am curious about the logic re: pitot sampling. With three pitot heads there are three possible pairs of "source", but when one head "fails" there is only one "source" left since the computer will fail both remaining if there is a "disagree?" Besides, why are pitots still "pointy"? Ice starts at the sharp bits.
Interface. The weakness as I see it. At the very least, I see the foundation as basically backwards. At a critical time in terrible weather at night, I don't want the box to fall all over itself telling me what's wrong, and what to do to fix the computer. I want the box to give me a few crucial cues so the a/c can continue to fly. In other words, give me the good news about what is working, not why the computer has to flop the a/c into my lap. The important cargo is not the computer.