Originally Posted by
msbbarratt
To an outside observer this all seems like pilots sometimes have to infer important information (e.g. the pitots are iced up) indirectly and somewhat subjectively from other data.
Not really - the point is that as long as procedures (e.g. automatics off, stabilise flightpath) are followed, it's a fairly simple and straightforward process of elimination. More to the point, if it is an ice crystal blockage, then there's a good chance it will have cleared by the time the aircraft is stable. Apropos of nothing, I'm inclined to think that inferring important information from other data (e.g through the instruments) is part and parcel of what a pilot is expected to do as a matter of course!
Additionally, I'm pretty sure that as far as the A330 is concerned, the only incidents where two or more pitot sensors froze over involved the Thales AA probe - which was being taken out of service even before AF447 happened. After AF447, I believe the Thales AB model was also withdrawn from service and all aircraft retrofitted with the Goodrich probe.
[Pitot-mounted cameras] - Crazy idea? Unnecessary idea?
Not so much that as a case of the "using a sledgehammer to crack a walnut" analogy. Being a techie myself, I'm well aware of a natural inclination on our part to want to find technological solutions for a problem right out of the gate - as such, one of the things drilled into us by our Software Engineering teachers was to recognise that tendency and give the matter serious thought, because additional complexity means additional components capable of going wrong.
In this case, the rarity of even one of three pitot tubes icing over is a strong indication that adding complexity for its own sake would be the wrong thing to do. You also have to bear other things in mind - for example that the conditions in which icing occurs increase the likelihood of obscuring a camera's view, or that UAS can also be caused by a fault in the air data computer module.