Originally Posted by
Old Smokey
The AoA indicator is the only suggestion that bears some merit.
Of course those ice up too, if things aren't going well. And it's hard to provide accurate airspeed to AOA relationships at high speed, because the change in AOA for even moderate speed changes is small, and you start getting into instrument accuracy ... it's hard to get much under 0.5 deg total AOA system accuracy, but 0.5 deg out at Vmo-type speeds is a LOT of speed shift. AOA isn't bad at low speeds, but its pretty useless at cruise speeds. Especially if you need accurate data for other systems, like rudder limiters and the like.
AOA alone at low speed has its own problems too - with just AOA it's harder to know how oversped one might be - so landing distances might become more problematic. Also, failure cases where we today routinely apply speed increments become more interesting when we have to think in AOA increments. Not saying thats impossible, but its certainly different.