The following paper proposing an asymmetric memory-hard PoW had been
recently published:

http://eprint.iacr.org/2015/946.pdf

My intent is not to promote the paper as I have not finished studying it
myself. I am however interested in the dev-list's stance on potentially
altering the bitcoin PoW protocol should an algorithm that guarantees
protection from ASIC/FPGA optimization be found.

I assume that, given the large amount of money invested by some miners into
their industrial farms this would represent a VERY contentious hard fork.

It is, however, also true that a novel optimization-resistant algorithm
could greatly ameliorate decentralization in the bitcoin network due to a
resurgence of desktop/cellphone mining.

Where do the core devs stand on this matter, hypothetical as it may be?

Dpinna