Now I ask, wouldn't it be better if we went straight to the ASIC friendly, but easier to commoditize mining chip production algorithm, than going through hardfork after hardfork?
Agreed. Monero's idea of slightly altering the PoW parameters every X months is just whack-a-mole, and doesn't sound like a proper development plan for any serious tech person. On the plus side, at least they are admitting that the PoW they devised to be ASIC-resistant in the past isn't very resistant now, because (omg) technology moves on. (Gosh, is it really 7 years since we developed the first Bitcoin FPGA miners? I still have that dev board lying around.)
For something arguably more ASIC-resistant, Boolberry has had its Wild Keccak for a few years, but I guess it's NIH for everyone else. Zcoin's upcoming MTP also looks interesting with its multi-gigabyte memory requirements. But even they won't be ASIC-proof forever -- remember when 640 KB was enough for everybody?