Even here in HN, some obviously true statement like "if one wants better resistance to flaws in hashing algorithms, one should XOR the results together of different algorithms, since then to break the whole every individual part must be broken", gets serious pushback as being unnecessary.

I suspect that pushback comes from NSA shills who don't want to see it becoming standard practice to stack encryption or hashing in a way which makes it harder to break.

If you had stacked MD5 and SHA1, it wouldn't be broken today!

And that's why I am so skeptical of new post-quantum protocols. They are much more complex than current protocols which creates much bigger surface vector for somebody trying to sabotage them or just unexpected mistake.