>2. If the IOTA project wishes to ensure that trinary logic is involved in the proof of work and signature hashing process we suggest that a trinary function co

>2. If the IOTA project wishes to ensure that trinary logic is involved in the proof of work and signature hashing process we suggest that a trinary function could be composed with a secure hash algorithm in a construction such as: >Hash(msg) = MD6(msg || TrinaryFunction(msg)). Haha this is a wonderful suggestion. "If you weirdos think ternary arithmetic has any benefits, just combine it with a crypto primitive you didn't handroll, ya dingus" And then they say it's okay because of "higher level checks" that a colliding transaction wouldn't validate, so then when one is demonstrated, they say "it's okay because we'd just decide to reject the bad one even if the hash is the same" and the final defense is "something something distributed ledgers Satoshi"