There is an existing question about why Iota chose their specific post-quantum signature scheme. But my question is, why use a post-quantum signature scheme at all? Were the motivations discussed anywhere?
My understanding is that many of IOTA's unique design choices were driven by a need for performance on resource-constrained IoT/embedded devices. But as far as I know (perhaps not far), post-quantum schemes have worse performance characteristics than the schemes they're replacing, increasing the bandwidth and memory requirements for these devices. It may not be a huge difference, but other performance choices like the use of ternary were seen as worthwhile despite the benefits being relatively small.
(Post-quantum schemes are certainly cool. Like many people, I "independently re-invented" (ha!) a Lamport-style signature scheme when studying undergrad cryptography, and it was fascinating. But I assume "it's cool!" is not the reason here.)
What benefits justified the performance and immaturity costs of a post-quantum scheme?