Ah, search engine indexing does seem like a big deal. That never even crossed my mind. I was just doing some reading on rel="canonical" since I’d never seen that before, and it certainly is interesting. It seems like this is a valid “workaround” of sorts if you can’t / don’t want to redirect, so maybe that’s why Amazon doesn’t get penalized much (besides their size).
For Discourse, I imagine it could be difficult to use effectively (assuming you even wanted to) since each post has its own URL and theoretically should be separately indexed.
Anyway, thanks for the explanation. 