@lcamtuflcamtuf
I know IEEE 754 NaN isn't reflexive because it represents an error condition. Is that true for Inf as well? I guess I could look for docs, perhaps there's a whole class of error values in floating point?
If you have a fediverse account, you can quote this note from your own instance. Search https://recurse.social/users/shapr/statuses/114654388056286481 on your instance and quote it. (Note that quoting is not supported in Mastodon.)