Closed stephank closed 6 years ago
This actually violates RFC 5322, which says that the case-sensitivity of the local part is host-dependent (and thus we can't assume case insensitivity). However, as we know, almost every host treats is as insensitive, so we're probably good in practice.
I was looking for some earlier discussion around this, but can't really find it. It would be nice to have some mention of the RFC in the document, and why we deviate from it.
Technically, we've been doing it this way from the very beginning. It's definitely something I personally feel strongly we should keep doing, because it'd break a lot of expectations for our current users if we don't. (At Angry Bytes, but I imagine in general too.)
Yeah, I agree with you. Everyone treats the local part as case insensitive anyway.
Alright, I have a verbal ok on this one from @onli, and the related PRs are all approved. I'm going to merge this work, make a broker 0.2 release, and update the public broker. :)
Specs.md
listing our own specs.Specs.md
listing terminology.