It's the "simple" text-based nature of HTTP that keeps biting it. It's the source of parsing complexity (it's really nasty when you need to handle all the weird

It's the "simple" text-based nature of HTTP that keeps biting it. It's the source of parsing complexity (it's really nasty when you need to handle all the weird "web-compatible" edge cases and not be vulnerable to things like request smuggling). In this RFC it increases complexity due to the need for normalization.