Re: FYI... Binary Optimized Header Encoding for SPDY

On 2012/08/06 8:22, Adrien de Croy wrote:
>
> personally I see little value in allowing a "safe" character to be
> encoded with %, and specify it has no semantic meaning (that it is
> encoded rather than "native")
>
> Why not simply deprecate such things for 2.0, and when it comes to
> putting together a 1.1 message from a 2.0 message, it needs encoding at
> that stage, and at that stage, there's only 1 allowed way to do it, e.g.
> safe chars MUST NOT be %-encoded etc.

It's possible for HTTP to forbid such useless %-encoding. Please note 
however that URIs (and IRIs) allow it, and HTTP can't change that.

Regards,   Martin.