Re: draft-ietf-httpbis-http2-latest, 5.5 Extending HTTP/2

On 14 July 2014 13:09, Kari Hurtta <> wrote:
> So it is implied (but not said) that implementations MUST NOT generate
> undefined pseudo-header fields.

The doc says:

"Implementations MUST NOT generate header fields that start with a colon"

How is that ambiguous?  Would you prefer "endpoint" over
"implementation"?  Is that it?

Received on Monday, 14 July 2014 20:40:27 UTC