- From: Poul-Henning Kamp <phk@phk.freebsd.dk>
- Date: Sun, 10 Jul 2016 09:45:13 +0000
- To: Julian Reschke <julian.reschke@gmx.de>
- cc: Phil Hunt <phil.hunt@oracle.com>, HTTP Working Group <ietf-http-wg@w3.org>
-------- In message <5cdf0fa8-063c-7eaa-a9e3-fb6db7417254@gmx.de>, Julian Reschke writes : >> I would go as far to make all repeat header a failure condition under the >> new format. > >I'm not sure how that would be helpful, nor how that would be legal wrt >the base spec. In the case where we augment the existing spec with a new way of encoding the content in (new) headers, we can specify that repeat (new) headers encoded that way are illegal. That doesn't change anything in the base spec, it merely imposes a tighter restriction for a subclass of (new) headers. In the case where we implement a new semantic layer, that will by and large be a replacement of the base spec, so we're free to do what we want, subject to the necessary interop/compat considerations. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence.
Received on Sunday, 10 July 2016 09:45:39 UTC