- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Mon, 04 Jun 2012 18:16:17 +0200
- To: "Gordon P. Hemsley" <gphemsley@gmail.com>
- CC: ietf-http-wg@w3.org, Yves Lafon <ylafon@w3.org>
On 2012-06-04 17:59, Gordon P. Hemsley wrote: > ... >>> Are there reasons not to do this? >> >> The order is not significant. It never was. If w etry to help broken >> recipients we will break senders that do not sort. >> >> That's a normative change for which I simply see no compelling reason. > > Well, I'm confused: I thought SHOULD meant that it was preferred, but > not required. How would that break senders who do not sort? > ... They wouldn't be conforming anymore; you may want to read RFC 2219 and the definition of "SHOULD". BTW this exactly why we're trying to reduce the amount of SHOULDs in the spec. Best regards, Julian
Received on Monday, 4 June 2012 16:45:23 UTC