- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Sun, 11 Jan 2015 10:41:16 +0100
- To: Bjoern Hoehrmann <derhoermi@gmx.net>
- CC: RFC Errata System <rfc-editor@rfc-editor.org>, fielding@gbiv.com, barryleiba@computer.org, presnick@qti.qualcomm.com, mnot@mnot.net, bjoern@hoehrmann.de, ietf-http-wg@w3.org
On 2015-01-11 02:30, Bjoern Hoehrmann wrote: > * Julian Reschke wrote: >> Again, it is not an error. The spec is correct the way it is. I see your >> preference to reference instead of copying the rule (and I might >> actually agree with it). > > So what is your desired outcome here? Right now anyone combining the > ABNF fragments from the HTTP/1.1 RFCs is likely to come across two > independent definitions for the `method` symbol and will have to check > what is going on. They will likely notice that the specifications do First of all, combining the ABNF is intellectually interesting, but not very useful. The use of "method" in the request format happens at a layer different from the use in "Allow". Furthermore, the check will show that indeed the ABNF is the same. So there is no problem here. > follow a clear pattern for importing rules, just not for `method`, and > then try to find out what's special about it, possibly ultimately con- > cluding that there is nothing special about it and they can treat it > as if it had been "imported" like any other rule, but they can't really > be sure of that conclusion without independent verification. I would > like the errata to provide this verification. There is nothing wrong with the spec. You are asking for an editorial change (that I might actually agree with), but the errata system is not the place to record proposals for editorial improvements. (That being said: we planned to open an issue tracker for the next HTTP/1.1 revision, and *that* place would be the right one to record this things like this) Best regards, Julian
Received on Sunday, 11 January 2015 09:42:10 UTC