- From: Mark Nottingham <mnot@mnot.net>
- Date: Thu, 30 Aug 2018 10:22:59 +1000
- To: Patrick McManus <mcmanus@ducksong.com>
- Cc: Mike Bishop <mbishop@evequefou.be>, Alexey Melnikov <alexey.melnikov@isode.com>, "Julian F. Reschke" <julian.reschke@gmx.de>, HTTP Working Group <ietf-http-wg@w3.org>
On 30 Aug 2018, at 3:46 am, Patrick McManus <mcmanus@ducksong.com> wrote: > >> Personally, I really wonder if existing implementations have seen enough deployment to preclude just changing the value. > > the 6455 implementations all (to some close approximation of all) use lower case websocket.. the new protocol is just matching reality. let's not double down on the registry problem in the new protocol. Ah, I forgot about that. It sounds like we need to: - register 'websocket' pointing to 6455 - file an errata on 6455 recommending using that for interop, but noting that some might send other casings of the token - perhaps modify the registry entry for "WebSocket" noting that it's not the preferred protocol artefact / historic / whatever Alexey? -- Mark Nottingham https://www.mnot.net/
Received on Thursday, 30 August 2018 00:23:26 UTC