- From: Carsten Bormann <cabo@tzi.org>
- Date: Sun, 13 Sep 2015 16:12:05 +0200
- To: "Nilsson, Claes1" <Claes1.Nilsson@sonymobile.com>
- CC: "public-wot-ig@w3.org" <public-wot-ig@w3.org>
Generally, yesterday's IPv4 NATs require the use of TCP for reasonably efficient NAT traversal. The need for supporting that legacy is one of the two reasons why we are standardizing CoAP over TCP (the other one is for backend usage). Websockets is not a protocol I would expect to see used on a constrained device. (There is indeed a draft specification for CoAP over Websockets, but that is really for use in Web browsers, which as of today don't speak CoAP over UDP or DTLS directly.) Grüße, Carsten
Received on Sunday, 13 September 2015 14:12:42 UTC