Review

[bernard_aboba@…]

Overall, this document does not cover some important aspects of
of web evolution, such as the addition of additional transports
(UDP, DTLS, SCTP, SRTP), the evolution in HTTP (Websockets,
HTTP 2.0), and the emergence of HTML5 as a general application
development platform. Overall, these trends argue that the web
is becoming more of a general purpose platform, subsuming existing
applications, rather than forcing applications to adapt to a
more limited HTTP 1.1/TCP/Javascript paradigm. IMHO, these
wider trends need to be incorporated into the document.

[bernard_aboba@…]

Document review

[bernard_aboba@…]

The basic premise of the document is that we have entered an age of "post standardization". However, at least in the medium term, the opposite appears to be the case: that we have entered an era of rapid change in web protocols, where standardization is more important than ever.

While signaling is out of scope for RTCWEB, the group has spawned a large amount of new protocol work on issues such as security, RTP multiplexing and congestion control. So instead of introducing an era of "post standardization", RTCWEB has increased the need for standards work (at least in the media plane).

Similarly, with groups such as Hybi and HTTP 2.0, we are seeing an increased need for standards work.

Maybe once this flood of standards work is completed, and the browser incorporates the huge increase in functionality that all of this work implies, we will indeed begin the transition to a world in which mobile code has obsoleted the need for development of new protocols.

But in the web world we live in now, standardization has never been more important.