Re: [Last-Call] [Gen-art] Genart last call review of draft-uberti-rtcweb-rfc8829bis-02

[Cullen Jennings <[e-mail]>]

Comments inline ....

On 4/17/2022 12:37 PM, Christer Holmberg wrote:
> 
> My question is why JSEP uses an INFORMATIVE WebRTC reference WITH a version, while other RTCWEB RFCs use NORMATIVE WebRTC references WITHOUT a version...
>

I think most those documents are simply wrong and would challenge people 
to show what normative part of them is not implementable without being 
able to read the WebRTC spec. We have many example of fully working 
RTCWeb applications that don't even use javascript or any of the APIs 
from W3C and that was our design goal to start with to allow non browser 
clients. I'm not going to speculate on how the documents ended up with 
normative references as right now we just have to deal with 8829bis.

My view is that the references to the W3C WebRTC stuff should be 
informative not normative in 8829bis but if someone wants to point out 
specific required text in 8829bis that requires understanding the W3C 
spec to implement, I'm happy to change my mind - clearly not a huge deal 
one way or the other.


> 
> RFC 8829 references https://www.w3.org/TR/2020/PR-webrtc-20201215/. I just want to verify that there is no text etc in 8829bis that is not aligned with 20210126.
> 

I have been watching the diff as they go it to try and make sure they 
are aligned. I also just looked at all the changes from 
PR-webrtc-20201215 to the REC and I don't think any of them would cause 
any changes to the IETF specs or to the changes from RFC 8829 to 8829bis.

It's hard to review the stuff do to some bulk moves of text and my eyes 
glaze over after reading it way too many times but I did do a review and 
did not see anything that looked like it might be an issue.