Re: [webrtc-extensions] Consider an API for non-delivered messages in SCTP datachannels (#50)

> I'd say the question is whether the importance of the use cases outweighs the potential pitfalls (and I think there will be more once we examine this as part of the WebRTC state machine) and complexity. 🙂

I second what @alvestrand and @lgrahl said: keep the API as simple as possible for the required use case. I have seen a lot of cases in which the SCTP socket API, which is specified in an RFC, was used incorrectly and therefore the application behave sometimes in a strange way. Some developers start to work around the strange behaviours since they do not realise that the application has a bug the y need to fix.

@boivie : Can you precisely describe the need the application has and which you currently can not implement using the data channel API? I think this is the API currently exposed, not the SCTP API. One just needs to figure out how to provide the data channel API by using the SCTP API. At least in my view...

-- 
GitHub Notification of comment by tuexen
Please view or discuss this issue at https://github.com/w3c/webrtc-extensions/issues/50#issuecomment-865146114 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Monday, 21 June 2021 15:50:43 UTC