Interest in WebTransport + WebCodecs?

[pthatcherg]

I gave a presentation at the W3C games conference (slides) about WebTransport + WebCodecs: . It was mostly targeting at gaming and cloud gaming, but it should be very well suited for low-latency live streaming.

I gave a presentation at the IG previously about this topic, and WebTransport + WebCodecs is hopefully a better way that is progressing in both standardization and implementation. We're looking for input from potential users of these APIs as we shape them.

If you are interested in using these APIs or have opinions about how they should be shaped, live a comment :).

[chrisn]

Thanks, @pthatcherg. Here are the minutes and slides from that IG call, also @johnluther's introduction slides.

[chrisn]

As an update, both Web Transport and Web Codecs are now being actively developed in the Web Platform Incubator CG (WICG).

There were breakout sessions at TPAC 2019 on both Web Transport, Web Codecs, and related discussion in a breakout covering Efficient audio/video processing.

Great to see these moving forward! I'd like to invite IG members to give input via the linked WICG GitHub repositories.

[tidoust]

Please note that the WebTransport API is now being proposed as a standardization item in a new dedicated Working Group charter, see:
https://lists.w3.org/Archives/Public/public-new-work/2020Jun/0004.html
(copied below)

I note that whether the WebTransport API ends up being a good solution for common low latency media streaming scenarios remains an open question at this stage. It will depend on whether the API (and underlying protocol) features the right semantics to address scalability/cacheability issues. In turn, that will require requirements to be brought early on in API design discussions.

Hello,

Today W3C Advisory Committee Representatives received a Proposal
to review a draft charter for the WebTransport Working Group:
https://www.w3.org/2020/06/proposed-webtransport-charter.html
 
As part of ensuring that the community is aware of proposed work
at W3C, this draft charter is public during the Advisory
Committee review period.

W3C invites public comments through 2020-07-27 on the
proposed charter. Please send comments to
[email protected], which has a public archive:
http://lists.w3.org/Archives/Public/public-new-work/
 
Other than comments sent in formal responses by W3C Advisory
Committee Representatives, W3C cannot guarantee a response to
comments. If you work for a W3C Member [1], please coordinate
your comments with your Advisory Committee Representative. For
example, you may wish to make public comments via this list and
have your Advisory Committee Representative refer to it from his
or her formal review comments.
 
If you should have any questions or need further information, please
contact Wendy Seltzer, W3C Strategy Lead [email protected].
 
Thank you,
 
Xueyuan Jia, W3C Marketing & Communications
 
[1] http://www.w3.org/Consortium/Member/List

[tidoust]

I note that WebCodecs is now being tracked in a separate issue #40.

The WebTransport Working Group was created earlier this year as proposed. The group published a First Public Working Draft of the specification early May:
https://www.w3.org/TR/webtransport/

The specification has made a lot of progress since last discussions on the topic in the Media & Entertainment IG, notably in terms of effective transport semantics. As part of TPAC 2021 plans (#71), it could be worth considering a joint call with the WebTransport Working Group to review progress and assess whether WebTransport can be a useful brick from a media streaming perspective.

On top of my head, some considerations/questions that could perhaps be useful to discuss:

• Scalability: Main envisioned use case for WebTransport in media contexts is for low-latency streaming. Can WebTransport leverage CDN/caching infrastructures in low-latency scenarios? Also, to benefit from caching, would the resulting latency be similar to that in low-latency DASH and HLS? In other words, what added value could WebTransport bring in that space? (for instance, would the ability to create "unreliable" transport be a major boost?)
• Integration with MSE: One envisioned path for realtime scenarios is WebTransport + WebCodecs. But media companies are used to rendering media content with MSE+EME. Would WebTransport + MSE work as well? How simple is it to plug WebTransport and MSE?
• Benefits over HTTP/3: WebTransport builds on top of HTTP/3. When HTTP/3 becomes widely available, what would be the added value of using the WebTransport API rather than fetching content over HTTP as currently done in adaptive streaming solutions?

What about it? Are there other points to discuss?