This post is also available in: Français (French)

My twitter is on fire! This is one of the most awaited for news in the ecosystem and for many but not all, it comes way earlier than expected.

I. history and context

Earlier this year, speaking at NTT’s webrtc conference in Tokyo, I had listed ORTC only as a possibility for 2015 (slide #17 and #30). Later on, at the San Francisco webrtc meet up hosted by tokbox in june, I had repeated this prediction, which was fitting what most were thinking back then (around 1:45). The release of the GetUserMedia API earlier in May should have been a hint. I was wrong.

As august and september passed by, more and more hints in the ecosystem pointed to a fast(er) release of ORTC in Edge. A lord of questions were still open: which codec will be used, how close to webrtc NV API will the ORTC API be, etc. Actually, some of those are still unanswered today. By september, it was clear everything was done and ready, which made some devs happy, and some standard committee members less happy. It shall start to make sense why I chose to spend one week in Redmond before the W3C meeting. 🙂

II the facts

On september 18, 2015, Microsoft announced the availability of ORTC for Microsoft Insider Program members. They announced it in several places, but mainly on the windows/Edge and office/Skype Blogs. On the official page of microsoft, a demo with twilio wass accessible. The same day, blogs posts on webrtchacks and other very visible venues started to appear. Very clearly, some were in the known, as the body language of many at the Kranky geek show when the subject came on the table had hinted at.

Both &yet’s simpleRTC (with its audio-only, chrome-only-for-now, demo) and twilio’s ortc-adapter were providing when this article was written ways to interoperate between webrtc and ORTC already.

As announced by Bernard Aboba at the Kranky & Geek show, and as I reported on the corresponding webrtc-discuss mailing list thread, MS plans to support it’s own flavor of 264 SVC first, called 264UC. For the hardcore media developers that already know about 264 AVC/SVC and want to see the  difference with 264UC, here you go. It then plans to support plain H264 (AVC) for interoperability with firefox (today) and chrome later. As far as support for 264 for webrtc in chrome, android and iOS versions of the libwebrtc library already support them through the OSes APIs, only desktop versions of chrome do not support it, and have a bug open and already assigned. In brief, they are working on it as we speak, and patches are already available.

In the mean time, MS plans to use a media server to transcode between the different flavors of H264 out there for Skype, as illustrated above.

III Consequences and discussion

A. short term

First and foremost, go read the webrtchacks article to get a quick overview of what you need to do to have a simple 1:1 call. Get an aspirine, come back here. 🙂

The ORTC API is more complex, and being lower level will need you to define many more things than the webrtc api would for the same use case. The advantages or those API only become obvious for more complicated cases like codec manipulation, connections with media servers, multi-party with one peer connection, simulcast, SVC codecs …. For most, it’s just going to be superfluous nonsense. In which case, go with a shim like twill’s ortc-adapter and you will be taken care of. It’s very likely that most Vendors will update their JS SDK in the next weeks to do just that as a first step. Note: that will only work for audio, so prepare some aspirines for the web developers that will have to handle all the browsers and all the media cases 🙂

If you are using an IE plugin, you need to make sure your market need it. Entreprise market might need it as they are captive of older versions of IE, but social website, and common websites should not care. Moreover, the two main plugins out there are still stuck with older versions of the specs (before promises and other things were added, roughly half a year ago), and show no sign of preparing for the 30 new Objects and API ORTC and webrtc NV are bringing.

B. medium/long term

There are two main questions here you need to ask yourself for the long term. 

1. Am I limited in any way by the webrtc API?

If you have already everything you want, just use a shim, you’re good. If you are limited in terms of network control (security or bandwidth), codec, number of streams, etc, then ORTC or webrtc NV will bring something for you. If you’re using a media server, chances are also high that you will benefit from the new APIs.

2. How convergent webrtc and ORTC really are?

That’s the key question. If I wait, will this interoperability problem get easier or disappear? Most JS developer are ok with some degree of incompatibility anyway, but incompatibility on the wire is not something you will be able to deal with at JS level, so one has a real problem here.

First demos and example seems to indicate that interop, as far as signaling, API, and audio media is concerned, is already a reality. Until MS implements plain H264, interoperability for the video will be a question mark.

Finally, I am not aware of anybody having looked at the webrtc / webrtc NV / ORTC APIs to check how far appart they really are. My feeling is that they are close enough for some JS magic to make them disappear, but again, I have not checked yet.

IV Conclusion

Since august, it was obvious for many that this was coming. On my most recent (public) presentation, my last slide was clear (here, slide #12): ORTC is a great API, and it’s needed for a lot of use case, but many vendors are not ready for the tsunami and the amount of work needed to support all that.

My prediction was and still is that the webrtc PaaS ecosystem will shrink by 25% (in number of companies) as the smaller one are already very very lean, in a market were traction is still the exception, and funding can be hard to come by. The IE plugin vendors will be even more impacted, as the market just got smaller, and they will need to have different sources of traction to hope raising money. We’re living interesting times.

This work by Dr. Alexandre Gouaillard is licensed under a Creative Commons Attribution 4.0 International License.

This blog is not about any commercial product or company, even if some might be mentioned or be the object of a post in the context of their usage of the technology. Most of the opinions expressed here are those of the author, and not of any corporate or organizational affiliation.

This work by Dr. Alexandre Gouaillard is licensed under a Creative Commons Attribution 4.0 International License.

This blog is not about any commercial product or company, even if some might be mentioned or be the object of a post in the context of their usage of the technology. Most of the opinions expressed here are those of the author, and not of any corporate or organizational affiliation.