In this interview, host Aaron and guest Tsahi Levent-Levi discuss two unrelated but timely topics. First, they examine Safari 26.4’s new support for WebTransport, explaining that the API lets browsers use QUIC directly for custom transport scenarios, enabling high‑throughput, multi‑stream connections and potentially powering Media‑over‑QUIC (MOC). Tsahi argues that while the Safari addition is a useful step, WebTransport’s real impact will depend on broader browser and developer adoption and on native MOC support, which he views as a solution looking for a problem. Second, they shift to a legal analysis of Dolby’s lawsuit against Snap for alleged AV1 and HEVC patent infringement. Tsahi outlines Dolby’s strategy of targeting a smaller vendor to extract royalties, the role of the Alliance for Open Media’s patent pool, and the complexities of retroactively extending patents to cover AV1. They discuss how the case could affect codec adoption—potentially boosting VP9 if AV1 faces legal hurdles, or, if Dolby loses, accelerating AV1’s dominance and challenging the future of HEVC/VVC. Throughout, they stress the importance of monitoring the lawsuit while continuing to adopt AV1, and they advise developers to prioritize codecs (VP8, H.264) if short‑term risk is a concern.
Chapters
Thanks for joining me today, Tsahi Levent-Levi. We're going to discuss two different topics today. First, we'll talk about WebTransport, and then we're going to talk about patent lawsuits. But I want to be clear for our viewers up front, those topics are not related to each other, but they are both important. So let's see if we can cover both of them. Let's talk about WebTransport first. So Safari 26.4 supports WebTransport as of March 24th. 2026. So why is that a big deal, Tsahi?
Well, that's a question. I don't know if it's a big deal or not. Um, and let me try to explain. Okay, WebTransport allows you to use QUIC as the transport protocol within what you're doing in the browser. So HTML3 runs over QUIC. That's available in all browsers as far as I know. But now the actual interface is an API to developers is there. So you can build your own things and not use it as an HTTP transport, but rather something else similar to how WebSockets is used. So that means that now I've got a connection from the client to my server that can run over HTTPS— sorry, that can run over WebTransport instead of WebSocket or HTTP. So the big deal is that you can do things there that you couldn't earlier, especially high throughput, multiple channels, things going in different configurations and ways. Essentially everything we want from MOC, Media Over QUIC, that runs over WebTransport. Right now, having that on Safari means that now you can have theoretically MOC run on Safari as well.
Yeah, as well as iOS devices now too.
Right? Yes. Now, here's why I don't think that that's a big deal. Okay, and that's me. First of all, we've seen with WebRTC, the developers care a lot about Safari, but they would ship products if Safari doesn't support it. Even so, for many years, Apple didn't support WebRTC, then it did, but not exactly, then it did, but still not exactly. And even today, there are things that doesn't work quite well on Safari browsers, which makes Chrome the best browser around for things related to WebRTC still, even on Mac devices. I guess the same can apply to WebTransport. It would still work in a lot of places. We're still going to have different, you know, backward compatibility for mock scenarios that will include WebRTC and HTTP. In HLS and MPEG-DASH and all of these other protocols. So the fact that it supports Safari adds some kind of a stamp that WebRTC transport is happening— sorry, that WebTransport is happening. But is it going to happen tomorrow? No. Does that mean that MOC is there? No. Because MOC can be implemented on top of WebTransport. Through a lot of WebAssembly code. I think that we're all waiting to see that browsers start supporting more of MOC internally than letting the developer do everything by hand, because that's not going to be fun to anyone. That's my view.
Well, it makes sense. But certainly you could say if, if, if someone watching this is interested in seeing, you know, in using MOC across all browsers, then this was an important step, right? Because MOC sits on top of WebTransport. So now at least we've got that ground layer implemented that we could use MOC, but we could also get a lot of, you know, some of those same benefits if we're willing to, you know, hand roll it ourselves using WebTransport essentially, and being able to do some things around multiple streams in the same connection, for example.
Yes, I'm probably not that enthusiastic because not that enthusiastic from MOC to begin with, because again, I still believe that it's a problem— like, it's a solution waiting for a problem. And that's the issue that this industry still has, and it will have even with Safari supporting it.
Yeah, no, that's fair. I mean, I think for most use cases, we're still always going to be using WebRTC, um, as kind of a default. And so definitely, but there are those edge cases where I think WebTransport, unbundled WebRTC, Mock are going to be interesting and definitely opens that up. So we'll see more of those edge cases, but WebRTC, Mock, WebTransport, they're all going to be living side by side together for a long time. So maybe let's switch topics now to things that apparently are not living side by side as nicely as we thought they would, which is our legal topic. Topic for today. So possible patent infringement. In a recent issue of your WebRTC Insights newsletter, you noted how Dolby is suing Snap over AV1 and HEVC patent infringement. So what's going on here, Tsahi? Because I thought that AV1 is royalty-free.
Yeah, so first of all, I'm not a lawyer. Let's start with that. But on face value from the story, the way that I understand it, Dolby decided that they have a few patents, one of them related to AV1 and two others probably to HEVC, and they are shopping around to get, let's say, royalties for patent payments from vendors. Now, when you start doing that for codec or for anything else for that matter, you say, well, I've got the patent, someone is infringing it, or many vendors are infringing it. How am I going to go and win that patent? And the way to do that is to go towards the smallest, weakest vendor in the link and try to win over him. Okay, so you take him to court. The first thing that you are assuming is not going to fight back. He's just going to come negotiate and pay. If you get that, that's a win because now we can find the next weakest link that is slightly bigger and then you go and chew them one by one. I had that done to me when I worked at the company once. It's not fun. We were about second or third or something like that, or we even help someone do that to others so they don't touch us. So there's a lot of politics going on in there between businesses. Now, I think the issue is here is slightly different because AV1, it's not that it doesn't have any patents in it, it's that a lot of the companies and vendors that are investing in AV1, and that's most of the industry today, created a patent pool in the Alliance for Open Media. And what are you essentially saying there, as far as I understand, is we all got— have patents for AV1, we're giving it to anyone else within this group, and we're giving to everyone else outside of the group to use these patents specifically for AV1 implementations. And by the way, if someone is going to come and sue one of us in the group, we're going to use our pool of patents to go sue them back or something like that, or to protect them. I, I don't— that nuance I'm not sure about.
Yeah.
Um, and that's where things become interesting because Dolby decided to go after a vendor, right, in the US and in Brazil, so specific countries. And the way they did that was say, we've got one patent that relates to AV1 and we're going to sue for that. And they went after Snapchat, which are a vendor within the Alliance for Open Media. Now, I don't remember that they were a vendor in the Alliance, so it might be that they joined because they knew that they are going to be sued, or that they joined before that.
Yeah, not aware of that.
Okay, so it's one of the two. Now, here are the things that, that are mystifying. First of all, Dolby has been using WebRTC for a long time, which means that they're on the path of the things that happen in WebRTC, and AV1 is one of these things that's happening in WebRTC. In a way, they're now pissing all of the other vendors in the ecosystem that are out there that deal with AV1.
Yeah.
And WebRTC. Why? I don't know. Really don't know. Um, I think it also means that they don't intend on using AV1 in the near future.
That Dolby doesn't intend on using that, you're saying?
Yes, I think so. Because let's assume that they start using AV1. Yeah, what would Apple, Google, Facebook, Amazon— who else is there?
Samsung, Netflix, Mozilla is in there.
Yeah, with their AV patents, would they go after Dolby at that point in time? I'm like, um, not a smart move from that point of view. Yeah, then there is a part of when you try start reading around that, the first thing is, well, Everyone is saying, well, this is interesting. Someone is going to check if anyone has issues with patents. So everyone is going to look at that specific litigation that is taking place. We had that once years ago with Nokia going after HTC, I think, with a, with AV, with the VP8. Nobody still cares because VP8 is still considered royalty-free. Apparently didn't work out. Will this work out? We don't know. I can only assume, but I don't want to assume. I hope that it won't work out. Yeah, because that would cause another set of headaches. Now, within the industry, when you look at it, there are a lot of vendors, companies, ecosystems that are going to switch towards AV1 in this year in one way or another, and that might hurt them. And if this case is won by Dolby, that's going to be an issue in the industry. If they lose, then AV1 is going to explode. It will probably kill future adoption of HEVC, and it will also put in question the continued effort on VCC or VVC, which is H.266, the next codec after HEVC. Why? Because who wants to pay for patents when you can use AV1 today and AV2 tomorrow, which You know, they're already talking about in the lines for open media.
Yeah.
Now, looking at the actual patent that they are using, it's an old patent that got continuations more than once. It's a technique within the standardization bodies that within the patent office where you can start a patent and then after a couple of years just update it a bit. And what they did in the updates were increase what it is supposed to do so that it would fit into AV1 as well. Again, I'm not a lawyer. That's my understanding of what I read, which means that it might be hard for them to come and, come and say that this is a patent and they own it because they made the changes after the fact that AV1 started doing things that they didn't like. So how is that going to fare for them? I don't know. I hope that not that well. And it's going to be interesting times in the next few months.
Yeah, for sure. And definitely, I think, you know, we don't want to go back to like the bad old days of Flash and things like that that WebRTC was supposed to get away from, where we had very vendor-specific and license-specific implementations, to be sure, right? And most WebRTC traffic, to be clear, is still over VP8 and H.264, so not directly affected by this. I suppose that at least in the interim, while people are worried about AV1, maybe this makes VP9 become a little more widely adopted. What do you think? I see everybody holds their breath and see how the court case goes.
My money is that people are going to be two things: they are going to follow what happens in this case, but they are going to ignore it all the same in parallel, and they're still going to continue with AV1 and being hopeful and optimistic. Yeah, um, that's my guess. There's just too much going on with AV1 and around AV1 to lose at this point for such a reason.
Yeah, yeah, yeah, certainly a lot of momentum behind it. But I guess if you're watching this and you're concerned in your code, you can set kind of preference order in the code and you know, rely on VP8, H.264, exclude AV1 if you're concerned in the short term. But, you know, we'll see how it goes through the courts, I suppose. So, and what all those other big names in the Alliance for Open Media, how they respond to this, it will definitely be interesting to see. I personally would not want to start a lawsuit that goes up against the likes of Google and Apple and Microsoft, but—
You actually do, but you want it to be small. Like, $100,000 is nice, that they might decide not to fight back. This one, that's a bit too big to, you know, to ignore.
Right, right. Yeah, good point. Yeah. All right, all right. Well, until I come up with that $100,000 lawsuit that I can do against Google, um, and advance my retirement a little A little bit. We'll have to continue our conversation next month. In the meantime, as always, thank you for your insights today, Tsahi. For those of you watching, to learn more about Tsahi's work, check him out at bloggeek.me. To learn more about future episodes of WebRTC Live, check us out at webrtc.ventures. Thank you, Tsahi.
