Video: Apps Are the New Middleware, and OTT is the New Broadcast
Tim: Welcome back to Streaming Media East 2017 and I have a very special guest with me today. Avni, introduce yourself. Tell the audience what you do.
Avni: Thank you. I'm Avni Rambhia. I'm an Industry Principal with the Digital Transformation group at Frost & Sullivan. I cover encoding and transcoding technologies as well as DRM and conditional access technologies. Many exciting things happening in our world.
Tim: Let's talk DRM first. We'll take it in reverse order. Christopher Levy was on just a moment ago. Matt Smith was on from Brightcove. We talked about the differences between encryption and DRM, and Christopher did a good job of explaining how encryption fits within. What's the state of DRM these days? It seems like paywalls are coming down which may mean that there's more of a need to protect content that's not behind paywalls. Tell me what you're seeing in the industry.
Avni: In a nutshell, what we're seeing is that apps are the new middleware and OTT is the new broadcast. As you start to see set top box become a smaller and smaller percentage of where content is consumed, you need to take protected content and put it to more and more devices, more and more browsers, more and more use cases, many different networks. DRM is that flexible, all-purpose glue that secures content anywhere, anyplace. We're seeing a lot of the traditional cast providers having to diversity and very proactively shifting towards DRM and, more specifically, multi-DRM.
Tim: In the enterprise space, which I spend a decent amount of time in, sandboxing, especially with bring your own device, sandboxing became a really big deal. In this assumption, if I understand what you're saying, we're looking at a completely unprotected device so the content itself has to be highly protected.
Avni: Yes. It's what we call the hostile attacker model. In the vast majority of cases, your user is a good user. It's a valued customer. You want their experience to be good, seamless, transparent. You don't want to fall into the better than original traps that we had a decade ago where you protected the content so heavily that users couldn't reach it at all. But, at the same time, you do have to keep tabs on where it is, how it is, and more sophisticated apps are helping with that. To me, IP connectivity is great because you can do a lot of dynamic checking and you can do a lot of dynamic reactions that you couldn't do in a one way broadcast world.
You have a mix of all these different approaches. That's one of the big parts of the strategy that we're preaching for content protection. It's not just locking it down. It's securing it. Whatever it is, whenever it is, and whatever secure means in that particular context. For enterprise, for example, it'll typically mean LDAP-based integration for the authentication, some kind of measurement of who's watching it, when they watched it, where they're watching it. You do have some users of DRM now, with distance learning for example, where it starts to look much more like a paid TV service than it does an enterprise video conference. You start to see a merger of disciplines there.
Tim: Essentially, that would be subscription-based or paywall. You mentioned the whole locking it down from a decade ago. One of the things that frustrated me in the industry, it still frustrates me, is if I have an Amazon Prime membership here in the states and I happen to travel abroad, I really would like to be able to watch my Amazon Prime content. The original argument had been licensing wouldn't allow them to let you see the content outside the US but even the Amazon original content or the Netflix original content, which they completely own the licensing to, is not available to watch overseas. Do you think we'll get to the point where the devices will be trusted enough with some form of two-factor authentication to allow you to really actually take TV anywhere or see the content anywhere you are? As opposed to TV anywhere meaning within the 48 contiguous states, etc.?
Avni: It's funny. I think I'm somewhat of a heretic on that front. I think the secure devices are already there. Samsung TVs are sold as much in India as in Vietnam as in Afghanistan as they are in Brazil or the US. Apple iPhones may be a little bit different distribution but Chromecast is wildly global. I think the devices are there. Where the gaps are in my perspective are in the law enforcement, in the tracking, and the tracing, and also not necessarily in the content agreements but in the talent agreements or the resale agreements. Part of it is the business aspect of it. Part of it is the law enforcement partnership. Technology can go so far. Part of it is actually also the delivery side of it. If you look at the data transport rates in the US, they're dramatically lower compared to anywhere else in the world. For a given prize, if you're going to stream a certain volume of content, you have to look at those balances too.
I think it is definitely growing. The growth for anyone is outside the US. We're a very saturated market here. Whether you can take growth and make it profitable, that's where, I think, the disconnect is right now.
Tim: You have ex-pat communities of all different flavors. You have iPlayer from BBC. People who are paying a license fee but they can't necessarily consume the content outside. You mentioned data transport rates. That gives us a beautiful segue to HEVC. HEVC has been an area you've focused on, had a conversation yesterday with a company that wants to create a survey and do a report like have done with some other companies. Their basic statement is it's time for HEVC. What's your take on whether that's reality or whether the licensing issues or technical issues or playback devices still need to mature to be able to get us to a point where HEVC is used?
Avni: I think I would take a step back and say it's time for a successor to AVC. AVC has taken us very, very far. It's an outdated-
Tim: That's a great political answer. It's time for something else.
Avni: But the fundamental with the problem is AVC is very, very good for SD. It's pretty good for HD. If you throw a lot of computational cycles at it, as of today you're still able to get as much out of AVC as you theoretically could out of HEVC.
Tim: Absolutely. As a matter of fact, I worked for about seven months as a strategy officer for a company that does video optimization. They were able to essentially give you HEVC bandwidth levels for 1080p.
Avni: You have all the cool jobs.
Tim: Yeah, I don't know. But 4K seems to present a bit of an interesting challenge.
Avni: It is a bigger problem. What happens with HEVC's, in the ideal situation, and I was one of the first people to get on the soapbox and say we've learned our lesson with licensing. We don't expect it to be an issue with HEVC. It was all set to absolutely skyrocket. Internally at Frost, we keep a directory of who has made announcements and who's shipped product. Between 2014 and 2015, we saw this massive drop where everyone who had announced plans in 2014 suddenly put them on hold in 2015. The reason was the licensing snafus that started to come up. I understand where the tech companies are coming from because you've got millions and millions of dollars in MPEG-2 royalty revenues that are drying up so how do you keep an old school business running and profitable?
However, this is the age of the internet. Try telling Cisco and Google that they are going to beholden to a couple of IP houses where they've got billions of dollars of R&D resources of their own. I said this at Streaming Media West last year: I think there is a window of opportunity where HEVC can still get its act together and start to leverage the massive investments that we have-
Tim: But if it doesn't?
Avni: If it doesn't, then VP9 is here, AV1 is right around the corner. For the next two years, AVC is not all that bad. 4K is aspirational. VR I think is aspirational. But you don't really have the business model or the bandwidth or the content creation. Those blockades are still there.
Tim: It's interesting that you say that. The 2014-15 where the licensing snafu happened, I wrote an article around that time where we went back and looked at what happened with h.264. You remember there was VP7 I think at that point. There were a couple other competing codecs. Once the AVC group got their act together on licensing, there were some sacrificial lambs on the table, of On2, etc. My thinking was when we came forward, that with the VP9, AV1, that it would force the licensing, that group to go back and get their act together. It hasn't' necessarily seemed to happen.
Avni: I think it hasn't because you have two distinct ecosystems now. You've got the linear broadcast ecosystem where HEVC is still going to be the way forward. You've got governments negotiating down content royalty rates and so forth. But the problem is that's such a small percentage of our overall ecosystem now.
Tim: That's true.
Avni: You've got this ubiquitous AVC that's common across enterprise and media entertainment. I don't think we're going to have that luxury moving forward. That age came and went. When Adobe put AVC into Flash, that essentially brought those two ecosystems together and we had this golden era.
Tim: And Gary Williams taking the ITU standards. Because I was in video conferencing and we had H.264 long before we had it in MPEG. What's fascinating to what you say is when we put the Level 3 survey report out a couple weeks ago, it was very clear that, overall, the respondents said that 4K broadcast was going down in importance for them and 4K streaming was rising for them. You say it's aspirational. To me I wonder if we're going to reach a point where, if HEVC doesn't get its act together, and AV1 continues to improve, that, essentially as you say, we have a replacement for AVC but it may not be HEVC.
Avni: I think that's true. That's the other difference between OTT and traditional linear broadcasts. You don't have billions of dollars in equipment out there and sitting there. You can change formats on a dime.
Tim: You're not required to meet the same standard as an NTSC broadcast.
Avni: No. This is the internet age, right? Just repurpose a whole bunch of servers and rewrite a JavaScript and you're done. The battles have been fought at the processor level but those guys have enough footprint now that they can hedge their bets. I still think the preferred alternative is to use HEVC. Maybe they will get their act together but, if they don't, you've got a very credible plan B. Regardless, I think the power base has shifted away.
Tim: Even in the time from 2015 to now, with these quality optimizations of AVC, we're finding it has more legs than we thought that it did necessarily.
Avni: Because the CP was given a lot more room to grow.
Tim: Exactly. Let's say tomorrow HEVC licensing gets dealt with. Would we start to see significant adoption because devices are capable of playing it back or would it still take time to get the licensing deals into silicon?
Avni: I think it should be fine. There's a lot of capable but not enabled out there. Apple is a classic example of that. HEVC was in FaceTime, now it's not, but you know it's capable. It's just not enabled. I think a couple of sticking points in that the DRM stuff may have to be done. Libraries may have to be re encrypted. Some of the analytics, some of the data gathering, some of the caching, but those are little problems and people know how to deal with them.
Tim: As we think, moving beyond, when do you expect 4K will move from aspirational to standard? Especially if you're saying consumption of set-top boxes compared to the overall is dropping. Do you think we will see standard with 4K or do you think it's going to be that ultimately everybody's still comfortable with 1080p on smaller devices?
Avni: I think it depends. HD, if you look at it, it had an initial surge but then you still have SD channels 10 years later. But it will creep up there. One of the very immediate applications of 4K workflows is actually VR because, for an HD VR view, you have a 4K workflow. We're seeing 4K workflows fly off the shelves not because of the linear 4K experience but because of the immersive VR experiences.
Tim: Are you seeing traction in VR video?
Avni: We are.
Tim: Okay, interesting. Because I've talked to another of companies who have said, "Got a lot of tire kickers but we're not necessarily selling a whole lot of ... " This is more the service delivery side than the production side.
Avni: It's an interesting statistic. If you go to Google and look at 360 videos, there's about half a million of them now up from a quarter million maybe six months ago. Only 8,000 of them are actually longer than 10 minutes. In the grand scheme where you're streaming a billion SD/HD minutes of video a day, this is still small. But if you look at the trajectory, it's promising. If you look at the user engagement levels, it's promising. I think that the key differences, can you cross the bridge between mainstream and mass market? That's where-
Tim: That's a pretty big chasm to go across.
Avni: It is.
Tim: One final question, 1080p, HDR, HFR, and, for the audience, HFR is high frame rate, HDR is high dynamic range, more like film than video. Where are you seeing traction there? Because I'm hearing from delivery companies that they have paying customers who want HDR/HFR, where they don't necessarily want 4K or VR.
Avni: That's exactly spot on. HD with HDR is getting much more favorable reviews than just straight 4K and it's much, much cheaper to deploy, much more immediately responsive from the customer.
Tim: Yeah, the customer can sense a significant difference. Interesting. Do you think that that will continue to give legs to AVC for a period of time or do you think HEVC, with HDR and HFR might also be a potential answer?
Avni: I think at least from what I'm hearing from the encoder vendors as well as the middleware and the app developers, HDR plays nicer with HEVC. It's more clean in terms of the workflows. You have fewer workarounds. I suspect it's going to be in that context.
Tim: High frame rates of course, obviously, the higher the frame rate, the more data you've got, therefore the better for HEVC or AV1.
Avni: But sometimes that actually favors AVC because the implementations are much more efficient than the HEVC decoders. With high frame rate, it goes back and forth. But for HDR, HEVC is better.
Tim: Well, we've actually done almost eight minutes now. We could probably do another 30 or so but this is Avni from Frost and Sullivan. I'm Tim Siglin, contributing editor with Streaming Media Magazine. Thank you all for watching this interview.
Avni: It's been a pleasure. Thank you for having me.
Tim: Absolutely. Yeah that's ...
Avni: That went by fast!
Related Articles
EZDRM's David Eisenbacher and Streaming Media's Jose Castillo discuss DRM solutions and key players, and emerging codecs in this interview from Streaming Media East 2017.
29 May 2017
BuyDRM founder Christopher Levy and Streaming Media's Tim Siglin discuss the history of BuyDRM, the evolution of DRM technology, and misconceptions about encryption and DRM.
27 May 2017
BlueFrame Technology's Chris Knowlton and Streaming Media's Tim Siglin discuss BlueFrame's subscription-based cloud service and ProductionTruck app and how the company is approaching the worship and college sports markets.
26 May 2017
Frost & Sullivan's Avni Rambhia discusses the pros and cons of browser- and app-based content delivery strategies in this presentation at Streaming Media East 2017.
22 May 2017