Voices of Video
Explore the inner workings of video technology with Voices of Video: Inside the Tech. This podcast gathers industry experts and innovators to examine every facet of video technology, from decoding and encoding processes to the latest advancements in hardware versus software processing and codecs. Alongside these technical insights, we dive into practical techniques, emerging trends, and industry-shaping facts that define the future of video.
Ideal for engineers, developers, and tech enthusiasts, each episode offers hands-on advice and the in-depth knowledge you need to excel in today’s fast-evolving video landscape. Join us to master the tools, technologies, and trends driving the future of digital video.
Voices of Video
Greening the Streaming Industry
Power consumption might be the most underrated challenge facing the streaming video industry today. As Dom Robinson, founder of Greening of Streaming, reveals in this eye-opening conversation, the energy footprint of streaming video could represent 2-3% of global electricity consumption – a figure that demands attention from everyone in the ecosystem.
What began as sideline discussions among network architects has evolved into a 30-member organization tackling streaming's energy challenges systematically. Robinson walks us through the complex relationship between video quality, network architecture, and power consumption, challenging many common assumptions along the way.
Perhaps most surprising is the weak correlation between data volume and energy usage in transmission networks. "It's not as simple as reducing bitrate," Robinson explains, describing how energy efficiency requires examining the entire delivery chain. A change that saves power in one area might increase consumption elsewhere – like when a codec change forces millions of consumer devices to work harder on decoding.
The conversation explores fascinating territory, from the "gold button" concept (defaulting to energy-efficient streaming with the option to upgrade) to comparing distribution models like unicast and multicast. Robinson questions whether we need UHD for all content types and shares how Greening of Streaming is conducting rigorous research to provide evidence-based guidance rather than assumptions.
Ready to think differently about streaming's energy footprint? This conversation reveals both the challenges and opportunities in building a more sustainable streaming ecosystem – one that delivers amazing experiences without unnecessary environmental cost. Join the growing movement of companies tackling this crucial industry challenge.
Stay tuned for more in-depth insights on video technology, trends, and practical applications. Subscribe to Voices of Video: Inside the Tech for exclusive, hands-on knowledge from the experts. For more resources, visit Voices of Video.
Voices of Video.
Dom Robinson, Greening of Streaming:Voices of Video.
Jan Ozer, Streaming Learning Center:Voices of Video.
Dom Robinson, Greening of Streaming:Voices of Video.
Jan Ozer, Streaming Learning Center:Welcome to NetEnt's Voices in Video, where we explore critical streaming-related topics with the experts who create and implement new streaming media technologies. One issue that should be paramount to all participants in the streaming media ecosystem producers, engineers, product and service providers, consumers is power consumption. An organization called Greening of Streaming was recently created to address growing concerns about the energy impact of the streaming sector. Today, we're speaking with Dom Robinson, who is the founder of Greening of Streaming and chief business development officer of Ideas. Hi, dom, thanks for joining us. Hey, john, how are you doing? I'm good. So some housekeeping notes If you're watching and have a question, please post them as a comment on whatever platform you're watching, and we'll answer those live, as time permits. If we don't get to them live, we'll get to them via email after the show. Okay, don, what's going on with you? How are you? I'm good.
Dom Robinson, Greening of Streaming:I'm finally in my office, sat down for the first time in about seven weeks, just doing normal day's work at my desk, which is good.
Jan Ozer, Streaming Learning Center:Listen. Why don't you if you've been in the industry, you know probably as long as I have tell us about Greening of Streaming and how it was formed when, who was involved? How did it all come about?
Dom Robinson, Greening of Streaming:Sure. So you may remember, for a while I was running the Content Delivery Summit for Streaming Media Magazine and conversations in that group kind of started amongst the sort of network architects and CDN architects, amongst the sort of network architects and cdn architects, and streets of encoding, uh, cloud architects started up about their sustainability, uh, and so on, and that over a couple of years that sort of grew into its own event and as I was standing up that event I sort of thought this is really big, let's maybe set up a trade association event. I sort of thought this is really big, let's maybe set up a trade association. And what was once a sideline conversation, the conference, has now become a fairly fully fledged members association. We've got about 25, 30 members growing monthly at the moment and it's keeping me very busy, that is for sure.
Dom Robinson, Greening of Streaming:The main objectives of the organization is to foster collaboration across the industry, looking at energy efficiency principally, and the result of how we deploy streaming architectures to address that energy efficiency. What is the less accord? I've heard that mentioned both by you and by some other people if you were to optimize around that good enough quality, could you actually get to economies of scale where you were being energy efficient as well. We originally called it a gold button model because we wanted to offer the idea to the consumer instead of opting into green, was given green or given an energy efficient service, but could opt into a higher quality by pressing a gold button or some sort of interaction. Notionally, apparently, gold buttons don't translate to america very well. Apparently you get gold buttons for winning spelling bees.
Dom Robinson, Greening of Streaming:So we decided to rebrand it for the low energy sustainable streaming accord, uh, which which has the catchy title less and was uh, uh and sort of caught the imagination. So it's quite, it's quite. It's quite a mouthful to really explain what it's all about. But there are lots of silos of work across the industry looking at energy efficiency and sort of atomic scenarios of specific workflow processes. We're very concerned that we look at it systemically and that we're not just passing energy problems up and down the whole supply chain, and so the lesser code is an attempt to align everyone's thinking around specific areas of effort so that we can get all the best engineers to focus on on the right problems and collectively uh, and create some direction of travel for everyone together, rather than just lots of spidering um and slightly random walks into the new innovation space of making streaming energy efficient.
Jan Ozer, Streaming Learning Center:Can you give us an example of what you might reduce to reduce energy? I mean, are you talking resolution, Are you talking bitrate? I mean what do you reduce to save energy and reduce quality?
Dom Robinson, Greening of Streaming:That is the billion-dollar question. At the moment there are lots of assumptions that it's as simple as reducing bitrate. There's also a lot of conflation about what you mean when you reduce bitrate. Reducing bitrate for somebody doing encoding work means something different to reducing bitrate when you're transmitting data across a network, not in terms of bitrate is bitrate, but in terms of the effect it has on your power environment. So it's not as simple as reducing bitrate.
Dom Robinson, Greening of Streaming:One of the big experiments that we've been carrying out in what we call Working Group 4 has been looking at whether we can actually correlate energy bumps in operators' platforms with traffic bumps, and so far we've found very little correlation at all, which makes it complicated. When everybody's immediately sort of talking about this correlation between gigabytes and carbon or gigabytes and kilowatts, it certainly isn't a linear relationship, if there is a direct relationship at all in the transmission networks. And obviously your world, more your world, the encoding world, obviously if you're not encoding, you're not using power. If you're encoding, you're using power to some extent, and so you're using power to some extent, and so there's definite volatility in encoding and decoding and those ends we're paying particular attention to. But also we have hardware optimizations and strategies around how we use hardware resources on a sort of more macro scale. That can make big differences as well. So we're exploring all those things at the moment.
Jan Ozer, Streaming Learning Center:One of the presentations I think you saw at Mile High with me was from a different organization that was looking at power on TV, power consumption. So how do you? You know, I know that there's some standards organizations looking at power consumption how are you going to work with them? How's that all going to shake out?
Dom Robinson, Greening of Streaming:Yeah. So Greeting and Streaming is expressly not a standards organization. There's lots of them. They're very good at what they do. We would be, in their world, considered a user group. So we do have a working group within Greening and Streaming specifically focusing on driving energy as a KPI into protocol and standards development. So, by means of example, the pioneering project we're trying there is with the SBTA.
Dom Robinson, Greening of Streaming:They have their open caching system and within open caching there's essentially a routing protocol called Capacity Insights, and what we're doing is we're exploring where we can find energy data sources, energy mix data sources one of the places where we do talk about carbon that we can feed into the decision-making tree. So, anecdotally, if you're in Iceland trying to decide which cash you're going to pull a piece of content from whether it's London or New York everything's equal in price of performance, everything's equal in price of performance, but the New York cash is solar powered and the London one's fossil powered then it might be interesting to weight the load and the demand onto that New York cash. Of course, at this stage that's moot as to whether it would save any energy, but we might be able to determine its use by energy mix, the input into the system that's hosting those caches or whatever it may be. There's lots of really interesting explorations like that going on.
Jan Ozer, Streaming Learning Center:So if you meet somebody on a plane or in an elevator or offhand and they say what are the top X things I can do right away to reduce my power consumption? And say they're a streaming video producer that goes all the way through, they've got their own encoding farm, they've got their own encoding farm, they've got their own player, they broadcast around the world with CDNs. What are the top X things they can do to really make an impact as soon as possible?
Dom Robinson, Greening of Streaming:I'd say explore rather than do as Greening and Streaming. We're a long way from having best practice that we can recommend. There's so many divergent and complex strategies going on at the moment it's very difficult to actually say this is the best thing to do, because there's just a lot of depends, uh around any answer like that. But in simple terms, you know, the obvious thing is turn stuff off that you don't need. That underpins a whole discussion about redundancy strategy. So if you're a live engineer, I'll give you an example. There isn't a lot of conversation between a cloud and what you might run on the cloud. Yes, if you use a SaaS service from the cloud, like their own sort of home-hosted encoding service, you might find things are different. But if you just simply use a vanilla cloud computer to run your own encoding process, the cloud host has no idea what you're doing. So if you then need to set up a second one for redundancy, well that's sort of a very traditional thing to do in a live streaming environment. What we found in talking to cloud hosts is they have no idea what the application is. So they see an engineer spin up a computer. They don't know what it is, but they want to offer four, nines on that availability. So they spin up a hot spare. Then the engineer spins up a second transcoder and the cloud goes oh, I don't know what they're doing, but we'll give them a hot spare. And the cloud goes oh, I don't know what they're doing, but we'll give them a hot spare. So you've actually got four computers hot when you're only trying to run one workload which might have an outage of a few seconds. So we have to think and that's just a very simplistic example of where we might be over-provisioning redundancy Think about SLA. Sla is the networks themselves aren't that congested, they're just reserved so that people can hit SLAs. And so think about SLA. Think about whether you really need nine nines availability for live streaming. You know, your kid's birthday party or something. There are certain degrees of over-provisioning that have become natural because they're very commoditized and affordable. But they might be unnecessary and they might be. You might be able to halve your energy footprint simply by not using half that, half of your infrastructure. Yeah, and that goes down into into Codex as well. And there's going to be, and there is an endless discussion about quality, whether we need 2K, 4k, 8k, 16k and so on, and I think it's an interesting debate.
Dom Robinson, Greening of Streaming:There's not a great deal of clarity actually about what power those decisions make in terms of the distribution network. In terms of encoding, it can be quite vanilla. You can completely immediately see if you do less bit rates or if you do lower bit rates, or if you do higher compression even and produce lower bit rates. Sometimes you might have a different energy profile. But that's probably only a small part of the picture. If you save 500 watts on your encoder, it may be that something you've done has put 5% more energy on all the decoders and you've suddenly got 30 million viewers consuming another 30, you know another God knows how many megawatts off the grid for the sake of your half kilowatt saving at the core of the network. So it has to be really complete in the picture before we can really say do this. But those are the sort of areas that people should be exploring and understanding within their workflow.
Jan Ozer, Streaming Learning Center:Is there a simple answer to you know? Does it take less energy to transmit lower bit rates versus higher bit rates, or do you have to look at the decode side again, as you just mentioned?
Dom Robinson, Greening of Streaming:It's not a simple answer. It really isn't a simple answer. The network is not provisioned against bit rate. It's provisioned against expected peak capacity. So when you don't send data, you don't save any energy in the transmission network. So the bitrate doesn't matter to the telco or the CDN. The CDN caches are stuffed full. It's just how quickly you recycle things, but by not storing stuff in the CDN, you don't reduce your caches.
Dom Robinson, Greening of Streaming:But when it comes to the consumer device, you can set a flag in the codec which trips a decode on a laptop from GPU to CPU, and see your fans go on, just because something's changed. That might be bit rate, it might be you've switched HDR on. It might be you've changed your frame rate. It might be any number of factors that are just incompatible with the target device. So, uh, it's a minefield to come up with the best practice, um, and it's certainly a complex uh that there isn't really a do this. But what I would say, though, is don't over produce, you don't. You know, if you produce 3,000 runs in a ladder, you're probably going a bit far.
Jan Ozer, Streaming Learning Center:One of the things we talked about that was surprising to me in terms of your challenges was agreeing on measurable objectives, can you?
Dom Robinson, Greening of Streaming:go into that. There is a macro picture of how much of a problem are we trying to address and how can we measure success in that. So that tends to have the headline grabbing figures that I'm very terse about using these days. But at a macro end, the IEA produced a figure that 1% of the global final electricity supply is currently being used by data centres, that 1.5% is being used by telco and 6% of global final electricity supply is being used by ICT. But that ICT figure includes the telco and the data center energy consumption. So at a top level we're probably, if streaming is 60, 70% of the network, we're probably and this is conjectural we're probably using 2% to 3% of global final electricity supply. So in terms of measurable success for greening and streaming, if we could feel that in five years' time, three years' time, we've changed that trajectory and we're looking at something more like 1% 1.5% of global final electricity supply, we're probably doing the right thing, especially if everyone's still entertained and getting a great experience In terms of the sort of smaller, more microscopic, measurable elements.
Dom Robinson, Greening of Streaming:I've touched on this correlation between gigabytes and carbon and we're in caution about making any direct correlation. There are in caution about making any direct correlation. There we are looking at different measurement techniques for the different, but basically it's a bit like sort of deep physics. It depends on perspective. If you measure a system that's used by many different shared services, you have to try to work out how you can report on any one of those services in terms of its energy, and the models vary extensively. We are tending to try to take global sort of top level systems energy, ie comparable to what your electricity supplier would be billing you. Energy, ie comparable to what your electricity supplier would be billing you, and look for fluctuations in that based on load, so that encompasses a lot of the baseline energy which is really being burnt in order to make streaming possible. The problem comes when you want to really say well, how much did my stream use through that system and you have to apportion it. As soon as you set up an apportionment model and you get into deep complexity about when systems are being used and who by and what the other processes are and so on, and it becomes extremely difficult to come up with a clear cut way to say how much energy any one particular stream is using.
Dom Robinson, Greening of Streaming:It might be a different exercise than one we're kicking around in one of our projects that's come out of the Lessercord to explore what Will Law kindly nicknamed the breadcrumb model, where, as a file passes through an encoder, you kind of put a meter in the file and you start the clock reading the kilowatt hours as you start the encode and read it again at the end of the encode, and even if there's other encoding processes going on, that is the energy that's being used in creating that file.
Dom Robinson, Greening of Streaming:You record that increment of kilowatt hours as it moves into storage. You repeat the process and add it, and so by the time the file gets to its destination, you've got a sort of cumulative clock of how much energy has actually been consumed by the computers that have got that file to you. The problem is, if you have two files that come in parallel through that workflow, they're going to accumulate the energy twice, so it's going to give a very skewed picture of what we think the energy going into the system is. So perspective is complicated and it means that there's no one simple answer.
Jan Ozer, Streaming Learning Center:Yeah, one of the things that you've spoken about here and you spoke about in our meeting was the whole measuring, the systemic load. Can you load that image that you had, and I'm not sure how well it's going to show up in the Zoom, but I think you've been talking about how challenging it is and I think that was a great illustration of exactly what you're talking about. So why don't you assume that people can see the outline but not the detail?
Dom Robinson, Greening of Streaming:and just you know, take two minutes and walk through this I think I can zoom in a little bit like that as well. So yeah, so basically, when uh, we were once we formed the group and we were having our first meetings to talk about what the group wants to form as working groups, um, I, I've been reading a lot of the sort of science which is trying to relate streaming to climate science, and one of the things that stood out is I didn't feel people had a complete picture of what was going on, and so the first thing I thought was useful was to do an exercise where I redrew a live stream, as I've looked after for what 30 years now, from the perspective of every box that might go wrong, that might break the link. Ie, if I didn't have an internet, what boxes would I need to plug in to recreate a single point-to-point live stream? And so that's what this diagram sort of seeks to represent. So you know, and it also separates it into things like network layers. People talk about layer three and layer four. This is kind of the network layer view. I've also sort of sectioned it as we go along so you can look at where people are talking about contribution feeds or access loops or whatever. Some of the terminology is matched against those different stages of the network as we go right through.
Dom Robinson, Greening of Streaming:So, following from the top left where we've got usually two encoders I drew two to emphasize that it's usually a dual path you're feeding into, typically, maybe straight into some sort of system, sort of systems. Sometimes you're feeding into some sort of local digital asset management process, into switches, out of your router, into your first mile, if you like, up to your local exchange. You might go through some authentication up in layer four and then you go all the way up into the core exchange where you might encounter something. I drew this box on because I think it's probably best explained. If you think of a point of aggregation, like maybe YouTube Lives Hub or a point of origin, notionally that's a point of origin and then that feeds into your CDN. Obviously that would normally be a big cloud, but in order to stream in and out of CDN, potentially you might be able to hit the local edge and bounce back out of just one box. So, keeping things to a minimum the network carries on, you might have your exchange caching, you might even have some much deeper caching in the system and then, if I just move the video window right over on the right hand side, you're into the access loop now, not?
Dom Robinson, Greening of Streaming:It's funny, a lot of people who, who are involved in greening and streaming and beyond, I don't think they'd seen it really drawn out as a physical map before and thought about it in those terms. Um, that's what it looks like behind the code when it all gets delivered, uh, and though that gives some idea of the amount of kit that's actually getting the stream from A to B, when you actually include layer two in the internet, the physical layer underneath the internet, the things that I think one of the things that's really important about this diagram is, on the right-hand side, that last mile router and the set-top box highlighted in orange, the kind of dot, the thing that's going to do the donkey work of decoding in the facility or at the home. Um, that orange box goes on and off. You know it's probably on if it's a domestic box, maybe four hours a day or something, um, but everything else pretty much. Maybe the display goes off, but everything else pretty much stays on 24 7. So you have, if you start to look at the energy involved, everything else is is. You know you're taking the integral over time, so it's on a lot of the time. Uh, relative to the set-top box. The set-top box is probably computationally doing more work than anything in the network, apart from maybe if there's some transcoding going on in some of these early stages.
Dom Robinson, Greening of Streaming:But that's not really what we're trying to represent in this. It's more of a true internet end-to-end model here. So really the compute cycles are going on in the encoder and the decoder, um, but everything else is on, and also everything else is on and quite scaled up. So for example, here where I am outside my home, the um, the first amplifier at the top of the telegraph pole which splits my um, my copper line from me and my, my neighbors, that's running 400 watts, and then the first cabinets running a kilowatt at least, depending on how many ports they put in it. So you start to rack up quite a lot of energy in that infrastructure. Although it's shared, I take the point. But in terms of proportionality for a single user really the weight of the energy is off on the left here into the core of the energies, off on the left here into the, into the, into the call of the network.
Jan Ozer, Streaming Learning Center:Aidan McCullen. So you've described this. You know I'll call it a big elephant. I mean, how do you, how do you take a first bite, how do you, how do you approach it, to start making progress?
Dom Robinson, Greening of Streaming:David Fubini. You know, again, returning back to the lesser core, the effect. If I'm gonna, I'm going to share something again, if it's right, because I think this is the right time to talk about it. So, after talking to a really good cross-section of the industry's engineers about what their intuition is, about where they think the energy savings might be, one of the key ones is summarised there as our intelligent distribution model shifting. So it's a bit of a mouthful, but we have these camps of unicast and multicast. So networks are using their energy in capacity and the capacity is provisioned to cater for their largest peaks. So anything that can help mitigate those peaks is useful. And those peaks are almost invariably big live and typically football games or sports events. They're the really big hard-hitting impacts that networks really really get paid a lot of money to guarantee delivery of and they really provision themselves so that they can handle it. But there's these models like multicast and even peer-to-peer which appear every now and then and claim purport to be able to scale live streaming much more efficiently. Now efficiency is a big moot word efficiency in data, efficiency in economics and efficiency in energy are three related but not connected things. So simply saying one's efficient over another is a complicated thing. What we're going to set about doing is one of our key lesser-called projects over the next 12 months is getting some energy measurements of as close to possible, some apples to apples comparisons between those different models at different scales, to try to provide a bit of information which operators haven't yet had, which is when to choose to use those different models. Because, honestly, deploying to choose to use those different models, because, honestly, deploying a little bit of compute to everybody's home router or in every DSL exchange or wherever the fan out points are, in order to enable stuff like application layer multicasting or peer-to-peer or whatever it may be, those will increment the energy used in all those relay points. And if you're doing that across your whole network because you might have 30 people watching the stream, then it's clearly not as energy efficient as sitting with something that might be less, ostensibly less energy efficient, like brute force unicasting. However, once you get up into the hundreds of thousands of millions, it's much more complicated. You have to look at the regions, you have to look at the network access critically, you have to understand the devices and so on. So that's a big area of research to look into. Which are the big scaling models that can, or what are the energy decisions around when you use those big scaling models? So I think that's a big area of interest.
Dom Robinson, Greening of Streaming:Codecs and compression and network optimization is obviously ongoing. It's a big part of it. It's a big school of discipline within the streaming industry, and rightly so. It's where we use a lot of energy in encoding files, which I know you know about at NetIn, and we use a lot of energy in decoding the set-top box in the TV environment. That's much more complicated in the bring-your-own-device era but it's still an area we need to get much deeper understanding of. Most of the work on set-top boxes to date has been done under lab conditions. There's very little which has been done on the, particularly in the bring your own device environment, in anything like a real user measurement environment. So there's a lot of research to be done in that space.
Dom Robinson, Greening of Streaming:Measurement itself I've mentioned the breadcrumb model is another key area that we really need to come to a little bit more consensus over. However, I think part of the reason there isn't much consensus is because there hasn't been that much practical experimentation in the networks. Everyone's waiting for an academic decision that they can implement, but the academics don't have enough data from industry yet to really have those ideas refined, but the academics don't have enough data from industry yet to really have those ideas refined. So we need to close that loop, and measurement is really the central conversation in that discussion. And then, finally, the hardware and infrastructure optimisation. Atomically we can do quite a lot easily in lab conditions. We can measure things. But cool things like immersion, cooling and fundamentally changing how we design our build out of data centers and storage requirements and what expectations we have and how those affect those architectural decisions at scale. There's still quite a lot of discussion to be had there.
Dom Robinson, Greening of Streaming:There's some fun projects in that space as well. That's big iron stuff. That's where we get to play with the telcos projects in that space as well. That's big iron stuff. That's where we get to play with the telcos. So those, I think, are the hot spots. Um, those, those are really the hot spots that we're really focusing on. I think there are other hot spots in every sector. Every standards body has its own domain where they're the experts. Um, what we're hoping is we can wait, wave in our flags and say, hey, folks, latency is pretty boring now. There's loads of other innovation around energy to do and get the engineers to start focusing on doing exactly what they're doing today, with exactly the same latency, but doing it with half the energy. That's a great area to start.
Jan Ozer, Streaming Learning Center:One of the discussions that kind of got hot on LinkedIn maybe 18 months ago was VVC versus H.264. No-transcript. What are your off the cuff thoughts about the more advanced codecs that are much more difficult to encode and probably decode, I think?
Dom Robinson, Greening of Streaming:the one thing that really stands out for me about codec deployment is rolling out, churning out kits to support something new is probably takes many, many, many years of win in terms of an operational performance. To replace, replace if everybody goes well, I'm going to have av4, so I'm going to bin my laptop, I'm going to bin everything that runs av1 because it doesn't run av4 and churn it all out and my stuff doesn't get recycled properly and ends up in endless landfills or just, you know, being inefficient somewhere else. That that has to be part of the equation. So it's never as simple as this is uh, uh more in a five percent more energy efficient codec. You know there is a strong argument that h264 is probably um so ubiquitous and does the job. You know sufficiently well enough. Uh, that, uh, do we need anything more and are we actually focusing on the wrong problems? You know we are being challenging or purposefully wanting to ruffle some feathers. Is the return on investment into CODA really worth it when we could be investing in energy efficiency?
Jan Ozer, Streaming Learning Center:I'm reminded of looking at buying an electric car and someone saying the best thing you can do to save electricity is to keep using your own car and not buy a new one.
Dom Robinson, Greening of Streaming:Correct, one of the biggest. We could sidetrack into that. But buying a two and a half ton SUV electric hasn't really saved the problem. You've probably used half the planet in building the car itself, so we have to think about those embedded emissions, although Greening Stream doesn't talk about carbon. We have hardware vendors and chip manufacturers and infrastructure providers in all our working groups and they're constantly sitting there thinking about the physical resource, because it's a very complicated argument as well. One of the most subtle ones is actually driving.
Dom Robinson, Greening of Streaming:There is an argument that driving more bandwidth onto the networks drives innovation, and at different stages of innovation we can get exponentially more efficient with real-terms reductions in power. And this you can see it in fibre networks at the moment. As you go from one generation of fiber to another, you're quite often getting a hundredfold increase in capacity and, in real terms, a 0.75 reduction in power. So the question becomes how quickly can you migrate everyone off the old stuff and recycle the old tin, which means you need traffic to drive the economy to invest in that churn and you'll actually reduce the energy, even though the capacity will go up. It's not following Jevons paradox in its simplest way.
Dom Robinson, Greening of Streaming:Normally, like 4G to 5G, we've actually had a real-time increase in power of about fourfold and no traffic's moved into the 5G network. So we've got a potentially more efficient network, but there's nothing going on, so the efficiency's not really valid until we're at basically 100% utilization, or at least comparable utilization to the previous network, at which point today we're just investing more power and not really getting any gains. So yeah, I think those innovation cycles themselves are non-simplistic. And churning codecs yeah, if you can massively reduce the power in your data center, that's probably a good thing. Reduce the power in your data center? That's probably a good thing. But if your data center is consuming what? 30 megawatts? But you've just put 150 megawatts on everyone's decoding because of all those little step-top boxes working that much harder or having to be binned and replaced, you've got to think hard about that.
Jan Ozer, Streaming Learning Center:One of the things we talked about in our planning meeting was a player control that a consumer would have. That would increase the quality at an increased cost. Could you go into that? Is that something that's soon to be appearing, is appearing now, is kind of on the long-term roadmap?
Dom Robinson, Greening of Streaming:Yeah, so I think that's referring really to the gold button idea that we were talking about. So this is the idea that it's probably worth sort of going a little bit more specifically into the human side of that. Last summer, when we did our sort of public launch over here, we did a survey around that to find out how much the consumer was aware that streaming used energy and, in that awareness, whether they were, whether they felt empowered to act or to do anything about it. We we, we looked at whether eco mode, green button, those sort of opt-in services had any uptake. And really they don't. When people get eco mode, their TV is shipped in eco mode and it's the first thing they do is turn it off and put the brightness up. They just bought this high-performance TV. They don't want a dull, crippled picture in the name of eco. It's just not the way it is.
Dom Robinson, Greening of Streaming:There was a clear sense that the industry had to take leadership here, take responsibility in many ways and put a menu of energy efficient options in front of the consumer. So rather than having a green button, which you opted into if you're environmentally, socially responsible, the idea was we ship you a green stream and because we're the internet. We don't have to take away consumer choice. So all the people who go, ah, panic, consumer choice has been infringed. Well, it hasn't. We've given you a great big orange button or a switch or some sort of interaction of any sort you want, which basically allows you to upscale or upgrade or boost your signal for the duration of the match or the duration of the box set viewing or evening's viewing, because you're sitting there getting value from it.
Dom Robinson, Greening of Streaming:And while I want to save energy, I'm also a human. When I want a coffee, I boil a kettle, but what I don't do is I don't leave my kettle boiling all day, just in. When I want a coffee, I boil a kettle uh, you know we. But what I don't do is I don't leave my kettle boiling all day, just in case I want a coffee. Uh, that that doesn't have efficiency, uh.
Dom Robinson, Greening of Streaming:And what we are doing in real terms with a lot of broadcast and streaming technology strategies is we push everyone up to the highest possible run on the ladder. So my boy is dribbling on the carpet watching Peppa Pig at UHD. You know if that's got an eightfold increase in traffic and energy impacts and who knows what, do you have to stop at some point and say do we need to be by default offering my boy Peppa Pig in UHD? I'm sure he'll put up with it in 2k, and and so I think I think it's sort of a not an existential argument, but it's a directional argument that we have to ask ourselves, you know as a whole, when we're over engineering, why don't we divert that resource to doing, you know, doing either more energy efficiency or just brutally better engineering anyway? So I think those are the key questions and that's really what the gold button idea.
Jan Ozer, Streaming Learning Center:What's upcoming, what are the milestones you see hitting in terms of meeting and formulating and announcing and releasing specifications and best practices?
Dom Robinson, Greening of Streaming:So, after pretty much seven weeks on the road, as of today I'm chilling out, writing a couple of papers and then having a little bit of a break in August, because it has been immensely full on the last few weeks, a few months actually. What I'm hoping to do, though, by writing those papers, is get them out to project leads. We think we found we're trying to isolate project leads for those four projects that I showed you a minute ago. We've got, basically, volunteers for all of them and then to go through those papers and just agree the project scope and project plan, and hopefully, those project leads will engage with the members and some guests, because we've got gaps in our memberships and guests who are going to help fill the gaps, and so we can stand those projects up. And so what we hope to do is come to IBC this autumn and, fingers crossed, ibc are going to fit us into the programme somewhere, so that we can stand up and say, right, these are the full projects, this is what we're testing and this is how we're going to go about testing those projects. This is what we're testing and this is how we're going to go about the testing those. Uh, as a little bit of an opportunity again for the industry to say you've missed something or we think this, you know, just keep that feedback coming in. Transparency is important here. Uh, and then um, in providing that that we, you know, we feel that we're on to the right thing. Then through October I hope, the projects will sort of form, start to form, the technical work will start to sort of frame itself out and then the aspiration is in Q4 and Q1, when the streaming is sort of busiest in Northern Europe and America, we're going to get some real testing underway and that should bring data into our academic communities hands.
Dom Robinson, Greening of Streaming:I would have thought by the beginning of Q2, by the end of Q2, if we are lucky, we might be able to actually give some guidance in a model scenario. But the core system will be as near to production as possible. What will be modelled still at that point will be the client devices. They will all be going into test facilities, because it's going to be too complicated if we do everything in one go. But as we finish that project we should have finalised the planning for some research grants which we're looking to collaborate on, which will then hopefully take the whole project to the next stage over the following year and go into an exploration of what's really going on in the consumer premises equipment environment, so the player, the set-top box on the TV, the media player, the laptop, smartphone, all of those.
Dom Robinson, Greening of Streaming:We want to go from lab to real world measurement model over, sort of from summer next year forward, so that we can take our core models that we've hopefully settled on and test them in real user environments and actually check that by the time we bring this into a sort of OTT, bring your own device world, that those attempts to bring down the energy of our systems actually really work in the in the wild. Uh, so it's sort of a bit of a two-year plan, um, but it's also. It's also changing and full on on a day-by-day basis, email by email. It's. It's quite intense at the moment, so, um, so that the plan from Greening and Streaming's point of view.
Jan Ozer, Streaming Learning Center:Tell us who's in Greening and Streaming now and where to go to. What it costs to join, I guess, and where to get more information.
Dom Robinson, Greening of Streaming:So these are our current members. So there's sort of a couple of different groupings in there. The top two rows are our founder members. They came in on day one and then scattered Well, it's in order that they joined basically and then scattered amongst the members there after a few public service broadcasters who can join and get involved for free as long as they're participants for free as long as they're participants. And then likewise, more recently, we've extended that to affiliations with other peer industry groups like the Digital Television Group and IABM, and there's a few others coming in as well.
Dom Robinson, Greening of Streaming:So we just swap one secretariat membership so that we can keep an ear to the ground on what we're all doing and make sure that there's good collaboration between the organisations. So it depends on turnover. We don't really means test, but I think it's sort of people just play the honest card. So over 10 million turnover, it's £12,000 UK a year and it drops down to under a million turnover, which is about £2,000 a year. So we are, by our own admission, not a cheap organisation to join and that was by design. We did not want to become a green badge which people could buy and be non-participatory and tick the box and say I've done the greening and streaming thing, I'm now green. We have a fundamental rule in the organisation of no greenwashing and that applies to membership as well or specifically. So we want organisations to stop and think before they get involved and get some groundswell of support from across the organisation.
Dom Robinson, Greening of Streaming:So of the sort of 25 organisations we've got involved, there's about 130 signed up and active on the hub people and then in the background as well, it's always worth mentioning that I have about a dozen volunteer secretariat. I have about a dozen volunteer secretariat. So we don't have an individual member class, but there are some really good consultants and experts who are really well connected across the industry or have retired from leading jobs in the sector, who help me on a day-to-day basis with lots of the admin, lots of the sort of group leadership and setting up meetings and those sort of things, and they're invaluable. And so for people who are exceptionally keen to get involved and bring a lot of network, bring a lot of skill, expertise, running these sort of organisations because I sure as hell am making it up as I go then I do welcome people interested in getting involved in Secretariat, but I do ask them to do stuff for it. So yeah, that's sort of the big picture at the moment.
Jan Ozer, Streaming Learning Center:I guess, on a personal note, how are you? You know, ideas is launching their first product, or their first major product, norik. Now how are you balancing? I mean, you're the chief, what are you? Chief product officer, chief uh chief biz dev officer.
Dom Robinson, Greening of Streaming:So so, um, greening stream has definitely taken over my time. I thought for the first year of running it that I could run it arm's length and hire a friend who's a management consultant to run it. Uh and uh, he did a really good job, but he didn't know the sector and so there was so much referral back to me that we had to make a bit of a decision. Last year, part of my role with ideas used to be our marketing. We used to take a bit of a black ops, special ops approach to finding really specialised, big, scaled up live streaming platforms in the fintech and sports are even to help. That was cool that they came in but that had a lot of professional services on it and we wanted to really scale up the core licensable technology and ideas, which is what's Norse. So we've slightly repositioned the business and I'm a good guerrilla marketing guy I'm not necessarily a brand marketing guy and we thought Eric, who obviously you know well, was the man for that guerrilla marketing guy I'm not necessarily a brand marketing guy and we thought eric, uh, who obviously you know well, uh, was the man for that. He's seen every pitch under the sun and would probably guide us well out of the darkness and he's doing a great job in in, in helping us brand norskin and at the moment a lot of the effort is I'm raising that awareness on the marketing side of it. Um, at the end of the day, I wear an ideas t-shirt because I'm you know, that's part of my, my day job and everyone knows that I'm.
Dom Robinson, Greening of Streaming:Ideas is a member of just as much as the other contributors to greeting and streaming and members. I happen to wear the hat of running it at the moment, um, because we've kicked it off and that's the way the cookies crumbled at the moment. But you know, but it is a members-led organisation and we will be looking to sort of make sure that I'm not dictator in chief and not letting anyone else have a go in due course. So I fully expect to return to ideas over time, but at the moment, greening and Streaming I have the capacity four days a week, to be doing Greening and Streaming and keep my eye on the business as we started to build out on your product set and it is full on. It is very, very. I really thought greening and streaming is going to be a sideline conversation. I had no idea it would be two full time jobs in one. I think it's just off the moment, and a bit of serendipity, which is good.
Jan Ozer, Streaming Learning Center:I think it's just of the moment and a bit of serendipity, which is good. Okay, question coming in.
Dom Robinson, Greening of Streaming:Not seeing a lot of telcos on the membership slide. No, we're definitely weak on telco operators. We'd love more and we've got access to broadcasters through the EBU. The zone are a bit of a flagship content provider amongst our midst and a lot of our testing is we're turning to them to source. But yeah, we definitely want more engagement from telco operator and we've got a good relationship with content where we need it. But we are always interested in talking to content providers from the top line down.
Jan Ozer, Streaming Learning Center:Question what are the challenges of an organization that can issue recommendations only but not enforce them? So I guess, how are you going to? You're going to come up with a bunch of recommendations. What happens then?
Dom Robinson, Greening of Streaming:So I think in practice you can't really enforce technical standards anyway. So I don't even think of the standards body as being enforcers. Maybe they they protect patents and stuff. Stop people, stop people deploying things. But I I I've never really thought you can force standards down people's throat, unless you're sort of apple and you decide hls is the world, away the world, and so on.
Dom Robinson, Greening of Streaming:By engaging the big oil tankers like Akamai, intel, imd there's loads of them among that group and by engaging more and more of them, I think we can change the culture of the industry. And they used to call it in public affairs terms. They used to call it um, self-regulation, uh. That term tends to be favored more by uh, or the term industry-led reform tends to uh, tends to actually be more favored. And green and string definitely wants to lead the engineers to think in a better way. I don't think they need to be told what to do.
Dom Robinson, Greening of Streaming:Most engineers I know are intelligent and they love a problem, even if they're nutcase climate deniers. In practical terms, they also like making profits. So we sit on the fence with that. In practical terms, if you're all about lowering your bottom line by reducing energy, then find greener streams for you. If you're all about saving the environment by reducing energy, then find greening streams for you. It doesn't really matter. We can't force it on anyone. But what we can do is make the culture of the industry sit there and go. You know what that's really energy inefficient. I'm not going to tell the boss. We can do that, and that sort of direct action from the engineering floor will change things very directly.
Jan Ozer, Streaming Learning Center:Certainly it's a worthwhile cause. Couldn't have a better leader for at least the initiation phase of it.
Dom Robinson, Greening of Streaming:Yeah, I'm sure they'll get. Everyone will get tired of me soon, I'm sure.
Jan Ozer, Streaming Learning Center:I appreciate you taking time out of your busy uh dual. No, not at all.
Dom Robinson, Greening of Streaming:This is my busy job, so this is a pleasure.
Jan Ozer, Streaming Learning Center:Okay, cool, Good to see you again and, um, you know we'll be talking, I'm sure, in the near term.
Dom Robinson, Greening of Streaming:Speak soon.
