The OTT Lexicon For 2023
The world of streaming is defined by acronyms like SVOD, AVOD, FAST, OTT and more. But this leaves gaps and confusion in what is included in our OTT services. For example, what does a service like BBC iPlayer include? What is watching the World Cup on RAIPlay called? Do we clearly capture live video streaming in our standard terminology?
There is currently a morass of terminology in play, some of it enlightening, some of it confusing, and maybe it’s all become a bit “OTT”. As we begin what is bound to be another record-setting year for streaming, it seemed appropriate to start out with a review of how we describe ourselves in the streaming industry.
This article was inspired by experiences at an industry event I attended at the end of 2022. These experiences confirmed that there is a relatively high level of confusion about the terminology we use in the streaming media industry, even among some of the leading managers and practitioners.
The first experience involved watching a presentation describing the differences between FAST and Linear TV, where one of the differences between the two was identified as there being “no need for infrastructure” for FAST. As a streaming executive involved mostly in the “infrastructure” side of the streaming industry, I felt this was an interesting, and perhaps confusing, perspective. This subject was discussed in detail over dinner with industry friends, who like me had worked in the traditional broadcast industry before moving to work in streaming.
But that same dinner revealed that my friends, who are leading practitioners in our industry, had not heard of the term “BVOD”. This also resulted in a heated and enjoyable conversation as we discussed all the acronyms we use to describe different types of media delivery. As an editor focused on the streaming media industry, I was reminded that not everyone working in the industry reads the reports that our industry produces about itself, and not everyone uses the same terminology or understands the terms others are using to describe the space we are working in.
Following this conversation, I started to pay more attention to how streaming services were described in the media and industry articles. And it is confusing. FAST vs. AVOD vs. BVOD is just one simple example – a broadcaster’s OTT service can easily be described as any of these 3 terms. SVOD services on the other hand, now include Live and Linear content, so why is it VOD? Even advertising is being included in subscription-based services. The lines are blurring.
On top of this, from a technology perspective, some important terms appear to be over-generalized to the point that they always need further clarification – like Edge and Head-End – or that are emerging in an already congested space and could do with some positioning against their peers – like HbbTV and DVB.
So, with this inspiration from my friends and industry colleagues, I decided to address the terminology confusion that I’ve observed (and can sometimes feel myself), to at least serve as a reference point for fellow industry participants and analysts. Even if, as consumers, we can argue that we don’t really pay much attention to how content types and streaming services are classified when we turn on our choice of streaming device, and simply think to “watch Netflix” or “see what’s on BBC 1”.
We use a long list of terms today in our industry. But as the streaming industry evolves, and streaming services become “multi-everything” (i.e., platforms, content types, business models), some of these terms can become confusing and even misleading. Let’s start with the following list:
SVOD. AVOD. BVOD. FAST. Catch-up. OTT. Streaming. Linear. Live. Pay-TV. Free-to-air-TV.
Now let’s consider the following services and think about how we would categorize them:
Netflix. Amazon Prime. DAZN. Peacock. ITVX. BBC iPlayer. Hulu. Tennis TV.
Netflix and Prime have always been thought of as SVOD, yet Prime includes a lot of live content, and Netflix is launching live streaming. Should they both be called SVOD?
DAZN is a live sports streamer, that also produces DAZN Originals and has linear content on its platform. DAZN is primarily a subscription service, but why would we call it SVOD?
Peacock and ITVX are ad-supported platforms, with premium ad-free tiers. But they include linear, live and VOD content. Would we use the term FAST to describe any of their content? Are they AVOD? Is their premium tier SVOD, even if it includes linear and live content?
Hulu is a subscription-based, ad-supported service. So is it SVOD or AVOD? But it includes Linear channels as well as VOD, so what do we call that?
BBC iPlayer is license-fee funded, so no ads and no subscriptions. It has been labelled as BVOD (Broadcaster VOD) in most analyses (along with other Broadcaster OTT services that are ad-supported), yet it includes its linear channels and plenty of live events. So what should it be called?
Tennis TV is the ATP Tennis Tour’s D2C offering that includes live and recorded content for tennis fans, that are not part of rights packages in various markets. Highlights packages and some VOD content is free, while live content requires a subscription. So how do we describe this?
This line of thinking caused me to come up with a primary observation which is that, particularly from a broadcaster’s perspective, the terminology we use is very limiting, and specifically does not help us describe Live video streaming. The name “VOD” is almost synonymous with “streaming” and yet, as every one of these examples shows, VOD is always only one part of the content available on the platforms. And, more to the point, as we look to the future of streaming it is highly likely that live content and linear content will both be major aspects of how streaming services will grow and drive revenues for media companies in the future.
The development of FAST channels is a case in point. Two years ago, we had not described free, ad-supported, linear streaming services as “FAST”, but they still existed. They were the same as the linear channels that many people would watch over-the-air or in a free-to-air TV subscription. Every national broadcaster’s channels – literally all the household names you can think of per country – that have ad-supported business models and that were already supplying their linear channels as part of their free streaming services were arguably delivering FAST channels. It’s why the debate between colleagues who have been in the broadcast industry so long and yet have also worked in streaming for many years turned quite heated and entertaining as we discussed the concept of FAST vs Linear (and it’s why, to unpick some real nuances between linear and FAST, there will be a follow-up article on this subject). In the end, FAST may well replace the word “Linear”, but before we go there, I have tried to create a way of considering all these names.
A New OTT Lexicon
The lexicon of OTT, in other words the vocabulary we use, could use some clarification to create order from the slightly or strongly confusing situation we can find ourselves in as we discuss the various types of services in the market. The following approach to defining OTT services helps us map content types and business model types into a set of service types that can be offered. With this model we can see how all 9 types of services available in the market are offered, in different ways by different providers.
The terminology used above to describe the 9 Service types might not be the most gripping. “FAST” and “VOD” are certainly helpful in that they are simple to say and easy to remember. But they disguise the idea of Live video, which is a key area expected to drive much more value in the media industry over time. We should be able to easily describe how live, linear, and on-demand content features in a streaming service. It would help us all communicate about our streaming industry more easily.
And finally, I’d like to make a bid for TV or television, to remain in our vocabulary, and if it doesn’t then I’d like to see it replaced by something that could become as utilitarian and ubiquitous in the industry to cover all types of content we view. “Television” is from the Greek word “tele” meaning “far” and the Latin “visio” meaning “sight”. This word’s first recorded use was in 1900 in a paper presented by Constantin Perskyi at the first International Congress of Electricity during the World Fair in Paris. Over time “television” has come to be synonymous with the device in the living room or the content provided by broadcasters. But, whether we are streaming a video to our IP-connected device in the living room, our garden, or on the bus, or whether we are receiving it via satellite to a dish or via a transmitter to an antenna, it is still “content being seen from afar”.
So, while 2023 is bound to contain a host of innovations and new breakthroughs for the world of streaming, we can also remind ourselves that we are continuing to build on the long heritage of television which has its origins well over 100 years ago.
As the 2020s are an industry-redefining era for the media industry, and as we are at the start of a new year which is bound to be another record-setting time for streaming media services, it felt like a good time to reflect on some of the key terms we use and to create some clarity about what they really mean.
Possibly one of the most important concepts in streaming media is “Edge”. In its simplest sense, the Edge is about distributing data processing to a point that is as near as possible to the consumer. This is neatly related to the concept of just-in-time pull systems used in product manufacturing that leave the final assembly of a product to the latest possible moment that best combines flexibility, personalization, and efficiency. In the media industry, our product is a video, either delivered live or on-demand. As streaming continues to grow, the concept of the Edge, and what we do at it (i.e., the Edge functions), will become more and more important because the systems we use must become more distributed in order to be scalable and efficient for delivering lots of personalized content to very large audiences.
The word “Edge” has been around in the IT industry and the Media industry for a long time, but it has had a different evolutionary path in each domain, for a simple reason.
In IT, there is a concept that the source of data is the consumer (let’s include “employee” in this concept as well). This results in an exchange of data that is 2-way between the consumer and the receiving/responding entity (e.g., a business, government, etc.). In practice, a consumer enters data into a system (e.g., a search engine or an e-commerce website or a CRM database), and the system processes it and responds. To improve the responsiveness of systems, the ability to respond should be as close to the consumer as possible. And so Edge computing became a concept. Cloud service providers are major parts of the technical ecosystem to provide edge computing that makes the internet work quickly and efficiently. Building out compute locations that can host systems (i.e., software, applications) is at the heart of cloud service providers. But enterprise IT has been doing this in its own ways as well. For example, a retailer that deploys systems in each of its retail locations, on local servers, is effectively providing edge computing for its users. An oil rig operator with local servers deployed on individual oil rigs is doing the same thing.
Recent IT industry commentary has raised the idea that the word “Edge” in IT circles is so vague that it is almost becoming irrelevant. The point is made that Edge “products” have had various modifiers added to them in order to distinguish their purpose – for example, network edge, telco edge, near edge, far edge, industrial edge, and edge datacenters. It is understandable that in a vast 2-way communication ecosystem, while the concept of “edge” is straightforward, there can be a myriad of different use cases and deployment models. But in video streaming delivery, the situation is clearer, albeit there are still opportunities for confusion.
The Edge in OTT streaming video is the server that distributes the video to the consumer, and it is the last link in the CDN chain before the video is transported into and across the ISP networks to reach the consumer. Differently to IT Edges, a Video Edge is a one-way traffic system as video is requested but then pushed on-demand, without consumer-driven modification to the data-set, from the server to the consumer.
Figure 1: Edge locations in video delivery, used independently from one another or used jointly.
As shown by the diagram above, these Edge Servers are found in 3 locations today. The primary location, CDN Edge 1, connects to ISP networks in a general way, labelled as “transit”. This is a general-purpose connection method that has the lowest level of quality control compared to the alternative deployment locations. The CDN Edge 1 will connect to an Internet Exchange of some type that provides onward connectivity to all ISP networks in that physical location. This approach is typically used when delivery volumes or peaks are lower, and the broadest reach towards all ISPs in the market is important for the CDN.
CDN Edge 2 takes things a step further by connecting to the ISP in a direct “peered” way. This creates guaranteed bandwidth for the CDN into that ISP network, but it does not necessarily provide guaranteed bandwidth to the OTT service provider using that CDN (i.e., if the CDN is a multi-tenant platform). Direct peering is used when the scale of delivery by a CDN to a single ISP reaches a point whereby there are quality of service and economic benefits to be achieved by setting up dedicated network capacity for the Edge egress to use.
CDN Edge 3 takes another step. The Edge moves inside the ISP’s own data centers, further along the network path towards the consumer. This location offers lowest latency in responding to a request for video from the consumer, because it avoids sending the content from a point further upstream that can potentially suffer from network congestion as all the video is transported downstream back towards the consumer. The most advanced streaming services, like Netflix, YouTube, DAZN, and BBC iPlayer use this model in a Private CDN mode (i.e., Edge server capacity dedicated to their own use). Some Public CDNs like Akamai have deployed Edge servers inside some ISPs around the world to provide best possible service to their customers. In addition, an IPTV or CableTV service, like BT TV or TIMVision, that uses a private CDN platform only for that particular service will have Edge servers deployed inside the ISP’s network. In a typical set-up within a single ISP’s network in Europe, Edge servers can be found in 5-15 locations (PoPs) to serve a single country.
CDN Edge 4 is a future possibility that could see Edge servers deployed at the closest possible position relative to the viewers. A country like the UK is expected to have about 1000 exchange locations in the future, as fiber to the premises is fully rolled out. This would offer potentially 1000 locations for Edge servers that could serve very localized markets. For Live and Linear streaming in particular this could be highly efficient as a “stream propagation method” that still supports timeshifted viewing (i.e., pause, rewind, fast-forward), while for VOD content there could be layered caching rules where most popular content would be highly distributed according to demand and less popular content would be cached in more centralized locations. See article Telco Access Networks (& The Growth Of OTT).
So when we talk about Video Edges, which are the last point of video management before sending the stream across the broadband network to the device, we need to be aware that the video Edge could be in a number of different locations.
Head-End is a long-used term, originating and still heavily used in the CableTV industry, that incorporates multiple functions that prepare video for delivery to the consumer. In simple terms, after receiving the video signal from its source, the head-end receives them and converts them into a format for playing on televisions or other devices. But the head-end, like the Edge, contains a long list of functions (Figure 2) and options that can make one head-end very different from another. And these functions, rather than being fully centralized, can be performed in multiple locations along a delivery chain before the final stream is originated to the consumer.
Figure 2: Head-End functions that can be centralized or distributed according to network topology.
In many video delivery chains, there can be multiple “head-ends”. A contribution head-end could receive signals, then decode and multiplex them before onward delivery. A primary distribution head-end could receive that multiplexed signal and encode it for adaptive bitrate delivery. In VOD workflows, the head-end could ingest VOD assets, transcode them and store them ready for onward delivery when requested. A secondary distribution head-end could package and encrypt the content before originating it to the CDN. Other functions like ad insertion and content filtering could be applied at different steps in this chain.
In OTT video streaming, the concept of a head-end can easily be called an origination platform. An “Origin” can be a single term that denotes packaging, storage management, encryption, ad-insertion and stream delivery, although encoding and transcoding are typically kept separate from the concept of an Origin. See article OTT Content Origination.
While head-ends and origins are already being deployed in cloud platforms, which brings a range of operational and commercial benefits compared to fixed location and fixed capacity deployment models, several of the functions that we include in the head-end are candidates to become (or are already becoming) Edge compute functions. They would therefore take place inside the CDN rather than at the Head-end. Packaging and Encryption/Protection are two such candidates. CMAF (Common Media Application Format) could be used while streaming content from an Origin to an Edge, at which point JIT (just-in-time) Packaging into HLS, DASH or MSS would happen at the Edge. Protection methods like watermarking and tokenization are also classic candidates for deploying at the Edge, given that individual streaming sessions are initiated at the Edge.
So, while the word “head-end” gives the impression that it is a highly centralized location for all manner of video processing functions, in reality those functions are already often distributed across multiple locations, and some of the key functions are already making their way to the Edge. So how long will the concept of a “Head-End” remain as we move towards a world of Edge computing? Time will tell.
Technology is full of opportunities to mystify and confuse. It’s why we need engineers and developers to do their incredible work deep in the depths of software and electronics to deliver content to our screens. As a non-engineer, despite working in advanced technology companies my whole career, and as someone who focuses more on how the technology is used and delivers value to all of us in our everyday lives, I am continuously fascinated by how our content is produced and delivered.
The evolutions in Head-end and Edge technologies and deployment models are inevitably becoming more distributed and closer to us as consumers. Achieving maximum efficiency while operating at high performance and with more personalized services relies upon this more distributed delivery model. That said, there is a balance to find and one of the most interesting trends to watch in the streaming industry will be to see where, in the end, all the work to deliver streaming video actually gets done.
You might also like...
Learning From The Experts At The BEITC Sessions at 2023 NAB Show
Many NAB Shows visitors don’t realize that some of the most valuable technical information released at NAB Shows emanates from BEITC sessions. The job titles of all but one speaker in the conference are all related to engineering, technology, d…
The Streaming Tsunami: Part 1 - Seeing The Tsunami Coming
Streaming video is on the cusp of becoming a major problem for broadband networks. Up to now we have been dealing with a swell in the streaming sea that has caused a few large waves to crash on to the…
Machine Learning (ML) For Broadcasters: Part 10 - Automating Workflows For Compliance & Moderation
Machine learning and some other AI techniques are increasingly being applied for many aspects of media content compliance and moderation, ranging from technical compliance for parameters such as loudness, to vetting for hate speech.
Compression: Part 6 - Inter Coding
The greatest amount of compression comes from the use of inter coding, which consists of finding redundancy between a series of pictures.
FAST vs. Linear TV: Part 2 - What Are The Real Differences?
FAST can deliver the promise of true personalization intertwined with mass market reach. This heady mix can maximize advertising revenues from the large audiences currently turning away from traditional Pay-TV services towards free streaming services.