streaming is entering a new phase - csi magazine · and their stories, with over 90 minutes of...

ww

w.c

sim

agaz

ine.

com

June 2016

The magazine for global video

8K T

VIn

tera

ctiv

e st

oryt

ellin

gAT

SC 3

.0

April 2019

Streaming is entering a new phase

• OTT, VoD, Cloud TV, Cable, Satellite, IPTV, DTT, IoT • The latest news, views and features

The leading media for video delivery technology

TV Broadcast • Streaming • Satellite • Cable • Hybrid

App

le S

VOD

ana

lysi

s

1_Cover.indd 1 19/03/2019 10:59:19

WHEREBROADBAND

MEETSCONTENT

EXHIBITION & CONFERENCE4 - 6 JUNE 2019

REGISTER ONLINE NOW!

Cologne / Germanywww.angacom.de

BROADBANDTELEVISIONONLINE

47 % INTERNATIONALPARTICIPANTS21,700

CONFERENCE ATTENDEES

2,300FROM 37 COUNTRIESEXHIBITORS500

ANGA Services GmbH

Nibelungenweg 2 · 50996 Köln / Germany

Tel. +49 (0)221 / 99 80 81-0 · [email protected] www.angacom.de

Perspective Publishing3 London Wall BuildingsLondonEC2M 5PDwww.perspectivepublishing.com

Editor’s report:As OTT approaches maturity, the streaming market is entering the next phase in its evolution. On the one hand, this will involve increasing business model diversification, including experimentation with ad-supported tiers. DAZN is the most recent platform to go down this route, and even Netflix is thinking about it in some shape or form. At the same time, a new wave of direct-to-consumer

services is coming, led by some of the world’s largest media & entertainment brands that will further shake up the market. The two developments are not necessarily mutually exclusive. Interestingly, these trends are coinciding with the launch of Apple’s long-rumoured and long-awaited video streaming service, more thoughts on which can be seen on page 8 in the analysis by Ampere’s Guy Bisson. Goran Nastic

Contents

05 News & analysis

06 Data corner: Apple SVODAssessing Apple’s video streaming service

08 COVER STORY: DTCGet ready for the next wave of (mega) direct-to-consumer services

12 Low latency streamingA CDN mechanism for reducing latency

13 Machine LearningFrom ingest and metadata to recommendations, discover how media companies are increasingly relying on ML

19 VPNs and geolocationIP intelligence and geo-location is growing in importance

20 Interactive TVWill Bandersnatch popularise branching narratives and storytelling?

21 Live OTT sports How to improve the Quality of Experience

27 Blockchain in DRM The role it can play in future digital rights

28 ATSC 3.0Is it in danger of missing the boat?

32 Building live OTT Live content distribution over the internet

34 CSI Awards 2019All the categories in full

35 Media in the cloudAccelerating file transfer in the cloud

41 Industry column: 8K TVThe DTG zooms in on 8K television

43 Connected entertainmentTowards a more connected world

50 Technology corner: IMFThe DPP on the progress of the standard

2019

ww

w.c

sim

agaz

ine.

com

June 2016

The magazine for global video

8K T

VIn

tera

ctiv

e st

oryt

ellin

gAT

SC 3

.0

April 2019

Streaming is entering a new phase

• OTT, VoD, Cloud TV, Cable, Satellite, IPTV, DTT, IoT • The latest news, views and features



App

le S

VOD

ana

lysi

s

1_Cover.indd 1 19/03/2019 10:59:19

EditorGoran Nastic

CommercialCamilla Capece Daniel Torelli

Design and productionMatt Mills (Manager)Jessica Harrington

Regular contributorsAdrian Pennington, Philip Hunter,David Adams, Stephen Cousins, Anna Tobin

CirculationJoel Whitefoot AccountsMarilou Tait, Mohamad Saidani

Editorialtel +44(0)20 7562 [email protected]

Advertisingtel +44(0)20 7562 [email protected] +44(0)20 7562 [email protected]

Subscriptionstel +44 (0) 20 1635 588 861 [email protected] Circulation manager: [email protected]

Subscription ratesPer year: Europe £88; UK £68; Rest of World £98. Cheques payable to Perspective Publishing Limited and addressed to the Circulation Department

Printed by Buxton PressManaging DirectorJohn Woods

Publishing DirectorMark Evans

ISSN 1467-5935

www.csimagazine.com April 2019 03



03_Contents.indd 1 21/03/2019 11:31:38

news in brief

JP Morgan forecasts 160m Disney+ subs Disney’s direct-to-consumer streaming service could accumulate as many as 160 million subscribers globally, thinks JP Morgan. Analyst Alexia Quadrani reckons Disney+ will ultimately be able to attract 45m subscribers domestically in the US and 115 internationally for a total of 160m subscribers, depending on when Disney gets some of its content rights back. “We view streaming services as a positive addition to Disney’s ecosystem,” Quadrani said. Previously, Moffett- Nathanson estimated Disney+ could attract about 2.4m members in the US by the end of its first year.

Bx-WiFi targets hotspot video congestionA new technology called Broadcast WiFi could offer a way of solving the problems of wireless reception and transmission in high-traffic venues like sports stadiums and music festivals. Global Invacom is partnering with Edgewater Wireless Systems for the further development of Bx-WiFi, a technology they claim enables live, high quality and large-scale event video streaming over an 802.11 network. “This partnership could set a new record number for video streaming to a device on each hotspot,” they said. Tested by the BBC, it could compliment 5G.

News

How do broadcasters approach AV1? Much has been written about how the AOMedia grouping could ruffle feathers with its new AV1 codec whose spec was frozen around NAB 2018 time. While the FAANGs are committed to deploying it, and the four main browser vendors are on-board, the big question is how AV1 might be adopted by traditional non-internet/OTT players.

“There needs to be an effort to support the ecosystem, because if something gets broadly adopted you need to work on maintenance and addressing additional use cases. And someone needs to put forward the resources and make sure these things are executed all the way to the integration, support those who want to use the technology,” argued Thomas

Stockhammer of Qualcomm at the DVB World 2019, a gathering of the global DVB-based broadcaster community “With MPEG it was very clear, there were go-to people. But here, who is doing the legwork, where do I go to, who supports me in all these detailed issues? If I have a user case or a problem how do I get into this, how can I be heard? Are things controlled or is it really open?” In reply, Nathan Egge of Mozilla said that the process of becoming a member of the Alliance is very much open. “I think we have been as forthcoming as possible to bring people in who have real concerns and make the format cover as many use cases as possible, I think there is an openness to that.

“I’ve been talking to broadcasters to understand what their needs are and I think the companies we had come together work on AV1 included profiles that are important for broadcasters. So when I talk to broadcasters, those looking at upgrading to a next-gen codec, this becomes an attractive path for them, even if it’s just in the back end. “There are companies who have invested significant resources in building this and they have committed to deploy and those people will be there throughout the lifecycle of this format.” Egge noted that the large streaming players chose AV1 not because it is royalty-free but because it is technically superior to alternatives, though this is up for debate.

SYFY and Sky to debut ambitious new VR and AR experienceThe SYFY is bringing a new science fiction experience, Eleven Eleven, in late May 2019, in a project influenced by immersive theatre, using storytelling techniques that enable users to view the action from multiple perspectives.

Launching on various virtual reality platforms and AR devices, the story is set to test the consumer appetite for long-form scripted content in VR and AR environments.

Eleven Eleven will launch with three custom experiences designed for tethered VR headsets, mobile VR headsets and AR

devices. It will blend elements of theatre, gaming and cinema with immersive technologies.

Tethered VR headsets will launch on Oculus Rift, HTC Vive/ Vive Pro and PlayStation VR, with a length over 90 minutes. Mobile VR devices (70 minute duration) will launch on Oculus Go and Samsung Gear VR. The AR version (also 70 minutes long) is developed for iOS devices using ARKit.

Users will have the ability to control their experience by following a group of

interconnected characters and their stories, with over 90 minutes of content to explore.

Sky VR will partner with SYFY in Sky territories in the UK and Europe (Germany, Spain and Italy) on the release of Eleven Eleven.

04 April 2019 www.csimagazine.com

4-5_News.indd 2 19/03/2019 11:51:02

UK-based satellite and broadband provider Sky has long been one of the world’s most innovative technology companies and this track record is continuing in 2019 with the launch of the Sky X streaming service in Austria and deeper integration with Comcast under the ownership of the US cable network.

First to Sky X, an IP-delivered pay TV offering that is something of a blend of Sky Q and Now TV.

The app will be rolled out to mobile devices, Samsung and LG smart TVs, PCs, and PlayStation 4 consoles, as well as through dedicated hardware in the form of the Sky X streaming device (pictured).

The full Sky line up of live channels and sports content is included, as are on-demand series such as Game of Thrones and movies, and popular channels from the likes of Disney, Discovery and NBCUniversal. It also comes bundled with broadcast free-to-air (FTA) channels, including ORF 1 & 2 in HD, RTL and Arte HD among others.

All three options are on a monthly rolling contract with fixed rates. The packages start at EUR19.99 a month, with a EUR24.99 tier and a EUR34.99 all-in package.

Analysts expect Sky to rollout its service in other markets if it proves a success in Austria. Sky’s footprint currently covers the UK, Ireland, Austria, Germany, Italy and Spain (that is expected to grow too).

“Sky X is not just a more expensive version of Now TV, nor is Sky X a poor man’s Sky Q, it represents the fundamental cannibalisation of a company’s core business, which significant market changes have forced it to do,” noted analyst Thomas Flanagan of Rethink Faultline.

“Over two years ago, when Sky first announced its plans to take Sky Q broadband-only, we shouted from the rooftops how the move was trendsetting and fearless – a term we don’t use lightly – and still Sky remains the only obvious tier 1 pay TV operator embracing online video with such open arms. Sky, of course has the resources, reach and funding to make such dramatic changes to its business model – even prior to Comcast ownership – but increasingly those operators which don’t follow suit soon do so at their own risk,” Flanagan said.

In related news, Sky and NBCUniversal are combining their advanced advertising platforms as Comcast pursues integration opportunities now that it has acquired the UK-based company. This unification of capabilities marks the first joint advertising initiative following Comcast’s acquisition of Sky.

The new joint set of

solutions within AdSmart brings together NBCUniversal’s Audience Studio advanced targeting solutions alongside Sky’s addressable advertising tools. Sky was one of the first movers in Europe with addressable advertising with its AdSmart solution, now over 5 years old, which breaks down customer demographics such as age, postal code and viewing packages to better segment the audience for advertising purposes.

Combined with Audience Studio, it can target audiences through the widespread reach of national TV, leveraging Comcast set-top box data in the US and Sky set-top box data in the UK for a total data set of more than 50 million households.

Other initiatives are already underway too. Sky will be bringing Comcast’s own voice interface to its Sky Q box later this year, and its broadband customers in Italy will be able to use Comcast’s xFi digital dashboard to manage their home WiFi. Sky CEO Jeremy Darroch said clear benefits are flowing in both directions less than six months after his company became a part of the Comcast empire.

news in brief

LTE-over-satelliteChipmaker Sequans Communications and aerospace giant Lockheed Martin are enabling LTE end-user devices to connect directly to GEO in what they describe as being the world’s first LTE-over-satellite solution. For this purpose, Sequans engineers modified their existing LTE chips to enable a new LTE-to-satellite communication specification developed by Lockheed Martin to take advantage of the huge satellite opportunity now developing for M2M and IoT applications. Startup UbiquitiLink is also testing technology with MNOs that could connect most existing smartphones directly to a satellite.

Canal+ rethinks SVOD strategy with launch French broadcaster Canal Plus has announced a new SVoD service called Canal+Series as a replacement of its folded CanalPlay offer. Prices start at EUR6.99 per month, with tiers up to 11.99 covering four user. The subscription will offer 5,000 episodes across 150 current series, with several new productions in the pipeline. Around 90% of the content will be exclusive to Canal Plus, including deals with Showtime and FX. It comes eight months after CanalPlay shut down. Netflix, which starts at EUR9.99/month has around 5m subs in France.

Sky continues OTT-focused evolution and service integration with Comcast

News


4-5_News.indd 3 19/03/2019 11:51:18

Is Apple’s imminent launch of an Subscription Video on Demand (SVoD) service the longest mooted product launch in history? It certainly seems like it. But where there was once only talk,

the last year has seen real progress, not least in original production. Apple is now a player of note alongside its FAANG (Apple + Facebook, Amazon, Netflix, Google) peers. With just 29 shows in production or development, it’s still early days to derive too much about its long-term strategy, but clues as to where Apple will position its offer are already apparent. And positioning is key now that the SVoD and streaming TV market is a pretty crowded place.

Late market entry can sometimes be a good thing, leaving the company time to work out what works and what doesn’t in a new competitive space.

But the issue for any new SVoD or Advertising-supported Video on Demand (AVoD) entrant is two-fold: there is early evidence that the number

of streaming services households are willing to take is capping out, at least in more saturated streaming markets; and with studio majors all advanced with their own plans to launch direct-to-consumer steaming offers, content acquisition is becoming more difficult.

The content challenge is one reason that original production is the new gold

rush. Apple’s move into original production is well underway, it is ahead of both Facebook Watch and Google’s YouTube in terms of shows in development (although both Facebook and YouTube already have a decent number of originals completed, where Apple has only two). But Apple is a minnow compared to both Amazon and Netflix in terms of its forward production slate. And Amazon and Netflix benefited from being early market entrants with arguably easier access to acquired content at a time when studios and other producers were eager to license to the new buyers.

Apple’s trump card, of course, is its device ecosystem. Reports suggest Apple is likely to be creative in the way it bundles access to its new service with device ownership. That will give it a turbo boost during early launch. But content is the real key to success and sustainability.

Looking at the production and development slate for Apple suggests it has Netflix and Amazon firmly in its sights with a heavy push behind scripted content. This contrasts with both Facebook and YouTube, the content slates of which skew more heavily towards unscripted content.

Data corner

Apple TV: Peeling back strategies for a crowded marketApple’s upcoming SVOD service coincides with the streaming market on the verge of a new phase of business model diversifi cation. Guy Bisson of Ampere Analysis delves into the company’s video plans, strengths and weaknesses


6-7_DataCorner.indd 2 07/03/2019 14:30:28

Apple’s forward production genre mix also points to a challenge to Netflix and Amazon, with drama, crime, sci-fi and kids’ content heavy in its development slate, all mainstays of the scripted output of the two SVoD majors. By contrast, Facebook and YouTube have much more documentary, reality and light entertainment in development.

Early reports of Apple’s launch plans suggest the device giant has ambitions for a global launch. But to date, its production slate shows little evidence

of addressing global markets, or leveraging local content for market entry. Compared to its peers, Apple’s upcoming production is distinctly US. Only YouTube is close in terms of its home-market focus. Netflix long ago hit a 50/50 split between domestic and international original production and Amazon is already at around one third international in its forward production. Facebook Watch has made international a definite focus which, with its reality and documentary mix, positions it nicely in its own niche.

Apple’s US focus is not too surprising given it will focus on its home market for its initial launch, but it will need to ramp up international original production rapidly if it has any hope of competing with Netflix and Amazon. And when you’re playing in the high-end scripted space, that is not a cheap thing to do. Apple’s mooted $1bn content budget may not get it far.

Streaming content business model reaching its next phaseA number of trends suggest that the streaming content space is on the verge of an impending adjustment. Consumer trends that point to SVoD stacking fatigue, emerging market power balances and the rapid expansion in number of players thanks to studio and platform-led initiatives, mean a new phase of business model diversification could be on the horizon.

Subscription has been the dominant

business model to date, and Apple’s own plans suggest that subscription is key. But its own suggestion of bundling services with devices could also open the door to experimentation with ad-supported tiers and a business

model mix that Hulu has proved can work. I’d say it would be foolish to bet heavily against Apple, but it’s going to be an interesting journey given the current development of the streaming TV market.

Data corner


Sour

ce: R

ober

t Fru

ehau

f / S

hutt

erst

ock.

com

6-7_DataCorner.indd 3 07/03/2019 14:30:41

Pay-TV operators have long grappled with cord cutting but ‘cord confusion’ may be the issue that offers them a new opportunity. The huge array of TV

and video services available baffles many consumers, as they are increasingly compelled to mix and match services such as Netflix, Amazon Prime, Hulu or NOW TV and devices ranging from standard customer premise equipment to streaming players Chromecast, Roku and Apple TV.

2019 is primed for SVOD overload as Hollywood majors near the launch of their D2C services. Disney+, an international rollout for NOW TV and another from NBCU join MGM’s Epix and AT&T’s WarnerMedia-fuelled direct-to-consumer (DTC) strikes. ITV plans an online paywall for its content if a united UK broadcaster service fails to fly. Then there’s Apple’s long anticipated debut, among many others.

The dichotomy is neatly put by Dimitar Serafimov, demand generation manager at OTT solutions developer Cleeng: “The rationale behind the OTT surge is to give the freedom to consumers to pick and watch their preferred video everywhere, every time, on any device possible. At the same time there are signs that the market is moving towards consolidation mainly through a string of acquisitions and the

emergence of services like Amazon’s subscription hub.”

Jeroen Ghijsen, CEO at user experience specialist Metrological makes a similar point: “Aggregation of video content is on the rise. Historically it was the MVPDs - and they are still trying to bring as many OTT channels as possible into their video offerings - but now there’s competition from OTT streaming devices which unite a variety of OTT channels.”

Roku, Netflix, Amazon, even Sling TV are slowly becoming what optional premium channel bundles were 15 years ago, observes Tony Maroulis of Ampere Analysis. “Netflix’s list of pay-TV partners has grown rapidly, and Amazon’s Prime Video has followed suit, albeit a little bit further behind.

“Simultaneously we’ve seen a growth in SVOD-stacking [consumer taking multiple SVOD services], with some countries exhibiting very high stacking rates amongst SVOD subscribers,” says the analyst. “As many as seven in ten US SVOD respondents take more than one subscription OTT service, and with heavy content hitters planning their own DTC services this is likely to increase.”

Then consider this statement from WarnerMedia’s chief exec about balancing DTC with existing partnerships. “While going direct-to-consumer gives us an additional opportunity to reach audiences that aren’t part of a traditional subscription

service, our wholesale relationships will continue to be an important distribution channel. So, it will be a priority to work with our partners to deliver

a compelling and competitive product that will complement our wholesale distribution, allowing us to reach the largest number of viewers,” says WarnerMedia CEO John Stankey. Similar agreements are happening elsewhere, but the

relationships are becoming increasingly strained as competition intensifies.

SVOD overload will strain existing business relationships furtherThe explosive growth in TV choices has seemingly escalated consumer frustration. A Hub Entertainment Research survey of US consumers last July found consumers hungry for consolidation: looking for fewer platforms — even a single platform — that will deliver their full array of content in one place.

“The opportunity is for pay TV operators to take advantage of this ‘cord confusion’ by providing a single platform that will build loyalty among existing subscribers, bring in new consumers and even win back lapsed customers,” suggests Mark Evensen, CTO at Amino.

The research indicates the time is right: among respondents, more than twice as many would rather access all their TV and video content from a single provider (69%) than through a range of sources (31%) “making it crystal clear that consumers long for simplicity when it comes to choosing and managing their entertainment choices,” says Evensen.

Laurent Maillot, marketing manager for pay-TV operator Orange agrees: “Consumers are looking for some flexibility. They prefer to have premium content from different providers but only one stop shopping. They don’t

Cover feature

Streaming wars set to enter the next chapterAs Hollywood majors near the launch of their DTC services, more consumer TV choice than ever is offering operators a new opportunity to re-aggregate on a blended content business model, says Adrian Pennington


8-11_D-TC.indd 2 19/03/2019 11:35:09

want multiple accounts.”He adds: “We are convinced that the

next phase of pay-TV strategy globally is about aggregated SVOD.”

There are other trends too. While the number of content providers is growing to respond to a clear demand from consumers for more targeted video content, many consumers want to enjoy that on the main TV screen.

“SVOD is mainly a complement not a replacement to broadcast TV for end users,” argues Jorgen Nilsson, CEO at Zenterio. “There’s a great opportunity for operators now, and it is a must for them, to maintain the role of prime provider of content and services to the home.”

Yet OTT subscription fatigue is real. “The ‘Netflix 20 minutes’ is when you spend a long time just deciding what to watch, instead of actually watching content,” adds Nilsson. “Users need and appreciate an easy way to discover content and manage multiple subscriptions in a straightforward and consistent way.”

The argument is that consumers need an aggregator to deliver an array of targeted content and services like contextuality, second screen and voice control in a single user experience that they can access without leaving the TV.

“This will ultimately satisfy their desire for easiest access to their changing content needs,” says Ghijsen.

“Integrating apps/OTT and live TV behind one remote control can help showcase a broad range of targeted, contextual content and significantly enhance the user experience with second screen, voice control, and universal search.”

Whether from an operator or the likes of Roku, the bottom line is that consumers are willing to pay for it if they can find and purchase their preferred content with ease of use and one billing system, integrated into a single UI.

“You will see OTT look like pay-TV with prices going up and pay TV more like OTT with prices and contractual terms more like an SVOD service,” predicts Jean-Marc Racine, chief product officer at video software provider Synamedia. “From a consumer’s point-of-view a [pay TV aggregation] will be attractive because it will be less expensive to access several SVOD services plus linear programming from one platform than if you shop for the same content and subscribe to them one by one.”

The benefit for pay-TV and OTT providers, especially the smaller ones, of a unified TV hub appears mutual. “Users who want to watch shows and movies from different producers need to multiply DTC platforms which makes their bill quite expensive at the end of the month,” says Cleeng’s Alexis Gaï.

“By signing partnerships with DTC providers, operators can create hubs and packages that will make the pill easier to swallow and give easy access to all the content they need. On the other side, operators will benefit from the fact that consumers will stay locked into their environment and will be willing to pay an extra dollar for the easiness of use,” Gai.

Viacom seems to have this strategy in mind from the get-go of its first DTC subscription service for MTV. Initially available only in the UK, MTV Play is designed to reach the mobile screens of millennials and Gen Zs will provide long-form content experience in a standalone app.

“We know there is a segment of younger potential MTV fans who sit beyond the basic pay-TV bundles,” explains Arran Tindall of VIMN.

Viacom nonetheless acknowledges the importance of Sky, Virgin Media and BT and said it was speaking with telco providers and other platforms about future distribution partnerships.

Stemming churn?A combined SVOD/pay-TV offer should help stem churn on both sides. As Gaï points out, whose company specialises in subscriber retention management for OTT services, when Netflix and pay TV are packaged together, the pay TV provider reaches more viewers who want that SVOD experience. Consumers can switch from a schedule to an on-demand experience within the same UI and an operator can deliver the latest UHD (HDR) experience for integrated SVODs.

Among other factors, DTC providers can guarantee a high QoE for viewers with pay-TV carriage, no longer being reliant on an external CDN to deliver their content to viewers.

“Operators can guarantee QoE on their managed network,” says Broadpeak’s Nivedita Nouvel. “DTC content providers have the possibility of following Netflix, for example, where they would provide operators with local

Cover feature


8-11_D-TC.indd 3 19/03/2019 11:35:25

caches to boost the quality of their video content. Combining local caching with a CDN service is a win-win situation for DTC content providers and network operators.”

Historically, it was difficult for operators to setup their own back-end systems to onboard and aggregate

myriad content and SVOD services. It is becoming easier and easier as technology matures.

“With cloud-based technologies, all that’s needed is a browser and a platform that can take care of the abstraction layer between the cloud and the underlying hardware, such as a STB

or smart TV, as well as the APIs underneath it for payments, DRM, etc,” says Ghijsen. “Similarly, where custom, native implementations used to be required, while the resources on the hardware were limited, now we are seeing HTML5-based cloud implementations.

Nonetheless, there will be pressure on aggregators to keep pace with every new content provider and integrate or onboard them in an aggregated offer. This creates business and technical challenges - and opportunities for vendors.

Zenterio, for example, offers its Zenterio Cloud and Zenterio TV solutions for premium OTT app integration, working with APIs to deliver unified search and discovery.

“The operator needs to find ways to make all content available easily and give the user the possibility to browse different content providers seamlessly,” says Nilsson. “This is mainly around how the operator can control the UI/UX to maximise content consumption, but there is a technical aspect as well to harmonise the metadata available from the different content providers.”

Bringing IP and broadcast, together, architecturally, is what Synamedia call Blended TV. “Whether the broadcast network is digital terrestrial, QAM, IP multicast or DTH, operators require next-gen platforms to aggregate services with unified metadata and content discovery,” says Racine.

Unified searchUnified-search is considered a key component of the additional value provided by aggregators. Should studios choose to keep all their content within their proprietary ‘walled gardens’, the value of a content aggregator will increase, reckons Ampere’s Maroulis, “assuming the aggregator can provide a greater degree of convenience for the customer.”

Since there’s no difference in the consumer’s mind about where content comes from, “the ability to aggregate

Cover feature


Media majors launching DTC/VOD services

Company/service Notes

Disney+ The big one. Disney will rely on exclusivity and originals to help attract and keep customers when it launches end-2019. Some estimate that Disney+ could attract up to 2.5m users in the US by the end of its fi rst year. Can Disney+ give Netfl ix a run for its money, in the US and internationally?

NBCUniversal Coming in early 2020, the company’s as-yet-unnamed service will employ a freemium model comprising a free AVOD part alongside an ad-free, subscription-based tier.

NOW TV Under Comcast’s ownership of Sky, there are plans to take it global in 2020, building on a strong brand recognition in the UK and Europe, though not so much elsewhere.

WarnerMedia Arriving later in 2019, it is expected to be one of the new leaders to emerge from the battle for the US market, under its new AT&T-defi ned, free-mium strategy. Rethink Research estimates it reaching 29.6 million SVoD homes by 2023.

Epix Now The MGM-owned service launched this February, priced at $5.99 per month, offering a library of MGM blockbuster fi lms, four live EPIX channels, and original content, some in 4K HDR streaming. Available stand-alone and via payTV.

T-Mobile (US) Original plan was to launch a disruptive OTT TV service nationwide, but gave up due to the huge saturation now experienced in this space, choos-ing instead to become a DTC aggregator like Ama-zon Prime Channels. “Reality topped rhetoric,” as analysts at TDG put it.

ITV Hub+ ITV’s premium online subscription service. For £3.99 a month, users get all the features of the original ITV Hub, but with ad-free content and downloads. Saw strong growth in 2018.

Apple The long, long debut is almost here. Late is not always bad, but it doesn’t suit Apple and might be a sign that it is stuck for ideas to innovate in this space. Much more analysis on pages 6-7

8-11_D-TC.indd 4 19/03/2019 11:35:37

metadata and go from search to directly viewing the content is essential,” stresses Racine.

What’s important is for operators to have the ability to react to consumers’ personalised, shifting content appetites — whether the consumer desires niche/localised programming or international content — with the tools in place to track and measure real-time content consumption, and modifying the content without disruption to the network.

“Simply put, related to the user’s personal preferences, if an application isn’t being used, then it should be removed,” states Ghijsen. “If an application is being used often, it should be integrated into the EPG, so it appears as if it was just a broadcast channel.”

“The key for the operator is to remain the main point-of-consumption for users and make sure users can find all content they love in one platform rather than many different platforms,” says Nilsson. “This is the differentiator for operators compared to DTC platforms.”

Content mixStriking deals for the right mix of SVOD services is another factor. Customisation and personalisation is key here as there’s no one-size-fits-all approach. Says Nilsson, “Different users will need a different mix in their UX so it is crucial that the operator can aggregate as much content as possible and be flexible in deciding how to present that. This way, every user can get the optimal mix over time.”

There are ‘must have’ services [like Netflix] which consumers expect to get access to, reckons Racine. “Don’t forget the importance of local channels either. FTA content still attracts the lion’s share of audience. The combination of these with catch-up and SVOD is where pay-TV can differ.”

“Linear and local TV is key,” agrees Orange’s Maillot. “Local content has a good ranking… we will keep on distributing FTA channels. Even if the

audience is increasingly fragmented and weakening every year, linear remains mass consumption and enables us to keep control of the TV. Our battle is to keep the user on our side.”

Maillot says Orange’s target is to offer the maximum available content. “There are some services which are a no-brainer, such as Netflix. With others there are trade-offs to be done. Amazon is a competitor on the aggregation side and that’s an issue for us. We don’t deal with them now…. maybe in future.”

There will be pressure for a clearer distinction between content providers and aggregators. “Hybrid models — where the same company is both content provider and aggregator — can work for very big players, but has its challenges,” outlines Nilsson.

“They will compete with their own distribution channels, because a user could find the same content in various places, at different prices. Another challenge is that there will always be some piece of content that users want but that is not available on a DTC store. It seems more likely that pure-aggregators have a better chance to provide a complete content offer to end users, and simplifying the discovery and access to it.”

The consumer wins… or do they?Inevitably, there will be casualties. Many SVOD-only services (ad-funded, transaction- or subs-based) simply won’t survive the aggressive content origination strategy of Netflix or Disney. There has already been a lot of consolidation and quite a few failures in the SVOD-only DTC space (Turner’s FilmStruck and the original CanalPlay). Even Amazon is looking to cull some of its SVOD Channels to focus more on exclusivity.

“Media giants with top-notch content (HBO, Disney) won’t license to the likes of Netflix which will in turn strip SVODs of a chunk of their subscriber base,” says Cleeng’s Serafimov. “That will increase competition among the best and will raise the overall quality of the service offered. That is always good news for the market and the consumers.”

“The winner will be the one that makes life easiest for the end users by removing as many friction points as possible,” says Zenterio’s Tom Keaveney.

Because, ultimately, while the onset of broadband, together with SVOD, skinny bundles and DTC platforms, has resulted in more choice, convenience and flexibility for users, it has also created something of a mess for both the industry and consumer in terms of app and platform fragmentation. Now - as OTT matures and the streaming industry enters a new phase in its evolution - the question is how the future of TV will be about deploying new integrated experiences that are truly personalised and contextual while keeping in mind some of these challenges from both perspectives.

Cover feature


8-11_D-TC.indd 5 21/03/2019 11:08:15

Here is a scenario that only OTT subscribers are familiar with: you are seating in your apartment and watching a dramatic tight playoff game of

your favourite team. Maybe even it’s the final itself. There is 1 minute to go and you are sitting at the edge of your seat when suddenly you hear hysteric screams of joy coming from your neighbours.

You immediately realise that there was a goal/buzzer-beater event, that your OTT service will be kind enough to show it to you in 60 seconds or so. This 1-minute-long limbo-zone, of knowing that your present is other people’s past, generates a variety of passing thoughts in your mind, starting from the obvious “which team actually won?” and “who scored?”, moving to frustration such as “when do OTT services will finally provide a real-time experience for their viewers?” and ending in existentialistic insights such as “if a goal was scored without me seeing it – did it really happen?”

Low Latency CDN MechanismMoving Back to TS over UDP. It is a well-known convention that in order to reduce the streaming latency, chunked HTTP protocols should to be discarded,

and the adaptive bitrate concept should be implemented on UDP and not on TCP. Therefore, It is critical to develop an streaming mechanism and protocol which handles MPEG-TS over UDP in combination with WebRTC and CMAF at the last mile. The protocol is enforced by an advanced CDN technology and player’s SDK.

Internal CDN Distribution. The CDN architecture is unique and innovative, as it introduces some patent pending concepts for providing best streaming serving decisions and performance AI algorithms (which will not be elaborated in this article).

In this low latency implementation, the use of ABR transport stream discards the need for a packager. The CDN’s Publish-Server gets from the transcoder couple of TS in different bitrates and profiles. It is then pushes these streams to the appropriate edge servers according to Vimmi’s special internal distribution algorithms, based on consumption patterns, end-users’ session management and overall system load. This CDN ensures a sufficient bandwidth between the CDN POPs and the publishing point for accommodating the sum of all profile levels of all channels.

Detecting Bandwidth Fluctuation and Switching Between Bitrates. The low-latency CDN mechanism treats a live playback request as any standard playback request: the CDN picks the best Edge Server to serve the specific request (based on Vimmi’s CDN grid-computing algorithm), it then generates and sends a token to the requesting client which relates to chosen Edge Server, the client and the request time. Edge server support three protocol to

delivery to client: MPEG-TS over UDP, WebRTC and CMAF. The client then generates a request (with the token) to start the stream from the said edge server. After the token is authenticated the edge server starts streaming the content to the requesting client, using MPEG-TS over UDP.

Once streaming has started, this low-latency CDN mechanism includes special data exchange between the CDN edge server and the served player, from which the edge server can conclude the current quality of the line and manage individual sessions.

This mechanism of constant monitoring, profiles switching and stream concatenation, implements a pure adaptive bitrate policy while maintaining a continuous connection to the client. The result is an optimal quality of experience for the viewer, both in terms of video quality, RTT and, most importantly, streaming latency - with an average of less than 1-second from the actual source feed in any unmanaged OTT network.

ConclusionAs the popularity of real-time live content, such as sport, reality voting and betting, is sky-rocking, and as the network’s infrastructure, capacity and quality are constantly improving, the common chunked HTTP protocols have become insufficient and the need for low latency streaming solutions has become a necessity.

Low Latency CDN is an inevitable evolution of the OTT streaming industry, and it is just a matter of few years until it will become a mandatory requirement for any content service provider which includes live content. Vimmi can be found at NAB Show 2019 on SU8622

Hearing future voices from the past“If a goal was scored without me seeing it - did it really happen?”

Low latency


Eitan Koter is Co-founder & Co-CEO, Vimmi Communications

12_Vimmi-v2.indd 1 21/03/2019 11:09:01

Sponsored by:

ww

w.c

sim

agaz

ine.

com

Looking at how ML is being used in video content

Machine Learning in broadcast

13-17_TiVo.indd 1 07/03/2019 14:25:07

AI and its subsidiary machine learning (ML) entered broadcasting well over a decade ago and have now infiltrated just

about every aspects of the video lifecycle and even started to determine what content is made in the first place. There is no single killer application, with benefits emerging more through coordinated application of different algorithms and methods across workflows and production lines in a case of the whole being greater than the sum of the parts.

We see this in metadata creation where AI tools are cutting costs and speeding up the process through automation, which in turn improves the viewer experience by enriching search and discovery. Users are more likely to find or be recommended something they really want to watch, possibly content they would not have searched for directly. ML can be employed to home in on the most effective mix between more obvious recommendations (for instance, based on known likes) and more intuitive suggestions, akin to thinking outside the box.

It is also true that some AI based ideas, such as artificial news anchors, might at first seem like gimmicks, but turn out to illuminate aspects of potential value, or give pointers to the future. That was the case in the UK when the BBC staged an experimental programme in May 2018 called Made By Machine: When AI Met The

Archive, which applied AI algorithms to dredge the broadcaster’s huge archive comprising 270,000 hours of video to identify content meeting the tone and style of its BBC 4 channel. This led to 150 programmes being selected, rather less than 0.1% of the amount analysed, which human observers agreed seemed very much in keeping with the spirit of the channel, which is focused on emerging arts, music and culture with a fair dose of comedy.

This was a once off exercise whose object was not to deploy AI on its own for programme selection but to discover how it could assist human editors to automate the effort of pre-selection from a vast archive. Two lessons were learnt, firstly that AI could provide valuable assistance, according to BBC Four channel editor Cassian Harrison, one of two architects of the programme. “AI can effectively analyse video content and edit and build its own image sequences”, he says.

The second lesson was that AI in this context was a powerful tool only when exercised judiciously by humans and would be useless on its own. “With AI, you have to build it with humans,” says BBC Research & Development’s head of internet research and future services, George Wright, the other architect of the programme. “Without human involvement, you would deliver quite a poor output.”

This is certainly true of such advanced applications where the algorithms are making the sort of decisions previously confined to humans, involving judgements with a subjective component through fuzzy combinations of attributes. Humans are

needed to act as conductors, directing the process, ensuring the algorithms deliver the results they would have achieved themselves given the time.

Along similar lines, some movie studios have been experimenting with use of AI techniques to automate production of trailers to generate as much suspense and

anticipation as possible without giving away the plot. IBM Watson Media has been working with studios on movie trailer creation, where the challenge is in conforming with particular requirements for context and tone, according to David Kulczar, the company’s Senior Offering Manager.

“IBM Watson created a movie trailer for the thriller “Morgan,” says Kulczar “To make the trailer, we essentially sent Watson to film school to train it extensively on what elements make a compelling horror teaser. Watson ran visual, audio, and contextual analysis on the film, to determine which scenes were suspenseful, and arrange visuals accordingly. The process to create the trailer took about 24 hours to complete — a significant reduction from what typically takes around 10-30 days.”

He admits more work was needed before video producers can adopt such technology at critical mass, especially to develop clear frameworks for training AI and machine learning modules to meet the needs of production teams. “In order to power that training, modules will require troves of training data to better understand everything from unique sounds to identifying exact objects or faces within a video.”

Such advanced examples do exemplify how AI will progress in established domains of video creation, production and delivery, according to Aditya Ganjam at Conviva. A major driver he reckons will be the ability of AI and ML to uncover more signals, in less time, than other techniques.

“Without AI and ML techniques, providers must spend a considerable amount of time and effort to review

Machine Learning

AI touches all parts of the video lifecyclePhilip Hunter examines some use cases of Machine Learning in broadcast


13-17_TiVo.indd 2 25/03/2019 12:32:57

data or configure and monitor manual alerts in order to predict successes or detect issues with the streaming experience,” says Ganjam.

“With AI and ML techniques, this process can be automated to proactively predict and discover, and reactively perform root-cause analysis. To consolidate and correlate across devices, channels, social media requires scale and real-time responsiveness of an AI or ML engine,” he says.

Such AI based correlation across data sources enables analytics firms such as Conviva to address the three fundamental issues of prediction, detection, and diagnosis in various contexts. “With detection, we are looking at identifying events of importance, in real-time, to determine factors that can lead to a poor user experience when streaming. Finally, we’re utilising diagnosis to analyse all the hundreds of millions of possible permutations for root cause of deterioration in quality in order to recommend action a publisher can take to improve the user experience.”

When looking at the implications of AI and ML for content, the main goal would be to understand how data can be used to decide what content to license or create, or at an even more granular level, how to tailor content to be most impactful with the target audience.

On the business operations side, AI and ML can inform investment

decisions. Conviva has seen that a single pilot often provides the detailed engagement data indicative of the show’s future success. Real-time analysis at the pilot phase of publication also gives creators the opportunity to edit or develop subsequent episodes based on those learnings.

This can then guide ongoing development of the content where that is under the service provider’s control, as well as how to promote it and target individual customers. “AI and ML techniques can be used to tie a customer to the content that would most resonate for them through deep analysis of their data, and the data of other viewers like them. The ideal is a closed loop where providers are both promoting and delivering content that excites their audience and ensuring that the experience of that content is at the level of quality the creator envisioned, all in real time.”

AI and the future of contentWhile being an opportunity, AI also imposes requirements that must be fulfilled in order to remain competitive and this applies not just to service providers or distributors but also content producers. In fact, it especially holds for content producers because unless their assets are “AI ready” in the sense of having the fine-grained metadata that makes them discoverable by contemporary search and recommendation algorithms, their

viewership will drop dramatically. This is the case for content producers at all scales, from casual uploading to YouTube through to major distribution deals with Netflix. In fact, the big SVoD players like Netflix and Amazon Prime insist their content suppliers label their content in very fine detail and that can only be done efficiently at any scale with the help or AI tools.

Algorithms similar to those used for metadata creation are also being used to generate captioning and subtitling which have traditionally been time consuming manually intensive processes. AI algorithms are automating language analysis, translation and speech-to-text, which in turn enables generation of subtitles and captions in multiple languages. AI tools can convert audio to text, so that the words can be delivered aligned with the video segments.

Such audio to text conversion also helps with content moderation, which has been a requirement for years but has become a rapidly growing burden given recent proliferation of both content and compliance rules, which vary significantly between countries. More advanced AI tools can now augment text analysis with sentiment, image and object recognition, allowing editors to detect more easily not just offensive language but also adult content deemed unsuitable for a given audience or window.

Another area where AI has been making an impact is colour matching during the editing process. This example again highlights how AI and ML work best in combination with human experts. As IBM’s Kulczar stresses, there is the prospect of AI and ML having a growing impact on the nature of content itself. “Looking ahead, AI and machine learning will generate new, realistic audio and visual content: think, voice overs, splicing together manufactured shots from pre-existing visuals, and automatically sequencing shots into new scenes,” Kulcar says.

Machine Learning


13-17_TiVo.indd 3 25/03/2019 12:33:11

Machine Learning


The terms Machine Learning (ML) and Artificial Intelligence (AI) are quickly becoming two of the most talked-about terms in the

technology sector, but what do they really mean for media? How are media companies like TiVo using ML and AI to make their businesses and the businesses of their customers more efficient? How do these technologies benefit end-consumers especially as they sift through the ever more-complicated mix of content choices and content platforms available to them?

This article explores those questions as well as how TiVo has invested in ML and AI throughout its workflows to help super-charge content discovery.

Content ingest and streamlining the processContent ingest – or the process where metadata aggregators like TiVo bring rich media into their systems – is the starting point for preparing the content needed for discovery engines. However, it’s become increasingly complicated as the needs for more and more comprehensive information has grown. TiVo processes information from hundreds of sources every day but sources have different ways of collecting content information and different formats, so it becomes difficult to ensure content sources align. Only recently could each content item be

automatically imported, and even then, a process of random checks meant that

incorrect or incomplete content could slip through the net.

In order to combat these challenges, TiVo uses advanced machine learning algorithms to assess the content that is imported from multiple sources and assign a confidence level to the content. Only if it falls below a defined level of confidence does a human editor need to get involved to check it. Content above the confidence level is able to be automatically processed and ingested into the content database.

The Metadata Iceberg – building out the detailThe content database is now bigger than ever before and requires a level of detail that many consumers will never be aware of. The days of time, title, cast and where content is playing is no longer sufficient to meet the needs of the industry. A paradigm we use at TiVo for our content is ‘the metadata iceberg’. The basic information is above the waterline for everyone to see in the discovery experience but below the waterline there is more information than you expect as themes, tones, product links, soundtracks, ratings, reviews and many other categories help to build out a resource that is fit for purpose in today’s discovery systems.

Making metadata voice-readyOne of those discovery systems is “voice-led discovery’ – an area where TiVo’s Conversation product is a world leader. With voice-led discovery, consumers are able to find content

using natural language; by asking questions and providing a set of search parameters that doesn’t necessarily include a title or actor’s name. Conversational services not only dramatically increase the number of searches a viewer instigates, it also broadens the type of search. With text search, the

viewer will typically search twice a month, using show titles. With Conversation, our research has shown this number jumps to an average of 26 queries and the type of search becomes more natural. Try typing ‘What is that movie with Peggy from Eastenders and Kenneth Williams?’ using text, and it quickly becomes clear why the natural, conversational search is changing how viewers access content.

Voice solutions rely almost exclusively on the data in the underwater part of the iceberg along with the weighted connections that form the basis of the TiVo ‘knowledge graph’, or proprietary machine learning content technology. It is this graph technology that is constantly looking for more information, understanding the context of what it finds and is continuously refreshing the content that forms one of the core areas that Machine Learning is making a difference for the consumer. Without the sophisticated algorithms that seek, sort, make sense and apply structure to the vast oceans of unstructured content we wouldn’t be able to deliver on the promise that consumers can ask questions such as ‘What is the film where Olivia Colman won an Oscar?’ or ‘What was the Friends episode at the beach?’ or ‘Show me some TV programmes filmed near me’, etc.

As part of any Conversation integration project, TiVo asks its partners for entertainment sources which are most relevant to their services and customers. By constantly updating these sources, this ensures that the tone and content of the service is maintained and enables natural, conversational content discovery.

How Machine Learning is super-charging content discovery for consumers By TiVo’s Charles Dawes

13-17_TiVo.indd 4 07/03/2019 14:25:56

There’s more to it than collaborative filtersNow let’s talk recommendations. We’ve become accustomed, as television viewers, to have a user experience show us new things we might want to watch. When it comes to content recommendations, because of the myriad of choices out there today, there’s a need to go far beyond the basic collaborative filters that show us ‘you watched X, people who also watched X also watched Y so you should watch Y’. There are many personal reasons we like to watch content and the algorithms need to be equally diverse. These sophisticated algorithms need to be powered by machine learning as the viewer progresses through the content discovery timeline. The systems will gain more knowledge about a viewer and tailor the content discovery journey to them individually. It’s this powerful personalisation that has showcased real results; users are eighty times more likely to click on

personalised content, twenty-one times more likely to view that content which will have a total lift of fourteen percent more unique viewers and a 129% lift in views overall.

The game is running late – what can we do?In the past, a major problem that would frustrate viewers on an individual level was changes in broadcast schedules that weren’t reflected in their personal recordings on their video recorder or DVR. More recently, this has become a much wider problem as service providers begin to implement start-over and network-based PVR solutions that rely on accurately being able to start and stop recordings at the time when they actually begin and end. In order to implement this at the speed and scale that a multiple channel service requires, TiVo has been working with partners to implement machine learning algorithms that are constantly looking at the content streams to identify the accurate

second-by-second start times of programs so recordings can be corrected in near real-time. This enables a seamless customer experience where they can replay a programme from the start and be assured that they won’t have to watch minutes of unnecessary content – or miss any of the content they love.

Closing the feedback loopNone of these improvements and advancements in content discovery would make sense without the ability to learn from the

consumers’ interactions as they find and enjoy their content. TiVo applies advanced machine learning to the return path data we receive in order to derive business insights that can help service providers and content producers understand how their services are being used. From being able to introduce a new sense of serendipity into the recommendations world through to helping get the right advertisements in front of the right consumers, machine learning and AI will continue to have a profound effect on how entertainment technology companies like TiVo do business. It will also continuously change how consumers find the right piece of content at the right time as the industry drives to create more and more great television for us all to consume.

Charles Dawes is senior director, international marketing, at TiVo

Machine Learning


13-17_TiVo.indd 5 07/03/2019 14:26:19

Drive engagement and loyalty.

For more information:business.tivo.com

As the expert in personalised video

discovery, TiVo can help you deliver

the engaging features and services that

drive customer loyalty. Our innovations

in entertainment metadata, search,

recommendations and conversation

are powered by machine learning and

provide the flexibility to add next-gen

services and grow your business.

Join the world’s largest, most

recognisable brands who turn to

TiVo for the ultimate entertainment

experience.

Untitled-1 1 04/03/2019 10:36:05

In February 2019, a development emerged that could have implications for the media & entertainment industry in Europe and globally. This happened when beIN Sports threw down the

gauntlet to media rights holders, choosing not to bid for and renew the Formula One television rights it held in the Middle East, due to the rampant piracy in the region. In a clear message, while most of the blame was aimed at beoutQ, beIN said it would pay less for rights that cannot be protected and that content owners should do more in the fight against piracy.

The call to arms can be seen as a wake-up call to the broadcast industry as a whole, which has arguably long had its head buried in the sand when it comes to new forms of piracy, choosing to seemingly ignore the lessons from music and how they could apply to video.

For a little context, consumers are trying to circumvent geographic restrictions and access video content from all over the world, with or without

content owners’ consent. Tools such as Virtual Private Networks (VPNs), Domain Name Systems (DNS), Tor, and proxy servers are helping drive a growth in the illegal viewing of geographically restricted content, posing a significant challenge and escalating to the media industry.

Whether it is the widespread theft of content as beoutQ is accused of perpetrating, or doing more to prevent content being geo-pirated and leaked by the widespread use of VPNs and DNS Proxies, it is arguably up to the entire value chain to work together more closely to tackle the problem. After all, payTV companies, sports league, big Hollywood studios, premium OTT providers and others with high value content all have a ‘stake’ in how that content is viewed, which needs to be protected from geo-piracy and geolocation-fraud. Ultimately, the entire business model relies on it.

Research by Global Web Index suggests that one in four Internet users use a VPN service on a regular basis and mostly to access restricted content, much of it live sports (although it is also known that Internet users in China are turning to VPNs to access social networking sites). Sports rights deals are in most cases handed out on a territorial basis, localised in a similar way to films, so live OTT sports services are employing technology to fight geo-piracy and tackle the unlawful use of VPNs.

The use of VPNs to bypass territorial controls and avoid geoblocking really first came to light as Netflix started to become more popular, but was yet to become a global OTT service. It was estimated that Netflix was being accessed by tens of millions of people in countries where it had no operation,

including Australia and New Zealand. Netflix established a team dedicated to combating VPN access to its content and others have followed its lead, with the likes of Hulu employing VPN detection to enforce geoblocking.

As the issues have become visible and the need to act more pressing, specialist vendors like Digital Element have come to the fore. Digital Element tackles geo-piracy through its NetAcuity geolocation solution, a proprietary IP geolocation which it claims allows customers to pinpoint users’ geographic location from just their IP addresses, as well as uncover numerous other actionable insights such as demographic data, connection types and proxies. The company works with brands like ESPN, Twitter, Discovery, Yahoo, VUBIQUITY and Disney Interactive technology, using what it calls ‘IP intelligence’ to “harness the power of location” to understand and connect with online audiences.

Doing so helps video distribution companies use effective technology to understand where their viewers are accessing content from and identify those who are falsifying their locations, to ensure content suppliers’ regulations are adhered to. This has risen to prominence as OTT proliferates and the migration to SVOD, direct-to-consumer streaming and other forms of digital viewing continues to rise. The aim is to help media companies balance the need between compliance with geographical rights and provisioning a better streaming service, using it as a power for good.

This balance is also reflected in the recent change to the law by the European Union that will see online platforms like YouTube and Twitter be forced to remove copyrighted material deemed to breach IP property. Overall, the EU is still keen to further its ambitions towards a common digital market, ensuring portability of content will promote legal access, but mindful of the territorial restrictions created by licensing schemes.

VPNs, geolocation and OTT video servicesTackling geo-piracy with IP intelligence

Security


19_DigitalElementByline.indd 2 20/03/2019 14:54:19

Whatever Netflix does is tracked very closely by the industry, so the

company’s recent foray into branching narratives has reignited an old debate surrounding this form of interactive TV. Will viewers embrace it, or do shows like Bandersnatch have limited appeal?

The interactive movie format was first tried in 1967 with Czech-produced Kinoautomat. The BBC experimented with a technology in the early 1970s that later developed into teletext, and then later in the decade the world’s first commercial interactive TV service launched in Ohio, USA.

The ‘Black Mirror: Bandersnatch’ episode was released in December 2018 by Netflix to its members worldwide, giving users the option to influence the story and in the process work their way to up to 12 different endings.

Depending on the actions the viewer took, viewing time could be anywhere between 40 minutes and over two hours. Viewers were presented with a series of A or B choices, from the small decisions that had no impact on the story (such as what cereal to have for breakfast) to more complicated dilemmas that influenced the fate of the main protagonist.

One argument in favour is that modern technology and the power of OTT stands to unleash the

opportunities of this kind of viewing experience as online video services enable interactivity and personalisation.

Charlie Brooker, the show’s creator, initially used Twine, an open-source tool specially developed for telling interactive non-linear stories, which was eventually replaced by Netflix’s own in-house tool. Having developed its own Branch Manager software, Netflix will continue to explore the possibilities further. Indeed, VP of product, Todd Yellin, promised more such programmes at a Mumbai this month: “It’s a huge hit here in India, it’s a huge hit around the world, and we realised, wow, interactive storytelling is something we want to bet more on. We’re doubling down on that. So expect over the next year or two to see more interactive storytelling. And it won’t necessarily be science fiction. It could be a wacky comedy. It could be a romance, where the audience gets to choose – should she go out with him?”

Given Netflix’s love of data, it will be interesting to see how the feedback affects any future experiments with interactive TV.

It goes at odds with the phrase “Netflix and chill” that has made its way into popular culture, with some users complaining that

they found making some of the choices in a limited amount of time (less than 10 seconds) too stressful.

Given the costs and complexity of producing and creating these types of stories, on top of most users’ preference for lean-back experiences, it is difficult to see them becoming mainstream, at least for the foreseeable future. “It certainly takes a lot of careful planning to produce a satisfactory experience, so it is a creative challenge as much as a question of cost and complexity. The main problem is perhaps the need for viewers to make conscious choices through the course of the narrative. For some that may be an attraction, for others it may be an irritation,” says Dr William Cooper of informitv.

Then again, computer games are borrowing heavily from film these days so why not have TV take some elements from that medium if the availability of choice means those viewers who want it are left more immersed and engaged? It also encourages repeat viewings...

“We have yet to see a defining success for this genre but I suspect there will be further innovation in this area. There are endless possibilities for narrative development,” surmises Dr Cooper.

In the end, content will always be king and if programmers like Netflix can pitch these shows as a compelling story rather than just a gimmick, then branching narratives could evolve into a viable TV phenomenon.

The future path to success, then, is very much open.

Interactive TV

Interactive TV and branching narratives: which way will they go?CSI asks if Netfl ix’s Bandersnatch will kick-start a new wave of interactive storytelling in TV


20_InteractiveTV.indd 2 14/03/2019 16:09:05

Sponsored by:

ww

w.c

sim

agaz

ine.

com

Improving the Quality of Experience

Live sports streaming

21-25_VNova.indd 1 14/03/2019 12:15:38

It’s easy to forget how quickly things change in this industry. For example, streamed VoD services have only been a mainstream proposition for a few years, even if our children now regard buffering

on Netflix as an infringement of their human rights. It’s also not that long since anyone who actively chose to watch an online stream of a live sports event would have been regarded as an oddball. Today, OTT streaming of live sports events is firmly established – in future it may well be an important influence on industry fortunes.

“Sports has always helped to drive subscriptions and revenue for media services: in radio, broadcast TV, satellite TV, cable TV,” says Lee Berke, CEO of US-based consultancy LHB Sports, Entertainment & Media. “The same is going to happen online.”

OTT streaming of all sports is growing in practically every market, according to Ian Munford, director of product marketing at Akamai. “The biggest video delivery records that we’ve hit are all to do with live sport,” he says. He contrasts average peak numbers of concurrent streams for matches in the 2014 FIFA World Cup – 5 million – with the equivalent figure in 2018: 9.7 million. “What we see as a peak in a given year then becomes the norm the following year. The viewing audience is transitioning to an online world.”

The push online may also be accelerated by the rapidly growing popularity of esports. Respondents to PwC’s 2018 Sports Survey say esports now has a greater potential to grow revenues globally than any other sport, toppling soccer from its ascendancy.

PwC estimates that media rights revenues for esports grew by 49.6 per cent between 2017 and 2018; and that the esports economy will double in size over the next four years.

The dynamics are different in every market, but pay TV providers in particular are coming under increasing pressure to create some form of sports streaming service. “I think most traditional players are developing some kind of online offering, partly because they’re under an obligation to show they are trying to reach audiences online,” says Tim Westcott at IHS Markit. “But I think you have to be careful about talking about this as a revolution. More of us are using mobile devices, more of us have moved online, but the TV still has a very important role in sport.”

Munford agrees, highlighting an increase in the average bit rate used by fans streaming World Cup matches in 2014 and those doing so in 2018, from 1.65Mbps to 2.74Mbps, which he says indicates a big increase in the numbers of viewers streaming to big screens.

But enabling a large audience to stream to big screens is both technically demanding and – perhaps more significantly – expensive. Many

broadcasters find that very off-putting when audiences for broadcast sport services are still very large. “There’s been a period when what’s been technically possible hasn’t been commercially possible,” says Westcott. Meanwhile, the number of

Direct to Consumer (DTC) streaming services is also rising, as more sports brands, including some of the most popular sports and teams, but also less mainstream sports, create their own OTT services. Notable launches during 2018 included F1 TV and Juventus TV. Westcott acknowledges that the number of these services is rising, but suggests that major sports are likely to keep using the same linear model as long as it is financially worth their while. “When their TV partners are offering them less money, that’s when they might think it’s worth making a change,” he says. That could happen. “Some broadcasters are looking at their catalogues and deciding that some sports are becoming unaffordable,” Westcott continues. “We are going to see some sports rights holders failing to obtain the revenues from operators they would like. Maybe they will decide there is a big enough online audience for them to change the way they license their rights.” Some sports may strike deals with major technology firms seeking to acquire a slice of the sports streaming market. The 2018 deal struck between

Sports streaming

A question of OTT sportStreaming of all sports is growing exponentially and this presents unique technical challenges, says David Adams


21-25_VNova.indd 2 14/03/2019 12:15:55

Amazon and the EPL to acquire the rights to 20 Premier League matches over two match days per season between 2019 and 2022, may one day be seen as the moment when Amazon started to turn itself into a significant player in sports. Since then it has created partnerships with other sports streaming services, including Eleven. Other notable recent developments have included YouTube acquiring the right to stream a range of major sports in the US and other markets; and both Facebook and DAZN making inroads into rights markets for cricket and football. StreamNet TV is preparing to launch a new international OTT service incorporating live sports and esports.

Service quality and latencyBut live sports streamed OTT will only acquire and retain a mass audience if those services match or exceed the performance of broadcast services in terms of reliability, Quality of Experience (QoE) and latency. Cloud-based technology in particular means that operators and sports organisations are able to launch new OTT services directly targeting the consumer at a much greater speed and a much lower cost. The cloud’s multiple

advantages include things like scalability, the ability to spin servers up or power them down to allow for audience peaks and troughs. “Quality is important and latency is really important,” says Guido Meardi, CEO and co-founder of compression specialist V-Nova. “A few seconds matter: the emotional charge from a bad experience is greater, if the game is happening right now and you cannot watch it.” The trouble is, there are a lot of things that can go wrong. “Some may be beyond the control of the service provider – the state of a broadband connection in the home, or slow clearance for a credit card payment, for example – but the provider gets the blame,” says Westcott. Everybody hears about it, because people complain on social media.” Technologies that might help a provider avoid this scenario include V-Nova’s PERSEUS Plus compression technology. It uses a unique hierarchical encoding approach to enable better compression on top of existing codecs like H.264 and HEVC, alongside reductions in the processing power required during encoding. By leveraging existing codecs and working within exiting streaming formats like HLS and MPEG-DASH it also maintains compatibility with existing devices. V-Nova says operators have used the technology to provide UHD and HD video at bandwidth speeds that deliver a good QoE even for viewers using devices relying on lower connectivity services like public Wi-Fi. It claims use of the technology can reduce storage and CDN delivery costs by up to 80 per cent. Open source and cloud technologies help V-Nova and other vendors provide flexible solutions that can scale up as audiences grow. A good CDN service is also essential, as is flexible backend technology, says Globecast’s Frederic Torasso. But arguably the most important thing any service can do is reduce

latency. No viewer wants to hear their neighbours cheering while on their own screen a footballer is preparing to take a penalty; or to see social media posts referring to events they have not yet seen on the screen. Latency can reach 50 seconds or more on an OTT stream. It is caused in part by legacy features in streaming formats, which use memory buffers within the workflow to reduce buffering within the player in a consumer device. With three or more chunks of six to ten seconds of content buffered in some formats, that delay stacks up quickly. Any attempt to reduce that latency ac acelerates the workflow, so risks an adverse impact on QoE. But there are ways to solve the problem. One is to use a multicast/ABR solution, featuring a unicast to multicast transcaster in the headend, with multicast then converted back to unicast equipment in or before the consumer device. Use of CMAF reduces the chunks of content in the workflow. Torasso claims that combining ABR and CMAF can reduce latency to between five and seven seconds. Akamai’s Adaptive Media Delivery solution uses ABR technology that can support HLS and MPEG-DASH, alongside CMAF. Its liveOrigin technology is designed specifically for live OTT content. Broadpeak offers a nanoCDN multicast ABR solution to achieve the same aim. “It allows you to handle unpredictable peaks in demand,” says Nivedita Nouvel at Broadpeak.

Finally, there are other technology factors that may also help shape the future of live sports streaming, such as 5G, which European operators are planning to trial for events like Euro 2020. Globecast suggests that major technology companies that buy sports rights may use VR to differentiate their offerings (StreamNet TV is already promising this).

All these technologies are impressive. But ultimately it may be the rights holders who determine the future of streaming live sports.

Sports streaming


21-25_VNova.indd 3 14/03/2019 12:16:08

Sports streaming


The growth of video traffic is dizzying, with 45% of people watching OTT video for over an hour per day. There is always talk of Netflix, but

most of the ‘valuable’ video content, such as sports, esports and news will remain, by definition, live. Major sports/esports federations and rights holders are increasingly going direct-to-consumer in a bid to challenge the established status quo of Pay TV operators and broadcasters. Indeed, for many sports federations or rights distributors, the tipping point has arrived where it is often more lucrative to monetise their content rights directly rather than licensing them to operators. In theory the internet allows you to distribute very high-quality video (even UHD HDR or 1080p50 HDR) to all devices. However, live distribution over the internet doesn’t compare to the quality of experience provided by traditional broadcasters. It also doesn’t deliver the same quality of service as VoD streaming services like Netflix, which can count on optimisation technologies such as multi-pass content-adaptive encoding, variable bitrate encoding and multi-level caching. Live streaming is a different ballgame: all users need to view the same

streaming chunks with the lowest possible latency, which requires single-pass real-time encoding, constant bitrate delivery and very reliable backbone connectivity to the CDN Points of Presence (PoPs). Caching and multi-chunk buffering also become impractical. Additionally, when lots of concurrent users generate peaks in network capacity utilisation, packet loss ratios skyrocket, generating lots of packet resends that further clog the network. Buffering or poor video quality leads to strong negative feedback from consumers, which ultimately impacts the economics of the service. Despite the myth of this problem being limited to developing countries, it affects strong infrastructure markets too: last year US streaming users experienced poor video quality and/or buffering more than 30% of the time; service outages during major sports events frequently made the news in almost all developed markets. To understand the problem and its corresponding solution, let’s imagine a sports streaming provider with a flagship property like the national football league for its territory, and walk through the issues it would face. To break even the provider needs to grow as quickly as possible to between 3 and 5 million customers. Most operators initially focus on ensuring the basic infrastructure to successfully authenticate and manage approximately

500k users (a realistic number for the first few weeks of the new service). However, scaling that service up to profitability becomes more and more about the bandwidth available across multiple points of failure, and the ‘network collapses’ that are experienced when certain bandwidth thresholds are exceeded. The math is simple: if an operator targets 5 million customers, it likely needs to be able to manage over 2 million concurrent users during primetime, and peaks of usage above 3 million concurrents. Live streaming services typically recommend 2-3 Mbps for SD and 8-9 Mbps for full HD and support passable video down to around 800 kbps. If we assume a target of providing HD to at least 70% of viewers (which would still be worse than what traditional broadcasters have been doing for the past ten years), then we would have 2 million users, meaning 7 Mbps x 2 million = 14 Terabits of traffic. In order to guarantee a minimum service to all, it would be necessary to use 0.8 Megabits x 2 million = 1.8 Terabits. This would have to happen during primetime, on top of the bandwidth needed by social networks, which also peak during sports events and all other internet services These figures, which when we account for packet loss are closer to 20 Terabits, are very high, and arguably unrealistic given the infrastructure available in many countries. The entire internet capacity of most European countries and US states is smaller than 20 Terabits. The entire global capacity of the biggest CDN providers is between 30 and 40 Terabits. With planned capacity increases taking many years to materialise, the day on which 20 Terabits can be taken lightly is not yet in sight. Of course, by that time operators will be serving more viewers demanding even higher quality too. Luckily a proven solution exists, and it comes from the correct use of two essential technologies: Content Delivery

How to successfully deliver live sports at scale over the internet – by the numbersBy Antonio Corrado, CEO and Co-founder, Mainstreaming, and Guido Meardi, CEO and Co-founder, V-Nova

21-25_VNova.indd 4 14/03/2019 12:16:35

Networks (CDNs) specifically designed for video (known as VDNs, Video Delivery Networks), and next-generation video compression codecs. Utilising these can increase the number of last-mile connections able to receive top quality profiles, facilitate the distributed delivery to CDN POPs and dramatically reduce the number of Terabits needed. That ultimately delivers a higher-quality, more robust consumer experience. CDNs have been quietly managing the static and media content distribution market for two decades. For historical reasons, however, most CDNs, including those that handle the largest volumes of traffic, are not optimised for video delivery. CDNs are online delivery systems that, through a technological infrastructure with different PoPs and servers, constantly copy content to bring it closer to the final user. Transmitting “LIVE” means exactly that - without copying - and it is for this reason that traditional CDNs have difficulty in meeting the needs of today’s market. When it comes to video, operators should use the specific services of a Video Delivery Network (VDN) like MainStreaming, which is natively designed to distribute real-time content such as live video. The second fundamental element to solve the problem lies in compression codecs. The one used almost exclusively by streaming services today, is called AVC (also known as H.264). To date, the market mainly offers four

alternatives: HEVC, VP9, AV1 and PERSEUS Plus. With HEVC there are compatibility issues with many of the devices in circulation because new hardware is required, as well as known license liability issues. VP9 offers moderate compression benefits in live video but also has compatibility issues with many devices. AV1 is not optimal for live video because of the considerable computing power required and the substantial absence of devices that are able to decode it before 2021. The answer comes in the form of a low-complexity software technology that can bridge the gap between codec standards by greatly enhancing video quality whilst leveraging the capabilities of existing decoder hardware. V-Nova’s PERSEUS Plus does exactly that by building on top of AVC and HEVC today (and in future any codec that has achieved hardware penetration). PERSEUS Plus AVC has demonstrated unprecedented compression performance, backward compatibility with the devices in circulation, and lower computing power requirements for encoding. PERSEUS Plus AVC is also the only codec enabling HDR 10-bit delivery in combination with 8-bit AVC. Operators have utilised it with live UHD video at under 8 Mbps, SuperHD 1080p50 under 4 Mbps, full HD under 2 Mbps, and acceptable quality services for a football match even under 300 Kbps. This triples the audience that can be

served and guarantees a good quality of service even to those with low bandwidth. When HEVC delivery is an option, PERSEUS Plus HEVC demonstrates even higher compression efficiencies using over 70% less processing power than native HEVC. Going back to the user volumes mentioned above, they would become achievable targets if they were managed through a VDN and with the best compression on the market. Serving over 2 million users simultaneously, with 70% viewing in HD, would require about 1.8 Megabits x 2 Million = 3.6 Terabits, ensuring a minimum service to all would require 0.3 Megabits x 2 million = 0.6 Terabits. Packet loss ratios would also be much smaller, making actual required backbone and VDN capacities close to theoretical numbers. Similar throughputs are feasible in the short term, while the previous ones are not. On top of this, last mile connectivity would be sufficient to guarantee the required percentage of HD viewing, while previous bandwidth requirements would make them a further barrier to quality of service. Change is essential for the entire industry to satisfy such large pools of users especially when more and more broadcasters and operators are gearing up for streaming delivery too. The growth of sports and esports is increasingly built on being able to reach fans with high quality video and low latencies wherever they are, across multiple devices and at a cost to suit everyone’s wallet. Thanks to interactivity, personalisation and ubiquity, OTT has the potential to become not just TV-like, but better-than-TV. However, large-scale live streaming operators and content rights owners must not underestimate the technical challenges of live OTT streaming, and should consider pursuing the technical solutions that are already proven to deliver the quality of service that everyone expects and deserves.V-Nova can be found on booth SU8213 at NAB 2019

Sports streaming


21-25_VNova.indd 5 14/03/2019 12:16:48

Untitled-1 1 18/03/2019 09:53:55

Blockchain is rising up fast on the priority list of the media & entertainment industry. Because controlling and licensing content is

becoming more complex in a digital world, one of the key use cases being examined by the industry is the role Blockchain can play in digital rights management (DRM).

A new model is starting to emerge that is more transparent, where the owners of the content have more control and can manage their IP intellectual property. That’s where new technologies that enable that, including Blockchain and smart contracts, come into play.

Dell Technologies has developed its own blockchain solution that enables media & entertainment organisations to increase the level of trust and transparency among all players in the content value chain, as well as reducing many of the inefficiencies, manual processes and overhead that goes ahead with managing content rights.

At BVE 2019, Dell EMC’s Blockchain strategy lead, Nkiru Uwaje, talked about the vision that the company has for the future of content rights. She noted that an estimated 25% of the royalties are still being left unpaid due to lack of being able to identify who

owns the content, the multiple owners who create music or films.

Who owns the content and where does the money go? How can technology enable companies to think differently about business models? This is where Blockchain comes in. Dell believes that technology, especially such as decentralised platforms like Blockchain, can enable companies and individuals to drive more trust and transparency within this ecosystem.

Dell sees a number of use cases having the most impact and potential within the M&E industry, including among them those around digital rights management (DRM), royalty systems, fighting piracy, fair trade for talent and micro payments for artists.

“By no means are we saying that Blockchain is that magic pill that you will have and everything changes. But what we are seeing is it represents a different type of logic that enables you to think differently,” Uwaje said.

Blockchain is potentially transformative of DRM and ownership. It is a decentralised digital ledger that gives everybody an equal copy registered across every block or participant, all on a single database. In a future scenario, content creators could upload content like video into a legally binding smart contract placed into the Blockchain network. Once recorded, a digital watermark is

embedded on top, acting like a digital serial number for content.

The system also performs a real time auditing and reporting feature that keeps time stamping consumption and distribution in a cycle. The entire ecosystem from start to finish is linked to the smart contract, and back to the content management

system. “This is the part that all content creators I think should care about and that makes me excited about

this platform,” Uwaje said. “Imagine a smart contract where every single word is code and algorithms, it means there are rules to it. These smart contracts can heavily automate processes but more importantly, because of the transparency of information, we can trust the contracts we put out there.”

The kind of information that can be stored includes ownership, publishing, usage rights, watermarks, payments and others. The algorithms behind the smart contract can execute on all these, according to Uwaje. This is also why smart contracts are also being introduced in other industries like insurance, property and finance. She mentioned an upcoming music festival later this year will be powered by Blockchain in the back end, acting as the smart contract to enable tickets sales based on that logic.

Blockchain, she argued, makes it a lot less likely to have content illegally downloaded and distributed.

These are the reasons why Dell has invested a lot of R&D in the last few years (its permission- based Blockchain is an abstraction layer of Ethereum virtual machine), helping content creators and big organisations working in the M&E space to visualise the change and monetise it too.

To this end, some of the capability Dell has built on top of its platform is focusing around digital rights, identity wallets, micro payments, real time transactions, even reducing some piracy in terms of prevention through account security methods.

Dell is working with very big media entertainment companies, well-known household names and brands, focused on using Blockchain for DRM and also payment reconciliation, joining many other similar projects out there, as the industry tests the technology and new business cases.

Blockchain and future DRMCSI hears Dell’s vision for Blockchain in digital rights

Blockchain


27_Blockchain.indd 2 21/03/2019 11:22:24

Online video usage is undoubtedly increasing, at the same time as streaming is grabbing not just most of the

attention, but also R&D budgets and technology innovation. Mindful of these market and industry dynamics, the Advanced Television Systems Committee set about modernising its core standard to embrace broadband as much as traditional broadcast.

“We don’t have to wait for full broadcasting to make full use of standard. It’s a suite of standards,” notes Alan Young of Crystal. “If you want to use only internet delivery you can, there is nothing to stop you from doing that,” he says. “But if you want mass market through broadcast, infrequent as it may be, you can - you get to decide most efficient way of delivering your content.”

An enhanced version of what is now retroactively known as ATSC 1.0, ATSC 3.0 brings a number of benefits to broadcasters, including a more robust and efficient transmission system that is comparable to DVB-T2, by leveraging their licensed spectrum. It employs IP packets for delivery, rather than a DVB Transport Stream, and is arguably more aligned with online and mobile services.

There is a much more robust modulation scheme that enables mobile television to handhelds and in-vehicle devices moving at high-speed, according to Matthew Goldman and Lisa Hobbs of MediaKind, which has been closely involved in both the creation of the ATSC 3.0 standards and in the development of ATSC 3.0 broadcasting and receiving solutions. The company has encoding and packaging solutions (the latter a requirement in all IP systems), as well as targeted advertising solutions that broadcasters are looking to deploy.

ATSC 3.0 offers support for newer technologies such as HEVC, which is

more bandwidth efficient, and can support video resolutions exceeding 2160p (4K UHD) at 120 frames per second. It also provides support for HDR and wide colour gamut (WCG) immersive video technologies, Dolby AC-4 and MPEG-H 3D immersive audio technologies, as well as datacasting capabilities.

The modulation scheme supports layered division multiplexing (LDM) and multiple physical layer pipes (PLPs), each with its own trade-off between bits/second/Hertz and robustness (signal-to-noise ratio), thereby enabling a large variety of different services to be offered under different reception conditions simultaneously.

ATSC 3.0 is based on a native IP transport layer and utilises the DASH delivery protocol for OTA channels, according to Harmonic’s Jean Macher, enabling technology such as receiver clients, DRM and targeted ad insertion systems to be reused and a straightforward path to OTT, where DASH has been used for years.

In addition, the standard enables broadcasters to deliver geo-targeted ads, emergency alerts in the most adverse conditions, VOD, and other interactive services using a broadband return path for viewers with internet access.

With ATSC 3.0, broadcasters can also explore OTA pay-TV options such as channel- or event-based pay-per-view. “The opportunity for dynamic or targeted ad insertion as well as DRM for ‘freemium’ or pay-TV services that require viewers to register is also possible,” says Macher. “ATSC 3.0 is leading the charge to the next era of broadcast television.”

All of these technological improvements sound promising on paper translate into better services and open up new business models. The main attraction appears to be using broadcast spectrum to offer additional services to mobile users and the opportunities for enhanced advertising.

DOA? Not so fast…Nevertheless, ATSC 3 has been declared in some quarters as “dead on arrival” (DOA), especially in light of 5G. The bottom line may be that 5G is already winning the content delivery race before it has even officially begun, Ooyala and other observers have speculated.

While Goldman and Hobbs think it is premature to declare ATSC 3.0 DOA, they agree commercial reasons will play heavily into its usage and deployment. “It’s all about effective network efficiency and the trade-off between unicast, multicast and broadcast utilisation. There is certainly a place for both technologies to complement each other, but again, it comes down to the commercial use cases,” they say.

Broadly, the main use cases are:• Localised broadcast streams;• Partnering to create interconnected

delivery across a broadcast, MVPD or streaming service providers’ IP stream on partnered stations;

• Alternate delivery or staggered delivery for live events by device types;

• Targeted advertising;• Advanced Emergency Alerts; and• Enhanced Video and Audio

delivery for a better viewing experience, including higher resolution (frame rates) and more consistent signal quality.

ATSC 3.0

Can ATSC 3.0 rescue FTA in an OTT world?Is the new ATSC standard in danger of missing the boat, asks Goran Nastic


28-31_ATSC.indd 2 21/03/2019 10:27:13

“These will help to shift the boundaries between traditional broadcasters and content service providers,” reckon Goldman and Hobbs. “Beyond the obvious, there are broadcasters looking at ways to pool the spectrum at their disposal and treat it as a data pipe. Automotive updates over the air (think navigation systems), corporate data transmission, etc.”

There are also plenty of opportunities for 5G and ATSC 3.0 technologies to complement each other. For instance, one case would be bandwidth sharing/swapping, with so-called ‘short tail’ wide-audience broadcasting benefiting with ATSC 3.0 and ‘long tail’ small multicast or unicast benefiting from a cellular-type network.

“5G is a great candidate to provide the broadband path to a hybrid ATSC 3.0 receiver,” says Macher at Harmonic.

Proponents, therefore, argue that ATSC 3.0 will allow broadcasters to implement new business models that will have a dramatic impact on the TV industry over the next decade, including more holistic hybrid broadcast and broadband services, blending the personalisation with mass delivery.

“5G does not make the ATSC3.0 transmission standard DOA; it just creates competition across different industries attempting to deliver content and information to consumers. This competition enables users to accept and become comfortable with various types

of devices for media consumption in different circumstances,” argues Ronnie Bell of Imagine Communications.

“Could this be VHS vs Betamax? Possibly – it could come down to ease of use, but might also be split by location, device and signal availability. It could also lead to better

alignment, for example, between using broadcast signals for mass reach and point-to-point for other purposes,” says Bell, adding: “ATSC 3.0 provides an opportunity for traditional OTA broadcasters to disrupt MVPD and Digital Advertising markets, and blurs the lines between linear and non-linear, offering the ability to stream cross-platform. “It creates the opportunity to reach the ‘cord-cutter’ audience.”

This is a sentiment echoed by Pearl TV, which sees the convergence of broadcast with streaming in ATSC 3.0 attracting ‘cord nevers’.

Indeed, there is logic to that as numbers from Nielsen on over-the-air and OTT viewing in the US show.

The number of broadcast-only households in the US has remained steady at around 11 million for the last three years, but Nielsen found the number of homes using an antenna to watch television has actually increased, from 11 million in Q2 of 2010 to 16 million in Q2 2018 (see chart above). Almost all the growth in OTA households has come from homes that also watch some form of internet TV. Most pair their OTA line-up with streaming services; as of May 2018, almost 60% of OTA households combine broadcast television with IPTV/OTT services. In total, broadcast-only, broadband-only and broadcast&broadband combined account for some 20 million homes.

Aggressive timelineThe move to ATSC 3.0 is not an enforced regulatory migration, but perhaps the adoption timelines show how pressing the broadcast community sees the need to go down this route.

The ATSC finalised the umbrella ATSC 3.0 television standard in January 2018, and the FCC started approving limited deployments a few months later. A year on, tests and trials have been underway in a number of US states to determine technical parameters and business implications, and these will go on throughout this year.

Major US broadcast organisations that cover virtually the entire US footprint have stated that they will be on-air commercially with ATSC 3.0 before the end of 2020. They include the Sinclair Broadcast Group (191 network-affiliated TV stations) and the Pearl TV partnership of eight broadcasting companies (representing over 220 network-affiliated TV stations) together with the technology vendor community.

“With commercial launches planned for 2020, this is a fairly aggressive schedule for deployment,” note Goldman and Hobbs.

With no extra frequency to manage the transition, broadcasters need to plan and work together in each market where they want to launch ATSC 3.0 services. This requires a new type of collaboration and significant planning.

As Myra Moore of Digital Tech Consulting points out, the last time US broadcasters transitioned from an old to a new system, conditions were vastly different, and this time there is no government mandate or organising body for a wholesale switchover.

“Although impossible to quantify, the new competitive video services landscape cannot be underestimated as a motivating factor for broadcasters to cooperate with each other - especially if that cooperation introduces technical efficiencies that can enable additional revenue streams, reduce ongoing operating costs, and create a platform

ATSC 3.0


Sour

ce: N

iels

en

28-31_ATSC.indd 3 21/03/2019 11:09:43

ATSC 3.0


by which a single market can transition from the old to the new system. DTT broadcasters know they have a high hill to climb and that they must upgrade their networks to see the other side,” she added.

One of the main challenges and drawbacks is that ATSC 3.0 is not backward compatible to the existing ATSC 1.0 system. So the cost of a new transmission chain for ATSC 3.0 is a limitation, as is the cost of better ATSC 1.0 compression in order to squeeze additional channels coming from the broadcaster hosting the ATSC 3.0 transmission, according to Macher.

This equates to investments needed to change and update hardware, meaning CFOs and CTOs will need to take a good hard look at the upgrade costs and ROI.

“Implementation will require upgrading signalling and transmission systems but the biggest challenge for broadcasters will be convincing consumers to use compatible receivers,” notes Dr William Cooper of consultancy informitv.

As it happens, while hardware support is not entirely clear, there are some positives to take in terms of release plans for some end-user devices.

Build it and they will come?Because broadcast groups and networks are committed to having a high number of markets lit up with ATSC 3.0 by the end of 2020, so compatible TV sets are likely to be available in retail markets by then. Compatible televisions are already available in South Korea and are expected to be available in the US from numerous manufacturers in Q4 2019.

All of the major CE makers were very active participants in the development of the ATSC 3.0 standard, and are now waiting for clarity from the market. “Many major consumer electronics manufacturers have said they will support ATSC 3.0 but ultimately they will be driven by the extent of adoption and consumer demand,” says Cooper.

“CE manufacturers are onboard but are trying to confirm what features and capabilities are needed to launch ATSC 3.0, as well as when it makes sense to introduce 3.0 products,” admits Macher at Imagine.

Connected TVs are the most likely candidates and, luckily, the largest TV maker in Samsung, as well as LG, are both based in South Korea, where ATSC 3 was first used (see box). For its part, Sony said it will supply TVs, demodulators and application development tools.

This rollout, on top of perhaps popular streaming hardware like Roku devices, will be the real litmus test for the success of ATSC 3.0.

With the current pace of change in the US TV landscape, the risk is that people start watching broadcast network content using D2C (direct to consumer) platforms, or SVOD platforms, bypassing broadcasters entirely.

“Some may regard it as already too late, but there is always a requirement to move standards on and in that sense ATSC 3.0 is a step in the right direction,” says Cooper.

“With the amount of technological change, there is definitely risk, especially with a voluntary adoption programme,” acknowledges Imagine’s Bell. “However, some of the models provided by ATSC3.0 are not easily replicable by other technologies, so we expect a lot of the work done on things like monetisation will be interesting to broadcast and other industries for many years to come.”

“It’s difficult to imagine what commercial alternative will become pervasive during this timeframe that would have the effect of making the

goal of these broadcasters obsolete,” agree Goldman and Hobbs. “That said, this always has been about having a viable commercial model for the varied TV and advanced services that ATSC 3.0 enables and it will be the roll-out of these services that will determine the success of failures of these endeavours.”

And, as Crystal’s Young proclaims: “We hear how broadcast is dead, OTT is taking over. OK, that might be the case but broadcasters won’t just sit there and let it happen.”

The fightback then, starts here, even if it is largely a case of if you can’t beat them, join them.

ATSC 3.0 globally South Korea has already adopted ATSC 3.0. Broadcasters there have been using it since just before the 2018 Winter Olympic Games in PyeongChang, mainly for UHDTV delivery, with the goal of full country coverage by 2021. It is interesting to note that this is in a country with one of the most advanced broadband infrastructures in the world, with high fibre penetration and some of the fastest speeds over fixed and mobile networks.

Elsewhere, Canada and Mexico are likely to adopt it, as they currently use ATSC 1.0.

“While the specific transmission standard may only be adopted in markets like the US, S. Korea and the Americas, we believe that many of the resulting capabilities that ATSC 3.0 is enabling will be standardised and adopted by other markets around the world,” says Bell at Imagine.

MediaKind’s Goldman and Hobbs point to some interests being expressed in Latin America (Brazil specifically, with on-air 4K tests being run in Rio de Janeiro) and even India, “but nothing definitive at the moment”.

ATSC3 could be a candidate in regions where digital terrestrial television is not widely adopted. Yet alternatives, such as the DVB standards out of Europe, are already broadly deployed and supported.

28-31_ATSC.indd 4 21/03/2019 10:28:02

Untitled-2 1 21/03/2019 14:03:13

Over-the-top (OTT) platforms have drastically transformed the video and entertainment landscape with new technologies, and it’s

projected to keep up this shift. In this new media consumption era, the viewer’s are the king, and their expectations of content and experience decide the fate of a platform. The race to please the viewer is quite evident by the fierce fight of OTT platforms for high-quality content, immersive experiences and value to the viewers. The motivation is the boom of the OTT industry and a very fruitful projected revenue of $129 Billion by 2023 with a user base of 777 million.

Throughout the years, OTT services have been synonymous with video-on-demand (VoD), but with innovations in delivery methods, technology, features, and user preferences, the industry is witnessing a shift. Viewers are now demanding the convenience and experience of VoD for linear channels and live events as well; this makes it crucial for OTT platforms to establish a solid business model targeting multiple screens. This growing demand for live content on the go is why many new VoD and digital multichannel video-programming distributors (dMVPDs) are also entering the arena.

Differences: VoD vs LiveAs compared to VoD, Live OTT workflows differ vastly. The variances start from the initial step of video ingestion to the final step of delivery of the stream. Besides the workflow, Live OTT also requires a different set of features, such as catch-up or personal video recorder (PVR) from the user

side, and blackouts, program substitution, and ad-insertion from the business side. Although many of these features

have been available in a linear TV platform for quite a while now, the expectation of a high quality of experience (QoE) from an OTT service, and low switching costs in the industry means that brands must deliver a seamless live experience over multi-screens as well.

Coverage of live sports or popular events may lead to millions of users hooked on the stream at the same time. This can cause multiple implications, ranging from scaling for content processing to the scaling of bandwidth. Streaming of valuable live content comes attached with the question of security; studios, broadcasters and event organisers want their content to be protected with the best security measures to circumvent illegal access and piracy.

As live video streams cannot be as easily recovered from a data loss and delays as VoD content, the chances of loss and failure are inherently higher in Live OTT. From a business perspective, live event streaming failures can ruin the experience for end-users as well as the brand image of the provider.

Another aspect that makes live OTT unlike VoD is the presence of time-based metadata. In VoD, the metadata comes in the form of assets, such as duration, title, language and much more, whereas in Live scenario the metadata also consists of timestamps or markers (SCTE-35) that help with ad-insertions, blackouts and nPVR.

Expectations from Live OTT Platforms:

Automation. To manage a large number of linear channels, attached

metadata or simultaneously broadcasted events, end-to-end automated workflows become essential. This requires automated workflows for all processes: ingestion, transcoding, encryption, packaging, and publishing of the content.

Robustness. A stutter free, high-quality and consistent video is the fundamental requirement for any platform. The platform must be highly-performant to deliver the live stream with zero downtime to have an optimal experience for the viewer.

Resilience. Smaller faults during processing or delivery of live streams can pile-up and cause more massive failures. Real-time monitoring, fault tolerance and greater support must be imbibed in the platform to guarantee a very high uptime to the viewers.

Scalability. As stated earlier, live OTT might mean serving hundreds of linear channels and catering to potentially millions of users when it comes to live events. The platform should be able to scale up the delivery and DRM solutions to accommodate the millions of requests for the video stream.

Extensibility. Capability to accommodate the changing user preferences backed by the drive through analytics, prompt issue resolution, and achieving new monetisation opportunities should be possible within the platform. The platform must be extensible to allow addition and modification of features, device compatibility and services (external or internal). There should also be a provision for adjustment and customisation of the workflow.

Modularity. Modern digital media solutions should be built on containerised micro-services and serverless architecture, enabling them to be highly modular as well as scalable.

Advertorial feature

Building Live OTTDemystifying live content distribution over OTT


32-33_Anixom.indd 2 18/03/2019 14:08:40

Such an architecture facilitates the platform owners to choose, integrate, operate and scale services that fit perfectly with their technical and business goals.

Protection. Content protection and security are becoming a growing concern for broadcasters and event owners. A live OTT platform must facilitate protection of content using digital rights management (DRM) standards, forensic watermarking and entitlement services. Multi-DRM service that respects cross-platform compatibility and can secure content across devices must be in place to attain continuous security of the content.

Monetisation. The end-goal for businesses is to generate revenue from their OTT initiatives, with Live TV the monetisation opportunities are limited to ad-insertion and selection of service model. However, in the case of advertisements, modern browsers and devices can circumvent many of the ad-insertions techniques; therefore, the platform must allow for dynamic server-side ad-insertion or ad stitching.

Axinom Live OTT ArchitectureAxinom live OTT platform is composed of several individual components that do all the complex tasks ranging from receiving the stream from vendors to serving it to the end-user devices. In a modern, scalable and extensible platform, these components reside as containerised micro-services. The figure below visualises the process for easy comprehension.

The basic media workflow for a live OTT starts with the process of ingestion where a video stream, provided by the vendor, is transcoded into streamable file formats. The source file metadata also include SCTE-35 messages; alternatively, the editor can specify the markers in the CMS for signalling the period and

duration for ad insertion into the streams. While packaging the content into adaptive streaming solutions such as DASH and HLS, the packager inserts SCTE-35 metadata into a manifest file or emsg boxes in case MP4 containers are used.

A third-party ad-service utilises the client manifest generated in the previous step. The ad-service can dynamically inject advertisements based on the SCTE-35 markers; this processed manifest is then passed on to a content distribution network (CDN) for playback of the live streams and the inserted ads on the user devices.

The multi-DRM service that comes with a developer-friendly unified API allows encryption using the popular modern technologies such as Microsoft PlayReady, Apple FairPlay Streaming and Google Widevine with CENC encryption scheme. Key rotation can also be performed to change the encryption key associated with the live content.

Live OTT with AxinomAxinom, with its modular and robust products and many industrial partnerships, provides the full media workflow that accompanies the live OTT as well as VoD. Axinom CMS can manage any video content, be it linear TV channels

or the EPG and time-based metadata associated with it. Catch-up duration, advertisement configuration, blackout regions and business and content rules can be defined within the CMS, along with providing accessto multiple users or external participants.

Axinom BIS takes care of all the integrations needed as well as the execution of business rules defined previously. The middle-layer facilitates ad-insertion, querying, filtering, erification and access to content assets and streams. In addition, it has several other features, such as authentication, entitlements, full-text search with filtering, DRM licensing, advertisement support, reporting, payments, monitoring and many more.

Axinom’s digital solutions have enabled media and broadcast organisations in reaching a global audience with their content. With the right mix of robustness, scalability, and extensibility, Axinom products fulfil viewer expectations in the choice of content and device, with unlimited localisations, not to mention supporting business goals with a spectrum of monetisation and operating models.

Axinom can be found at NAB Show 2019 on booth SU10810 and online at www.axinom.com

Advertorial feature


32-33_Anixom.indd 3 18/03/2019 14:08:57

Organised by:

www.NXTasiaExpo.com

Held Concurrently:

www.CommunicAsia.com

Join in the conversation:

#BroadcastAsia#ConnecTechAsia

A part of

Asia’s must-attend international event for the pro-audio, film, digital media and broadcasting industries.

www.Broadcast-Asia.com

CaptureT h e F u t u r e

18 to

20 JUNE 2019

SUNTEC SINGAPORE

www.broadcast-asia.com/register

Register Now

BCA19_VP_CSIMagazine_H271xW204_D1.indd 1 5/3/19 5:42 PMUntitled-1 1 06/03/2019 09:53:19

Sponsored by:

ww

w.c

sim

agaz

ine.

com

Accelerating video content in cloud environments

Media in the cloud

35-39_Aspera.indd 1 21/03/2019 14:16:16

Conventional licensed software is evolving into cloud-based delivery over the Internet, driven by the popularity of applications such as

Software as a Service (SaaS), Platform as a Service and Infrastructure as a Service and accessed via on-cloud service platforms like Amazon Web Services, Microsoft Azure and Google Cloud Compute.

As data-intensive sectors like Media and Entertainment increasingly migrate to off-premises solutions, data ingest becomes a significant challenge because large files must be relayed, often from multiple locations worldwide for use in time-sensitive applications, but conventional internet file transfer can be slow and unreliable.

Latency and packet loss are a major obstacle to long distance file transfers over conventional internet protocols. As a result, many businesses have chosen to adopt UDP (User Datagram Protocol)-based solutions that operates outside the traditional network overhead to significantly ramp up file transfer speeds.

The cutting edge UDP-based software technology FASP was developed by IBM Aspera to cut through broadband bottlenecks and achieve transfer speeds up to hundreds of times faster than traditional FTP.

The technology provides the foundation for the Aspera on Cloud platform, making it possible to shift large volumes of data regardless of file size, transfer distance and network conditions. Video processing company Encoding.com uses Aspera’s solution to migrate high-volume production video workflows from media companies to its AWS-based cloud platform and back out again.

Jeff Malkin, president of cloud media processing service Encoding.com, told CSI: “We process and deliver files to CDNs, cable head-ends and other

syndicated video hosting platforms. Moving large and valuable video assets through the public internet can be very slow and has security risks. Using Aspera, customers can deliver valuable video content fast and securely to our platform. The speed of overall video workflow, including file ingest, queue times, processing time and egress is critical for many large media and entertainment companies.”

Other media and entertainment companies working with IBM Aspera include Netflix, Universal Pictures, BT, Channel 4, Deluxe Digital Solutions, Australian Broadcasting Corporation, and Media Corp.

Overcoming latency and packet lossMultiple shifting dynamics in the media and entertainment industry are driving the need for faster file transfer over the internet. Video files continue to increase in size as content producers and distributors promote higher quality formats, using High Dynamic Range and larger frame sizes such as 4k and UHD.

The transition towards a highly decentralised and distributed model has resulted in a dramatic increase in the locations where content is produced, processed and distributed, increasing uptake of more readily available cloud-based storage and processing.

Jay Migliaccio, offering manager, Aspera on Cloud at IBM Aspera told CSI: “There are multiple trends at play,

including the switch from physical to digital distribution and production, the dramatic increase in firms that are able to produce, thanks to the internet, and the globalisation of production. Those trends are firmly in place so more horsepower is needed to move files around.”

Files transfers over the internet have traditionally relied on popular TCP-based protocols, including FTP/S, SFTP and HTTP/S, which facilitate communication between a web server and an operating system to organise how files are moved up or down stream.

TCP was introduced in the 1970s and has proved ideal for applications, such as email or web surfing, but when large file sizes are being sent longer distances a fundamental flaw is exposed. If the TCP layer in the OS (responsible for ensuring the reliable delivery of packets of data) detects that a packet has not arrived, a bespoke algorithm causes it to ‘throttle back’, retransmitting packets at a much slower rate, before slowly increasing that rate.

In the highly congested conditions of today’s Internet, where busy routers and multiple network stages increase latency and packet loss, TCP retransmission can dramatically slowdown transfers.

“Packet loss and congestion are a reality in the public internet today, but we shouldn’t have to let that slow down transfer rates. Aspera created FASP as a fool-proof high speed communications protocol to move bytes across the internet,” says Migliaccio.

Cloud focus

Media management in the cloudMedia companies’ increasing use of cloud storage and software services threatens to create bandwidth bottlenecks, but a new breed of super-fast internet fi le transfer can help. Stephen Cousins reports


35-39_Aspera.indd 2 21/03/2019 11:33:04

Aspera’s FASP protocol harnesses UDP to send packets at high speeds, however it cannot confirm if packets arrive safely. To do that, an extra layer of authentication and security is added to guarantee the timely delivery of every packet.

Aspera On Cloud has been a game changer for film distribution company Bitcine, which embedded the solution into its products used to move files into and out of AWS S3 storage. Colin Carter, Founder and CEO of Bitcine told CSI:.“We deal with files that are hundreds of gigabytes in size, so conventional transfer options aren’t even a viable option. Key advantages are the increase in transfer speed, auto- resume functions, and resiliency to poor quality connections.”

Another satisfied customer is Silver Trak Digital, which aggregates content for many large SVOD/OTT platforms. According to company COO Christian Christiansen, the ability to increase bandwidth during peak demand times using Aspera on Cloud is critical: “We operate a Mobile Application Management SaaS in the cloud where the bandwidth requirements are decided by how many files clients upload, download, and/or deliver. The ability to spin up Aspera On Cloud instances to meet demand is essential in providing a good service,” he says

Media use casesUse of Aspera on Cloud is mostly being driven by media customers via three key use cases. Content producers send video content back and forth for editing; content distributors, often with massive libraries, transfer licensed content to customers in markets worldwide. The third batch of users are SaaS vendors that QA and transform video into different formats, for example transcoding a master format into a format suitable for TV or OTT broadcast in different countries, or other asset management tools.

Cloud transfer offerings have had to evolve in recent years from the simple

movement of files from a PC to a virtual machine, to enable files to interface with a plethora of services currently exploited by media users.

Aspera on Cloud is based on a versatile software-based architecture that can be installed on any standard operating system. It is able to connect to users’ on-premises servers and those of all major cloud storage services and SaaS providers. The platform supports the full gamut of rich metadata available in object-based cloud storage and is constantly updated to reflect new additions.

Innovation at the core of FASP has enabled Aspera to expand into new transfer territory, says Migliaccio: “This year our focus is on adding automation to the platform, most of our customers have been asking how to automate different aspects to add efficiencies into their workflows. This is something we have included in our on-premises software stack for a while, so extending that concept up into the cloud is a natural progression.”

The Cloud offers opportunities to use different events as triggers to automate tasks. For example, when a file is uploaded to cloud storage, the start of the transfer can be an event that triggers the next stage in the file’s journey, such as delivery to its final destination, or communication with a SaaS application that will review or process it. An event could trigger an API call to any third party service to process the content.

Encoding.com is already providing support for this handy feature. Content owners with an Aspera High-Speed Transfer Server can utilise the Encoding.com API to automate the transfer of files from the source location to the firms processing centers, as well as deliver encoded content to specific destinations.

Aspera is injecting innovation into the live streaming space, which faces similar challenges in terms of guaranteeing speed and performance over long distances. Production teams require fast turnarounds to send live video, monitor, edit, and QA it and then re-publish the content for consumption by a downstream consumer audience. Dedicated fiber or satellite connections are a fast, but expensive option, FASP streaming over the public internet could offer a cheaper and more effective alternative.

Aspera’s first major live streaming test case saw FOX Sports stream live footage from the World Cup in Russia to editors in LA to live edit clips then re-publish as highlights on news channels. The production was so successful the broadcaster is looking to expand its use in future live events.

“Live streaming opens up numerous opportunities for major broadcasters,” says Migliaccio. “It goes back to the subject of decentralised production and the ability to cover live events anywhere in the world without need to license expensive transmission technologies. Although people aren’t likely to switch off satellites that have worked fine for the past 10 years, there will definitely be more tests in this exciting area.”

As the cloud continues to evolve and media companies seek to take advantage of its scalable and cost-effective services, fast file transfer will play an increasingly vital role in supporting its future success.

Cloud focus


35-39_Aspera.indd 3 21/03/2019 11:33:26

Focus feature


Video content is growing. The frequently coined phrase ‘Content is King’ has evolved to ‘Video is King’. Viewers want more

on-the-go video. And they expect to be able to watch video wherever they are, whenever they want, and on any device they choose. These demands require traditional broadcasters and a new breed of producers to create more video than ever to satisfy viewers — both on-demand and high-quality live streamed — in a growing number of formats, resolutions, and bit rates to support a variety of connected devices.

Producers must deliver faster and more efficiently than ever before to succeed in this highly competitive media environment. Content production networks are expanding, and new kinds of workflows are emerging to achieve quicker turnarounds. Media companies’ infrastructure, including servers, storage, and network resources, must accommodate massive volumes of video content. At the same time, security

requirements remain as strict as ever. To cope with all these pressures,

organisations are increasingly looking to move their workflows into the cloud, which can deliver many operational and financial benefits while helping companies scale out their video production.

Three key advantagesToday’s production networks consist of hundreds, sometimes thousands, of partners who contribute content on a daily basis. The rapid expansion of these networks demands self-service provisioning and enablement as well as seamlessly scalable technology offered by the cloud. Cloud infrastructure delivers virtually unlimited access to compute resources without the large upfront investments required for traditional data centres. From a business perspective, the cloud offers three key advantages: it removes the limitations imposed by computing and storage infrastructure within fixed footprint data centres; it eliminates the need to over invest in infrastructure to meet peak demands, which then sits idle during normal operations; and it improves cash flow by shifting

investments from a Capex to an Opex model with pay-as-you-go options that charge only for resources that are actually used.

The scalability of the cloud is particularly

important for live video use cases such as live sports broadcasts. The nature of these events brings huge spikes in viewer demand during the event, which quickly disappear once the game, match or

tournament is over. Cloud-based workflows enable companies to quickly scale production resources up or down as needed to match variable demand. This flexibility is what makes the cloud such a critical component of an architecture designed to service the changing needs of live video production.

Accelerating collaboration With media organisations needing to create more high-resolution video content in increasingly shorter turnaround times, rapid collaboration across the content production network is more critical than ever. Contributors need access to collaboration workspaces that facilitate exchange and delivery of video between users while always enforcing the strictest security standards.

New cloud services can transform cloud-based video sharing and collaboration by eliminating the complexities of sharing large video files with fast and secure transfer of the largest video formats, digital cinema packages, and associated metadata. By adopting a high-speed transfer platform in the cloud, organisations can easily enable members of the network to quickly and securely exchange and deliver video of any size, over any distance, and between end users across different organisations.

Faster file transfer in the cloudWhile the cloud offers many benefits, moving large video files into it can introduce delays that are unacceptable for fast-turnaround video workflows.

Empowering video management, processing and distribution in the cloudBy Jay Migliaccio, offering manager at IBM Aspera

35-39_Aspera.indd 4 07/03/2019 15:30:08

Legacy transfer technologies, such as FTP, are slow and unreliable, while shipping physical disks is time consuming, introduces additional delays, and exposes valuable video to unnecessary security risks.

Typical file transfer acceleration “gateways” upload video to cloud object storage in two phases. First, video is uploaded over the wide area network and written to the local virtual machine-attached block storage. In the second phase, the video is copied from the local storage to the target object storage. The result is a significant delay from when the transfer is started to when the video is readily available for use, as well as increases in local storage costs and compute time and costs.

To become a practical option for video workflows, cloud services need a secure, high-speed data transfer solution to move large video files and large volumes of video to, from and between cloud and on-premises environments. That mechanism must address two main bottlenecks—the degradation in WAN transfer speeds that occurs over distance using traditional transfer protocols, and the “last foot” bottleneck inside the cloud data centre caused by the HTTP interfaces to the underlying object-based cloud storage.

Fortunately, recent advances in high-speed transfer software services running in the cloud enable high-speed upload and download of large volumes of video and large video files directly to, from and across cloud object storage systems.

Security can be assured through secure access keys, and encryption in transit and at rest.

Managing complex workflows Creating, capturing, processing and

distributing video globally results in complex cloud workflows that require automation, visibility and management. Using an advanced workflow automation system to streamline complex processes eliminates human error, reduces cycles and guarantees rapid video delivery. A complete workflow automation tool provides a seamless end-to-end solution with third-party integrations that can scale to handle high-speed transfers of large file volumes between any location, whether on-premises or in the cloud.

The system can automatically invoke the right application—whether transcoding, watermarking or file format validation—and automatically manage the processing as well as handle conditional decision making. It can intelligently move workloads between cloud platforms and on-premises environments, ensuring companies can meet processing demands even when they spike to their highest levels.

Media companies can also use workflow automation to prioritise workflows based on user-defined characteristics, pushing the most important projects to the front of the completion queue so that high-priority work is completed first.

A workflow automation system can also monitor the lifecycle of video at any time or point in the workflow, as well as the system health and status of every job. It can automatically notify the team about processing errors and provide a recommended resolution.

And if a supporting application becomes unavailable, it can quarantine the application, notify the team, and re-route the workflow to alternative systems.

Moving content across multiple cloud environmentsWhile media companies are increasingly adopting a hybrid, multicloud workflow that uses a combination of public cloud, private cloud, and on-premises storage and compute resources, moving content between these environments has become much more challenging. Files that need to be exchanged are often stored in multiple clouds and on-premises systems. Traditional transfer technologies bridging these environments are slow and unreliable, and physical disk shipments between them are time consuming and expose content to unnecessary security risks.

IBM Aspera on Cloud overcomes the file transfer challenges of the hybrid cloud by allowing media companies to securely and reliably move their content across on-premises and multi-cloud environments at unrivalled speed.

With IBM Aspera on Cloud, media organisations can seamlessly access and share content stored across multiple clouds and on-premises data centres. It allows internal and external users to collaborate in a secure environment that tightly controls access to content and application functionality. Using Aspera’s FASP® protocol, large files and data sets are transferred across the storage environment securely – overcoming the limitations of other file transfer technologies to move content at maximum speed regardless of network conditions, physical distance between sites, and file size, type, or number.

The cloud offers an attractive option to keep pace with demands. Intelligent workflow automation solutions and advances in high-speed file transfer and collaboration services enable media companies to satisfy viewer demands and efficiently handle the increase in video volumes, file sizes and formats.

Focus feature


35-39_Aspera.indd 5 07/03/2019 15:30:39

It was hard to miss the sheer volume of 8K TVs that manufacturers had on display at CES in January, sparking excitement amongst consumers and TV enthusiasts alike. There is

always plenty of demand for a better viewing experience, and loads of new acronyms to go with it, QD, QUHD, OLED, SUHD are but to name a few. It’s fair to say that this is not only confusing for the consumer, but the industry could be forgiven for scratching its head a bit too. So what are the facts? 8K TV (also known as Super Hi-Vision) has 4 times the detail of 4K and 16 times more detail than 1080p. 8K resolution represents 7690 x 4320 pixels. What’s so great about 8K?8K is quite literally bigger and better than 4K; an 8K TV offers such a rich amount of detail that viewers can experience a quality picture on a 70inch

screen from only a few inches away. If the 70inch screen were in 1080p (HD), it might be difficult to concentrate without being distracted by the large pixels that make up the picture structure. Globally, TV sizes are growing significantly. In 2009, according to Statista, only 7% of households had a TV greater than 49”and in 2019 this figure is 33%. 88% of global television sales are now above 40”. Demand for bigger TVs is growing with consumers wanting to create a true cinematic and immersive experience within the home - 8K TVs make this possible.

The perceived quality of a TV picture is unique to you, with multiple factors altering your viewing experience. If you have 20/20 vision, the optimal viewing distance for 8K (Super Hi-Vision) is displayed in the diagram below, and as you can see, they are extremely short.

It’s all about the upscaling because the contents available today is HD or 4K at best! The upscaling technology within the 8K TVs allow viewers to

experience 4K content upscaled to 8K, producing a beautiful picture that attempts to mimic 8K quality, it isn’t 8K, but it’s a pretty close second. The best 8K TVs will use supersampling; a method used to remove jagged and pixelated edges (also known as ‘jaggies’) from images rendered in TVs.

In the world of transmitted images, there are no smooth lines, because lines are made up of many small squares, each with a single colour. Unless the line is perfectly vertical or horizontal, the lines will be jagged and supersampling aims to reduce this effect by calculating an average colour value via several points within the pixel.

4K TVs are already using supersampling technology to display lower resolution content and the advancements in AI and pattern recognition algorithms have significantly improved to make way for 8K upscaling. 8K technology won’t just allow for better viewing on a TV screen but will have a significant impact on the quality of VR experience. The 7690 x 4320 pixels could potentially offer a detailed and more immersive VR experiences than ever before with more pixels per inch. Although, evidently the technology to back this up isn’t quite there yet with Chinese VR company Pimax originally promising Kickstarter investors an 8K VR headset and instead providing a 5K VR Headset.

HDMI 2.1 is the new standard for HDMI connectors, confirmed back in November 2017, this technology has managed to keep up with the likes of 8K and VR with continual improvements. The new cables allow for faster refresh rates, catering for 8K resolution video at 60 frames per second, but more importantly for gamers and film enthusiast today, HDMI 2.1 allows for 4K resolution at 120fps and Dynamic HDR.

An 8K reality checkOne thing that is for sure, is that 8K equipment is expensive. Studios across the world are still in their 4K

Industry column

8K TV: Cutting through the confusionThe Digital Television Group (DTG) provides practical insight about what consumers and the industry can expect from 8K television


41-42_DTG.indd 3 07/03/2019 14:07:48

purchasing cycle and aren’t ready to adopt 8K technology just yet. Currently, the average 8K camera comes in at over £20k; the hope is to reduce the cost to a third of this price. The issue with 8K content production may be the prohibitive cost of cameras, but the manufacturers are seemingly living by the notion ‘if you build it, they will come’.

Japan and Italy’s RAI have both introduced a broadcast channel dedicated to 8K content, but it’s likely that this will only be used for large sporting events, such as the 2020 Olympics. There have been some movies filmed in 8K, Guardians of the Galaxy, The Hobbit, Transformers and House of Cards have all been produced using 8K cameras – but to enjoy all these films, you would need an 8K TV, a dedicated satellite or impeccable broadband connection

Unfortunately, it is not enough to film in 8K, a whole infrastructure needs to be put in place for consumers to enjoy it. When it comes to streaming 8K, internet connectivity is not yet at the same level of development. Even streaming 4K places a burden on bandwidth, making 4K streaming tough. You need 20mbps connectivity for 4k – and the requirement for 8K will be proportionally higher.

When you consider most households require multiple simultaneous streams,

it is likely that for successful 8K television there would need to be a gigabit connection in the standard consumer’s house. Even at this prohibitive size, the limits of the bandwidth will be tested. Then, most consumers would be daunted by the cost of this much data coming into the home. An hour of 8K streaming will require about 75 GBs of data. A long weekend of binge-watching could devour most people’s monthly data cap.

Streaming companies such as Netflix, Amazon, Hulu, and others, are struggling to deliver 4K content to consumers. This may be why they are more interested in developing HDR content at either HD or UHD resolution, which is more practical for delivery to customers. Netflix was quick to dismiss the potential of 8K; their reasoning? They focus on ensuring the content delivered is impressive. The number of pixels is irrelevant if the pixels being transmitted are of lesser quality. Netflix instead are favouring HDR, which offers brighter colours and deeper blacks, resulted in a better-balanced image.

Manufacturers are leading on 8K because they need to be ahead of the curve and create hype amongst consumers, generating demand, and in turn prompting the production of 8K content. It is important however, that the industry does not push ahead before

the infrastructure is in place. Universally agreed standards will not only create consistency across the industry and reduce confusion for consumers but will also allow manufacturers to exploit economies of scale across multiple markets. Once the DTG Widescreen Forum, then the HD Forum and now the UHD Forum, the group works hard to ensure that the ultimate UHD viewing experience is available to consumers. The DTG UHD Forum are now looking at ensuring 4K looks amazing on 8K televisions.

Why you should still be excited for 8KThe development of new generations of technology is always exciting, opening opportunities that were once thought to be the works of science fiction. 8K is most certainly here to stay, but we are unlikely to see content distribution anytime soon.

This kind of detail in images allow for advances not only in entertainment, but in the medical field too. 8K resolution is being used in surgical procedures to reduce the need for laparotomy surgery, an invasive procedure which involves cutting the abdominal wall to view the affected area. Instead, surgeons can perform endoscopic surgery, using 8K endoscopes to view the surgery on screen, this is less invasive as the incisions are smaller. The detail produced enables surgeons to see what is going on better than ever before, and the recorded 8K video can be used for research and educational purposes.

As with all new technology, there are teething problems that will take time to resolve before we start seeing 8K TV in our sitting rooms. The future of TV is one to be excited about as our home experiences become more and more immersive as next generation audio and vision evolve.

To see an 8K TV demo with your own eyes, come to the DTG Summit, New Horizons on the 8th May 2019 at One Wimpole Street, London.

Industry column


41-42_DTG.indd 4 21/03/2019 11:11:13

Sponsored by:

ww

w.c

sim

agaz

ine.

com

Converged networks, connected entertainment

New convergence paradigms

43_48_Access.indd 1 19/03/2019 11:45:04

Convergence is happening everywhere, evolving from a telco-centric version to one that is much more in tune with today’s complex

dynamics. The world is much less simple than even just five years ago. Broadband, the cloud, and connected devices like smartphones are causing industries to interact and collide.

Even with so much disruption happening in markets globally - companies like Airnbn and Uber are rewriting the rule book in their fields - television and broadcast is often cited as perhaps the most disrupted sector of all. The move from analogue to digital and then online has opened new opportunities while at the same time challenging long established business models and audience norms. Linear viewing is either flat or in decline, while SVOD and other forms of streaming are growing fast, driven by the proliferation of broadband enabled devices.

According to IHS, there are now more connected video devices in China than there are ‘dumb’ TVs in the entire world. By IHS estimates, there were less than 2 billion TVs in homes globally at the end of 2018 that can’t stream from the internet. This means there is new opportunity to reach people via non-broadcast means.

“Network capacity has made it possible for content providers to do what they couldn’t do before. It has become easier and more viable to go directly over the internet, and your viewers now have new devices they are using to consume content,” says analyst Tom Morrod.

In advanced markets, multiscreen has become the new norm. A survey just released by E&Y found that a little over a third of UK households (34%) often stream video content on multiple devices at the same time. Younger users are leading the way, with 62% of 18–24 year olds choosing to multiscreen.

TiVo, Netflix and other pioneers have shown just how much of a consumer appetite there is to be liberated of a strict linear schedule and they revealed a pent up demand to access content anytime, anywhere and on any device. This appetite for video and data across a wide range of devices will continue to grow over time.

Most broadcasters, in Europe at least, have woken up to this reality. “We recognise a new generation may not come into a linear environment at all,” says Dee Forbes, managing director at Irish national broadcaster RTE.

Swiss public broadcaster SRG is pursuing a strategy of becoming a ‘transmedia content provider’, as it calls it. “We want to say we are more and more moving away from being a broadcaster to something bigger, a transmedia content provider. Our vision is within 5 years for people to see SRG as a digital media company, as opposed to just a broadcaster,” said CTO Mathias Coinchon.

In last 2 years, RTE has similarly been restructuring its organisation to a more convergent, customer-centric model. This has led to the collapse and integration of 6 distinct divisions to remove silos, and Forbes notes how every content genre now has to work with a multi-platform perspective in mind. “We are platform agnostic in terms of content,” she said, adding: “It’s hard and takes time and thought.”

From the point of view of broadcasters, they are faced with the challenge and complexity of imp-implementing apps onto platforms, as well as getting their apps noticed. CE

fragmentation, meanwhile, doesn’t always make it easy to guarantee consistent experiences.

As Forbes suggests, a seamless user experience (UX) in a hybrid network environment is essential, but no easy feat to achieve. These challenges, will only intensify as the idea of a fully connected world starts to take shape.

Extending the multi-screen paradigmThe challenge for all companies is to keep up with the continued pace of innovation. With OTT maturing and streaming entering the next phase of its evolution, it is now a good time for media companies to embrace video and other entertainment in a wider connected context.

Operators like Sky and Comcast foresee a future where voice will evolve beyond today’s basic functions like changing channel or searching for programmes to more sophisticated control of smart home functions. These include opening doors, checking security cameras, setting the temperature and much more. Combining these with media experiences will difficult, but even more promising.

This vision is shared by large set-top box manufacturers like Arris who see the combination of TV, STB and smartphone as a good portal for smart home services. “You can use the real estate, the speakers and graphics of the TV, to interact with you,” said Charles Cheevers of Arris.

The nice thing, from a payTV perspective, is that the set-top box can

Access focus

Multiscreen 2.0: Where entertainment meets IoTNext-generation entertainment will see media blend with a wider set of connected scenarios. Goran Nastic reports


43_48_Access.indd 2 19/03/2019 11:45:16

become more than a video decoder. For consumers, the result is greater convenience and comfort.

Another smart home vertical is eHealth. “Service providers pushed smart home into the mainstream and now have the ability to do the same for home care monitoring services, particularly AAL systems,” says Jonathan Collins at ABI Research.

The any-screen mantra also extends to cars. In-vehicle entertainment is arguably a huge untapped opportunity for the broadcast industry. The inside of the connected car is becoming just another potential screen and connected device and there is an opportunity to move content to and from the home and the car for a more joint up experience, especially as driverless cars start to appear.

“With regards to connected cars and other IoT type applications it’s important to realise that the standardisation work going on inside 3GPP is very important in that space because we know that with 5G, and even 4G now - particularly the eCall directive - you have an inroad into many new cars which might not have existed five years ago,” said EBU’s Peter MacAvock, also pointing to the work happening with the upcoming DVB-I specification and bodies like the Car Connectivity Consortium (CCC).

“The media companies, on the other hand, are very anxious to ensure they can get into these cars with appropriate

content, which in the past may have focused on just audio. With the possibility of semi-autonomous and fully autonomous driving there is the option of delivering video, including high quality video, to a captive audience, which might not have existed before,” MacAvock said.

The airline industry has also started to experiment with a bring-your-own-device (BYOD) model for in-flight content, which may provide some lessons to the automotive industry.

“The BYOD idea in the car is very valid, but the car companies are trying to make sure that they don’t sell commodity transport vehicles, they are trying to sell a user experience,” said MacAvock. And he adds: “The car vendors themselves are struggling to identify open systems they can deploy which will allow them to prioritise their own user experience over, let’s say, the user experience of other vendors. And that UX will extend more into how the consumer is entertained, how they are cosseted inside the vehicles.”

ACCESS share these thoughts and considers one of the most exciting developments to be using the company’s multi-platform Twine product not just for video sharing between personal devices, but content consumption everywhere, in any scenario. Its vision sees next generation infotainment services enabling each journey to become a tailored, personal multimedia experience. “All of a sudden, media has

morphed,” as Dr Neale Foster, CEO of ACCESS Europe, put it to CSI.

Beyond the car, smart cities represent a logical and natural extension for even wider content consumption. Dr Foster offers some interesting ideas on the future intersection between entertainment and smart cities that you can read about in the next feature.

“The boundaries between the home and out of doors environment have blurred over the past few years. Today, the experiences we get in the home are absolutely fantastic: through our voice, we can turn on our TV, change channels and even switch devices when we move from one room to the next. This level of comfort has not yet been replicated well enough inside the car, which we see as the next frontier for entertainment: there is no reason that, in the future, we wouldn’t be able to get a voice assistant to turn on the ignition and start playing the opening credit of the next big TV hit,” says Dr Foster.

“The level of sophistication we have access to inside connected cars today is opening the door to new forms of entertainment, as long as the content offering adjusts to the device and the passenger. To achieve this, collaboration will be key: automakers need to be able to talk about content, and the media & entertainment industry will have to better understand the challenges and opportunities of connected, electric and smart cars.

“For us at ACCESS, the biggest hurdle to overcome is the notion that bringing a tablet with Netflix is enough to suit passengers. Replicating existing services will not work, simply because viewers don’t want to watch half of their TV show before arriving at their destination: they want a piece of content that will adjust to their personal preferences,” Dr Foster surmises.

Ensuring a seamless, high-quality and personalised experience that is available anywhere, anytime presents both a big challenge and a big opportunity. ACCESS can be found at NAB Show on booth IVE-4

Access focus


43_48_Access.indd 3 19/03/2019 11:45:31

Access focus


The rise of connected entertainment has radically transformed the way almost everyone accesses and watches their favourite content.

No longer tied to a specific schedule or the living room TV, consumers have the freedom to view their favourite programmes on practically any device, at anytime and anywhere. For the entertainment industry, this quick adoption of digital services has been a boon: today, any organisation with its own content catalogue can launch a video service. As we look towards the next revolution, where connected entertainment will move beyond the constraints of specific devices or designated apps, let’s zoom in on how multiscreen could evolve further and what needs to happen next.

Creating a true multiscreen environmentThe transition to the connected entertainment era has not been without its challenges: the industry has had to adapt from a few TV resolutions to support a wide range of devices, browsers, operating systems, screen sizes and security specifications. In addition, the proliferation of services and apps offering multimedia content has created a new challenge: increased complexity for viewers. Remembering which platform has a specific type of content – down to individual seasons of popular TV series in some cases – is simply not a good user experience. With studios and content creators entering the online video mix with direct-to-consumer offerings, the industry needs to find a solution to help consumers

navigate the fragmented connected entertainment landscape. Otherwise, we may see the reverse of the trend for cord cutting back to the convenience of pre-packaged entertainment offerings and a single monthly bill.

We believe a solution is achievable, and it can be found in the right technological tools and platforms. Today’s services can be created, tested and launched faster than ever but the missing piece is to centralise all subscribed services in one place. This will allow easy, convenient access to the services they want, and this is ultimately what all content providers and service operators desire. With our ACCESS Twine™ service platform, we have already taken the next step towards more cohesive, connected entertainment: combining a centralised service platform for unified content access with a platform agnostic technology solution.

Advancement in technologies are bringing us closer than ever before to a world where truly smart devices, smart homes, connected cars and smart cities are a reality. Here at ACCESS, we envision a world where media and entertainment services seamlessly integrate across IoT devices, inside and outside the home, including connected cars. TV programmes, Hollywood films and specially created – and curated - content will enable new business models for the automotive sector, such as subscriptions, pay-as-you-go, freemium and hyper-targeted advertising triggered by the consumer’s interactions with the real world. However, this vision runs the real risk of ending up completely fragmented if the automotive, IoT and media & entertainment worlds don’t collaborate.

This convergence needs to happen, and to be at the forefront of everyone’s thinking, so that media and entertainment can remain at the heart of consumers’ daily life.

It starts in the (smart) home Despite media and entertainment services being an integral part of our daily lives, they have typically been overlooked in the vision for an IoT enabled world. Consequently, consumers often have to use numerous applications in order to access and control their different IoT and multimedia devices and services. This is causing frustration for consumers and it contradicts the idea of IoT being the epitome of ease. By providing simplified and unified access to IoT and multimedia devices, futuristic smart home experiences can be brought to the here-and-now – and open the door to much more immersive connected entertainment options.

Breaking down the silos between IoT and multimedia services adds a new layer of convenience within the home. This will drive the consumer acceptance beyond the few early adopters that venture beyond asking their smart speaker to play a song or set an alarm. Consumers with a smart doorbell can already receive alerts on their smartphone, record a short video clip if motion is sensed, and play a short, personalised voice message when unavailable. This functionality is good, but it requires a specific app, which typically only works on a smartphone – and sometimes only one smartphone OS. With greater integration, operators can provide services that consumers’ desire, leading to increased brand loyalty and stable revenues. For example, when the smart doorbell is activated, live footage from the front door could appear on any TV screen in the home, pausing the viewing. This capability provides a much higher level of ease for consumers engrossed in a

The march towards a fully connected world By Dr Neale Foster, CEO, ACCESS Europe

43_48_Access.indd 4 21/03/2019 11:12:24

film or hosting a party, enabling them to quickly review who is visiting.

This is why we created ACCESS Twine™ for Smart Home. Our solution is designed to enable consumers to access and control all of their multimedia and IoT devices, even remotely. Support for voice assistants such as Amazon Echo/Alexa and Google Home/Assistant also helps make smart homes more intuitive. Extending this to multimedia enhances the control that viewer has of their experience by supporting rules and sequences that make sense in the real world. For example, in the middle of a film or a TV programme, consumers will be able to simply tell their voice assistant that they are going to make a drink or snack. The programme will automatically pause on the TV screen, switch on appropriate lights and even continue playing on the viewer’s connected device in the kitchen once they are sensed in the room, so they don’t miss a second of the action.

Hitting the road Extending the entertainment experience to the car is a new frontier for the automotive and entertainment industries and can leverage the experience from mobile devices. It’s crucial for the entertainment industry that Electric Vehicles (EVs), hybrid, ride sharing services and, soon,

Autonomous Vehicles (AVs) become an extension of the multiscreen ecosystem. By embracing the connected car as a new class of device, the entertainment industry can extend its existing offerings and generate new revenue streams in an environment where the consumer was previously out of reach. By making content available within the car, content creators, broadcasters and distributors can increase reach and engagement by turning vehicles into an additional platform for their assets. It could also foster new formats of content that are better tailored to in-vehicle viewing, boosting the industry further.

Thanks to ACCESS Twine for Car, our content service and automotive platform enabling secure in-vehicle services and connected infotainment, we’re striving to bring the content and automotive industries together. ACCESS Twine for Car also enables passengers to play media on any Twine enabled device within the vehicle.

Next generation infotainment services will enable each journey to become a tailored, personal multimedia experience. For example, a parent driving their children to school in the morning – e.g. for a journey of less than 30 minutes - will be offered family-friendly cartoon episodes or educational games, audio and video. All controlled from the comfort and safety of the driver’s seat and main console. The

content offering for a family embarking on a long distance journey to visit relatives could be wildly different, where audio books, podcasts - or feature length films for passengers - would be more suited for the duration of the drive. Through our solution, content owners will be able to work with the automotive industry to meet consumers’ expectations for a tailored in-vehicle infotainment experience that goes beyond simply mirroring the video services consumers already have access to in the home.

Arriving in smart citiesIn the not too distant future, consumers will be driving their connected cars in smart cities where buildings, traffic lights and road signs will all be connected. This is going to be yet another opportunity for the entertainment industry to develop new interactions with their audiences. Smart cities have the potential to make advertising more targeted than ever before by analysing consumers’ interactions with the real world. For example, a shopper passing a mall could notified of lunchtime deals or set menu prices. The entertainment industry could easily use this concept to provide a list of programmes that feature the mall or the restaurant that could be interesting viewing.

Smart homes, connected cars and smart cities are promising to revolutionise our lives by connecting things we use every day. The challenge is to make them more seamless, intuitive and efficient than ever before to foster wide adoption. One way of achieving this is to ensure a close integration with media and entertainment services. This can only be achieved if the entertainment industry works closely with manufacturers of cars, smart devices and smart cities. This will help speed the convergence between these worlds and ensure that media and entertainment continues to be at the heart of consumers’ lives.

Access focus


43_48_Access.indd 5 21/03/2019 11:12:44

In-vehicle experienceNorth Hall, N5306

Let us embark you on a connected enter-tainment journey that takes your content inside connected cars at NAB Show.

Join us for our exclusive panel discussion with Byton and France24 to learn how in-car infotainment can turn each journey into a tailored entertainment experience – only on April 8, 2pm PT in N3311.

Or come to our booth #N5306 in the In-vehicle Experience at any time to see our award-winning solutions shipped in over 1.5 billion devices worldwide.

For more information, visit eu.access-company.com or contact [email protected]

They say every story starts at NAB Show.

ACCESS_CSI_204x271_2019-03-04.indd 1 04.03.19 16:28Untitled-1 1 05/03/2019 11:08:52

Events diary 2019

Date Name Location Website

12-14 March CabSat Dubai www.cabsat.com

12-14 March FTTH Council Europe Amsterdam www.ftthconference.eu

6-7 March Cable Congress Dublin cablecongress.com

11-13 March DVB World Dublin dvbworld.org

26-27 March EBU’s Broadthinking Geneva tech.ebu.ch/events/broadthinking2019

27-28 March Connected TV Summit London connectedtvsummit.com

6-11 April NAB Las Vegas nabshow.com

8-11 April MIPTV Cannes www.miptv.com

30 April - 1 May SportsPro Live London live.sportspromedia.com

6-9 May Satellite 2019 Washington DC http://2019.satshow.com

9-10 May TV Connect London Olympia https://tmt.knect365.com/tv-connect

8 May DTG Summit London dtg.org.uk/dtg/summit.html

4-6 June ANGA Com Cologne angacom.de/en

10-13 June NEM Dubrovnik neweumarket.com

18-20 June Broadcast Asia Singapore broadcast-asia.com

6-11 September IFA Berlin https://b2b.ifa-berlin.com

13-17 September IBC Amsterdam ibc.org

30 September-3 October Cable-Tec Expo New Orleans expo.scte.org

9-10 October AI World Summit Amsterdam worldsummit.ai

14-17 October MipCom Cannes mipcom.com

13-14 November Cable Congress Berlin https://tmt.knect365.com/cable-congress

26-27 November 2018 OTT TV World Summit London ottworldsummit.com

26-27 November 2018 CDN World Summit London cdnworldsummit.com

November 2019 SportsPro OTT Summit Madrid www.sportspro-ott.com

For a full list of events taking place in 2019 please go to http://www.csimagazine.com/csi/events.php


49-EventsDiary2019.indd 3 18/03/2019 09:57:30

Technology is making the world seem ever smaller, yet perhaps now more than ever, audiences want their locality and their culture reflected in

the media they consume. From HBO’s German show Dark, with its fandoms in Chile, Bangladesh and Canada, to Netflix’s Spanish series Elite making waves in the US and UK, it’s no longer just big shows from the English-speaking world that have a global market. As Netflix likes to say, a great story can come from anywhere and travel everywhere.

In the world of high quality content, it’s increasingly the case that you can’t be a global player unless you’re a local player too.

It’s not just language that drives localisation, of course. Disney’s hit film Inside Out had 45 individual shots visually edited for differing cultural references. In some markets, hockey was swapped for soccer, while in Japan, protagonist Riley’s nightmare food was changed from broccoli to green

peppers. Why? Because Japanese children love broccoli!

The same challenges are becoming more and more common for TV too. For Doctor Who’s 50th anniversary show, 106 versions were produced, while BBC Natural History’s flagship series Dynasties required around 20 versions per episode, plus clean elements. Today, the process of creating versions is often highly manual task. Even though most of the audio and video is common between them, each version results in a new, complete video file, requiring storage, management and QC. It’s clear that we need more efficient workflows for dealing with this proliferation of versions.

IMF to the rescueThe Interoperable Master Format (IMF) is a form of component based media, allowing producers and distributors to manage multiple versions of content extremely efficiently. In an IMF workflow, we store and manage only the elements of video and audio that are unique between versions, reusing that which is shared rather than duplicating content. We use

Composition Playlists (CPLs) - data structures that act as the ‘recipes’ to define how these media components can be constructed into each required version.

This creates dramatic savings in storage and file transfer. If a new version is 98% similar to an existing version, with just a few shots changed, we need only store and manage the 2% of different shots, not a whole new copy of the programme.

IMF uses globally unique identifiers for every media track file and every composition, offering a host of benefits. It means the genealogy of content is clear. We can understand which content is shared between any two versions, simplifying content tracking and rights management. As new versions are created, knowledge is systematically preserved rather than lost.

Referencing content by unique identifiers rather than file names also enables more efficient exchange of versions between organisations. Often, new versions are created long after the original programme delivery. With IMF, we do not need to create and deliver a new full-length programme. Instead, only the new or changed content is delivered, along with a new CPL. The use of unique identifiers means that the CPL can reference video and audio which has been delivered previously, dramatically reducing the amount of video and audio that must be transferred.

When we don’t duplicate media, we don’t need to duplicate QC either. Back in the days of tape delivery, a small change could be performed as a ‘drop-in edit’, simply laying the changed shot onto the tape. We would QC the video or audio that had been changed, together with a few seconds of content on either side, but rarely would we review the whole tape end-to-end. Yet in current file-based workflows, small versioning changes result in a new, complete file, requiring a full new QC, at considerable cost. In an IMF workflow, however, it’s possible to

Technology corner

Localise or dieRowan de Pomerai, head of delivery and growth at DPP, looks at how IMF changes the game for content versioning


50-51_IMF.indd 2 07/03/2019 14:12:45

perform a full QC only on the changed media, perhaps backed up by an automated QC of the full new CPL.

IMF for broadcast and onlineBut while IMF offers huge potential to deliver workflow improvements for broadcast and online content, penetration of these workflows has been slow, apart from a couple of key players such as Netflix. So two years ago, the DPP set out to make IMF a reality for both our members and the industry as a whole, through technical work and implementation support.

The IMF standards were originally developed based on workflows from the movie industry. While it was clear that the same benefits are hugely applicable to broadcast and online content, some updates and improvements were required to fully meet the needs of that community. So, in 2018 the DPP, working with the North American Broadcasters Association (NABA), published SMPTE TSP 2121-1:2018.

This IMF for Broadcast and Online specification adds features that are important to broadcasters around the world, such as support for ProRes, high dynamic range using Hybrid Log Gamma (HLG), and control track audio description (AD). In addition, the specification constrains the parameters of IMF in order to restrict use to common combinations of frame size, colourimetry, sampling, and so on. By doing so, we enable true interoperability by aligning multiple users across the complex content supply chain around common use of the technology.

In February, we launched the DPP Certified programme, offering interoperability testing and product certification. This programme allows vendors to validate and demonstrate the quality of their implementations, and helps buyers to choose products with confidence.

Meanwhile DPP member companies from ITV to A+E Networks and BBC Studios to Turner are implementing or

piloting IMF in their workflows. From a single-episode pilot of a flagship drama to major programmes transforming end-to-end workflows across a whole organisation, 2019 will be seen as the year IMF became mainstream in global broadcast. Many of these companies will showcase their work with the DPP at NAB Show in April.

IMF in the cloudDramatically, many of the themes we’ve been discussing in our industry for the last decade collide in the implementation of IMF. Great IMF workflows rely on great metadata, and take advantage of universal identification of content. The storage and file transfer benefits of IMF make the format especially useful for dealing with UHD content. And cloud implementations built on microservices promise the most benefit across integrated supply chains.

It’s important to understand that component based media isn’t just a replacement for a flat file deliverable. Of course, it’s possible to use an IMF Package (IMP) in place of an air-ready master format such as AS-11. But to truly capitalise on the potential of IMF, it would be applied across the whole supply chain of generating, managing and exchanging versions. The most highly optimised IMF workflows place the content in the cloud, and use IMF as a framework to access, manage and reversion it.

Creating a new version becomes a case of creating a new CPL, and contributing only the new video or audio back into your cloud repository. In the complex multi-party workflows used by many, the greatest benefits are gained when all those parties - production company, post house, service providers, distributors - access the content in a common cloud system. Instead of shipping video back and forth, we simply exchange compositions. The unique identifiers in the CPL can be used to reference the media elements in that content

repository, meaning that media can be accessed when required, instead of repeatedly transferring entire versions between organisations.

What’s next?The benefits we’ve described aren’t years away. DPP member companies are already implementing cloud-based IMF workflows at scale. Technology vendors are already providing the tools to create and manage IMF, and content owners are building their next generation workflows.

The job isn’t done, of course. On the technical side, we’re working on exciting new technical developments with DPP members, including a new metadata track that will allow time-variant metadata, such as programme parting information or detailed logging information, to be stored in IMF. And we’re partnering on a project to develop common APIs for accessing IMF content, enabling true interoperability for IMF in the cloud.

To support implementation, we’re creating materials to help organisations to identify the return on investment and make the case for their investment in IMF. We’re producing informational materials and events to promote the benefits of IMF and educate the whole content supply chain. We’re developing common template delivery documents for broadcasters and distributors to use when commissioning IMF content. And we’re sharing information and knowledge about pilots and implementations of TSP 2121, to inspire and inform those who are just starting to consider IMF.

We’ll be sharing much more at our Breakfast Briefing, IMF for Real, at NAB Show in April, and on the Digital Production Partnership website, www.thedpp.com

Technology corner


Rowan de Pomerai is head of delivery and growth at the Digital Production Partnership (DPP)

50-51_IMF.indd 3 07/03/2019 14:12:59

© 2019 Synamedia and/or its affiliates. All rights reserved.

Embrace the age of Infinite Entertainment. We’re Synamedia.

We’re helping service and content providers around the globe to protect and monetize video content so they can win in the age of Infinite Entertainment. We do that with the world’s most complete, secure and advanced end-to-end open video delivery solution.

We’ll be at NAB 2019, booth SU10125, and we’re looking forward to discussing how to embrace this exciting age of Infinite Entertainment.

www.synamedia.com

streaming is entering a new phase - csi magazine · and their stories, with over 90 minutes of...

Documents