• What Google Should Do In Africa - Launch a Digital Television Station

AfricaFlixInspired by the recent successful launch of the Television White Spaces pilot in South Africa, I am once again tempted to engage in providing  mostly unsolicited advice about what Google’s strategy in Africa ought to be.  What I want to propose today is the launching of a Netflix-like streaming television service aimed at African markets and serving African and international content.  In this I take a more radical point of view than I did in my recent post about the Digital Dividend. This idea is borne of two recent experiences, one, some work I did recently looking at the migration process from analogue to digital broadcasting in Africa; and, two, my personal experience with Netflix.

Digital Migration

Most of the industrialised world has made a transition from analogue to digital broadcasting of terrestrial television.  The process in most African countries is much slower.  Because digital broadcasting requires a fraction of the spectrum needed for analogue broadcasting, a transition to digital television broadcasting:

  1. creates space for more television channels;
  2. allows the possibility of broadcasting in High Definition (HD);
  3. makes some spectrum available for other purposes such as broadband; and,
  4. allows broadcasters to keep pace with changes in broadcast technology.

Of the above benefits, I question whether a. and b. are relevant in Africa.  There has always been plenty of space for television channels.  The majority of African countries have four or fewer terrestrial broadcast channels.  As for High Definition broadcasting, it is hard to make either the social or economic case for HD broadcasting.  You could make the argument about Africa not being left behind or given second class technology but frankly I have trouble seeing the real market demand for HD television in the industrialised world, perhaps for sports broadcasting.  HD television feels like the upgrade that manufacturers tell you you want.  Which of course doesn’t rule out HD television via other means such as satellite.  So the real benefit of the switch-over to digital terrestrial broadcasting is the spectrum that is freed up for other purposes such as mobile broadband or TV White Spaces technology, etc.  And if that’s true, then does there really need to be a switch-over to digital broadcasting at all?

Across the continent, the switchover process is not going very well at all.  Very few countries seem to have come up with an effective strategy of how to manage the switch-over let alone finance it.  Few countries have an effective strategy of even how provide everyone with the Set Top Boxes (STBs) that are required to view digital broadcasts if you don’t happen to have a shiny new digital television set.  In South Africa, unspent funds from the Universal Service Agency are being re-purposed for the purchase of STBs for the masses.  The bottom line is that the economics of the transition to digital broadcasting simply don’t make sense.  It is an extremely expensive process for an, as yet, undetermined benefit.  Digital broadcasting is being treated as a giant “build it and they will come” project that will hopefully create both supply and demand at the same time.  And if you don’t buy that, there is always the argument that one must embrace the digital switch-over because it is inevitable.

Most African countries signed an agreement in 2006 that they would commit to the digital switch-over.  The general sense is that you simply have to get on board because analogue broadcasting will soon be obsolete.  It reminds me a bit of the furore a few years ago about the need to move to IPv6 because all the IPv4 addresses were about to run out.  Strangely the sun has yet to set on the IPv4 address space.

So what if there were another option, an option that didn’t exist or seem credible in the near future in 2006 when the switch-over policy decision was taken?  I am referring to Over The Top (OTT) broadcasting or the delivery of what was traditionally broadcast content over the Internet.

The Success of Netflix

Netflix is the most successful example of OTT service delivery.  As of the end of 2012, they claim over 33 million subscribers worldwide, most in the United States.  Not too long after moving to Canada from South Africa last year, I signed our family up for Netflix.  I was unprepared for how cheap and how efficient their service is.  $7.99 per month for all-you-can-eat television is pretty good, by any standard.  In fact, $7.99 looks roughly like mobile ARPU in Africa. People complain about the choice and how up-to-date their offering is but for our family it is the right blend of price and service.  What was even more impressive is how well it works.  We have a single 6Mb/s dry-DSL link that provides all of our household communication including VoIP phone service and Internet.  I was pleasantly surprised to discover that my kids watching Netflix did not affect the quality of my VoIP phone calls.

Last month, I attended an event at which Netflix’s Director of Cloud Infrastructure, Adrian Cockcroft, spoke.  He further impressed me with just how far and how fast Netflix has come.  One of the first things he said was that Netflix expects to become the largest Content Distribution Network (CDN) in the world some time this year.  That may or may not turn out to be true but just the fact that he thinks it likely is saying something given the likes of Akamai, Level 3 and others.  If you know anything about the distribution of video and other big media on the Internet, you know that it is dependent on CDNs which are gigantic local caches of content.  So when you request that YouTube video in Cape Town, it appears to stream internationally at high speed but actually you’re just getting a local copy from Google’s Global Cache CDN in South Africa.  This makes for much more efficient use of the main arteries of the Internet although how CDNs should be implemented and by who is a subject of much debate.  Netflix have built their infrastructure entirely on Amazon’s cloud infrastructure which makes them a completely virtual organisation.   It impressive just how flexible and efficient at content delivery Netflix are.  No doubt the competition is not far behind them but just to say that remarkable things have been achieved in the last 6 years in both video compression and content distribution.

An Opportunity for Google?

So, a bold move for an African country would be to scrap the digital switch-over entirely and invest all that money planned for a digital broadcast network into broadband fibre and wireless infrastructure to support OTT service provision.  Bold but inconceivable.  I think we all know that that is never going to happen.  The digital switch-over ship has sailed and there is too much money already committed to it.  It is officially too big to fail.  However, that wouldn’t stop a smart company from doing the right thing.

Google already operates its own CDN network in Africa.  There are Google Global Caches in South Africa and Kenya and doubtless other countries on the continent.  It is well-positioned to manage the delivery of content on the continent.  Arguably, as a company, Google understands African media markets better than most international Internet companies having invested significantly over the last 4-5 years in local offices and local talent in Africa.

An African digital television service could be a low-barrier opportunity for new content-makers and a great way to promote thriving national industries like Nollywood to the rest of the continent.  It would be an opportunity to innovate service delivery perhaps creating new super-low-bandwidth options or perhaps overnight updates or local WiFi-based mini-caches of television content.

Are Google the best company to do this?  They certainly have all it takes although it doesn’t mean that someone else might not do it faster and better.  The three takeaway points I want to leave you with are these:

  1. The switch-over to digital broadcasting in Africa is a slow-moving train wreck.
  2. OTT television services are growing faster and more efficiently than you might have imagined.
  3. The combination of 1 and 2 could mean a different television future for Africa.





Posted by Steve Song

@stevesong local telco policy activist. social entrepreneur. founder of @villagetelco #africa #telecoms #opensource #privacy #wireless #spectrum #data

  • Pingback: Why a streaming TV service should be Google’s next African play | memeburn()

  • Vukile

    The white spaces issue presents an opportunity to introduce new players in the dissemination of media. It’ll be fallacious to sell a universal notion of adding more foreign-owned media, it hasnt done much for the local market anyway. Whoever decides to come in, has to rethink the traditional rules of engagement, entirely.

  • Pingback: Media24 Markets – Why a streaming TV service should be Google’s next African play()

  • What was once analog television is now digital anything. High power brodcast mode digital bit blasters can be used to deliver inventive services to remote locations where long range inexpensive wi-fi microwave or expensive fiber or roads or electricity can not reach, all the while broadcasting a legacy audio/video stream. But, like TCI/IP broadcast mode, there is no response channel, unless the client receiver can also transmit and the central station can also listen. I think there are solutions to be found, if someone wanted seek them out.

  • afritek

    Why not an Africa compay,why does it have to be a Western company?When will Africa wake up and stop being a consumer and be a producer?The African govements need to support the the entrpreuners within Africa to create their own ventures and destiney and end “the economic slavely” bestwoned on Africa after independence.

  • stevesong

    I did think about that very issue as I wrote this and of course there is no reason why it shouldn’t be an African company. Certainly Netflix came out of nowhere and builds its infrastructure entirely on Amazon’s cloud. Why couldn’t the same happen in Africa? The key issue for me in these “What should Google do in Africa” posts is to think about how Google could shake up some very entrenched and complacent industries. Perhaps even worse than the incumbent telcos, the broadcast industry could use a little shaking up. I feel like I want to shake someone every time I look at what a colossal cock-up the digital switch-over is.

  • Existing WI-FI products such as made by Ubiquiti Networks have been nothing short of revolutionary, moving cost of a microwave link to 1% of what it once was. In much of the world most broadcast spectrum is unused. IMHO, for those places, “white space” represents a potential, new, high volume product line of low cost radio communications devices that would work on longer wavelengths. Longer wavelenths may offer some advantages or not. Remember, white space devices come with an overburden when dealing with interference protection to licensed users that may limit use even where spectrum is unused. I don’t know how they will work where existing communications are limited or do not exist.

  • OK.. I just have to make one last contribution. My idea of a return path to the central station was to use “HF”, better known as short wave. It would have advantages in un roaded not electrified locations. Thinking someone may have already developed a spread spectrum HF protocol, Google found this 1st try. Wiki says it can reach around the globe with 1 Watt. Data rate is as low as 2 characters per second, but that should be enough to order up a data file or video. http://en.wikipedia.org/wiki/Olivia_MFSK

  • stevesong

    I agree completely with your comments on WiFi and acknowledge that there are plenty of unknowns around TVWS but I don’t think it is an either/or scenario. I think it is worth lobbying for an expanded ISM band in the 5GHz range at the same time. One huge advantage TVWS has is that it does not require re-farming of spectrum, a process that history has shown to take a long time.

  • stevesong

    It is important to make a distinction between content and content-delivery mechanisms. Right now they are bundled up in a single notion of a “broadcaster” but why should they be?

  • i really really love your article…i like to read that kind of articles…..i like african culture and i love to watch Euro News

  • Dina

    Too big to fail? You mean like the Titanic? Stop me while you still can…

  • Pingback: Spectrum and the Paradox of the ITU | Many Possibilities()

  • Pingback: Spectrum and the paradox of the ITU – Research ICT Africa()