Guy Berger
Guy Berger

Cats, karakuls and communications scenarios

It was topsy-turvy at two conferences in Johannesburg last week. What you’d expect was not what happened:

  • Delegates at the event convened by the Commonwealth Telecommunications Organisation (CTO) concentrated on … broadcasting.
  • Those attending the Commonwealth Broadcasting Association included internet and mobile in their sessions.

It wasn’t quite the “Negroponte switch” where what’s on the ground (telecoms) goes to the air (broadcast), and vice-versa. Nor was it a convergence of internet and broadcasting.

Instead, it was all about what’s going to fly in the frequency skies.

The impending shift from analogue to digital TV will free up spectrum bandwidth — and the issue is what the “digital dividend” gets used for. Among the contenders are: A: Extra digital TV offerings, B: Wireless internet broadband, C: Digital radio.

The broadcast lobby predominated at the CTO event, and mainly dealt with the complex and mega-costly shift to terrestrial digital broadcasting.

I tossed a feral cat among the pigeons by fundamentally questioning the whole shebang. My reasons (pdf of presentation here):

1. Africa in general is still far away from experiencing spectrum shortage, so there’s no urgency in freeing up airwaves by ending analogue TV
2. As it is, most African broadcasters can barely populate their analogue TV channels. So we won’t see an explosion of new African channels simply as a result of technical possibility. As for the cellphone companies, they still have elbow room to leverage more services within their existing spectrum allotment.
3. With no real benefits to anyone, digital migration is an unnecessary mega-expense (mainly for taxpayers), whereas some of the cash could rather be spent on other communications capacities. (Satellite, anyone?).
4. Contrary to common belief, there is NOT a 2015 International Telecommunications Union deadline for all Africa to switch off analogue TV. In fact, almost 30 countries are explicitly permitted to stay with some non-digital transmissions to 2020.
5. Anyway, there are no seriously harmful international consequences for any country that misses the deadlines.

Aside from TV, my argument also opposed digital migration for radio (pdf presentation here). There is not even a frequency argument here, because unlike the case of TV, there’s not much alternative use to which you can put unused analogue radio frequency.

In addition, I argued, digital radio standards are still unstable globally. Africa should let the elephants elsewhere fight out their standards on foreign turf.


Most digital radio depends on using TV frequencies, and it’s therefore ultimately hostage to analogue TV being switched off and to whether policy decisions will allocate freed spectrum for this purpose.

Most likely, digital sound will be distributed largely through digital TV channels and via the internet — rather than through direct digital audio broadcasting.

Andrew Jones of Black Earth Communications echoed to CTO my general skepticism on digital migration for Africa by asking: “What problem is it supposed to solve?”

Few delegates could give convincing answers in regard to African conditions. But some reacted strongly. “The learned professor is saying that Africa should stay lagging behind, using 1960s technology,” responded one emotional delegate from Nigeria.

Someone reminded him that if we’re talking priorities, when his government (like many others on the continent) still can’t even get piped water to its people, does it make really sense to spend money on digital broadcasting migration?

Several participants representing tech companies with vested interests in the digital switchover opposed my input for “ignoring opportunities for Africa” (aka threatening their marketing of their technologies).

A more considered counterpoint came from Jason Lobel of NDS, who pointed out that new digital TV channels could come to be populated by popular content, a la Nollywood-style videos.

Fair point. But I’m not convinced that mainstream broadcasters like the South African Broadcasting Corporation are flexible enough to carry content with low-production-values, even once they have the [digital] channels. Instead, we’re likely to get re-runs, time-shifted channels, and some “fleas on the dog” — a reference to junk content that typically gets bundled with purchases of quality programmes.

So, instead of forking out now for digital migration, African governments should probably first invest in raising the quality and quantity of local content. Then we might indeed see digital migration delivering on its promise of better choices from TV broadcasting.

Digital communications are great, but the issue is where the digital investment should go. Right now, while the digital TV horse hasn’t quite bolted, it has been fed enough to stick its head out the stable.

If I’m right, though, there’s no problem with proceeding at snail’s pace until it eventually makes enough sense to have digital TV transmission in earnest and to start switching off analogue.

Likewise, instead of rushing to catch up, there’s good reason to put some investment into complementary alternatives — like boosting local content production and expanding the use of satellite and internet.

In this regard, CTO speakers from Telkom and Ericsson spoke of digital migration not from the point of view of TV benefits, but in terms of the value of releasing UHF spectrum for wireless broadband.

What they did not address, however, is how internet usage of spectrum measures up in comparison to that by broadcasters.

The issue is that the wireless internet is not as efficient a user of bandwidth as is broadcasting when it comes to delivery of audio-video on a mass simultaneous scale.

However, if the internet (especially over the air) is constrained in terms of its multi-casting ability, broadcasting (whether digital or not) is also limited in terms of being a one-way communications flow with a limited bouquet on offer.

We need both facilities — but we have finite spectrum. The tension can only be resolved by exploring how the two communications systems can optimally relate to each other.

Enter the potential of the Set Top Box (STB), as one of the possible ways to have the best of both worlds.

The current purpose of the STB is to convert digital TV signals for display on an analogue TV set. Most South Africans will need to get one of these items in the next three years if they want to see what’s being offered on digital terrestrial TV.

But the STB could also become what Jason Lobel at CTO called “a low cost computer”. Think of the TV set as the monitor, and the STB as a CPU into which you plug a keyboard, mouse, hard-drive and a modem.

That turns a device mainly intended to deliver digital TV into a way to bring large screen internet to the masses.

Configured like this, a STB could provide a return path for viewers to interact with broadcast content. Plus, if there’s nothing to look at on the traditional “box”, users could opt for watching video on the internet.

Given African conditions, such a STB would likely be linked to wireless internet. That means online traffic will overload if people decide to use it 24/7 for viewing IPTV, or YouTube (or let alone music downloads, or gaming).

Most likely, though, there’ll be some people using the STB for their digital TV programmes, which doesn’t cause congestion. In this way the Internet won’t then have to struggle to be the primary mass delivery mechanism of high-bandwidth content.

Each system can do what it does best.

But there are a few snags to this scenario. The CTO conference heard from David Hagen of M-Net that the currently specified STB in South Africa forgot one thing in its initial design: electricity for the antenna.

The result is that the box’s sole USB hub has to be used, inelegantly, to do the power supply job. This means that if you pull out the cable in order to plug in a modem, you lose your TV signal. If you want both services at the same time, you’ll have to buy a USB hub with multi-ports.

That kind of complexity could stop the STB from becoming a wonderbox that brings interactive TV, internet access, e-governance and the like to African households.

There’s also another limit as regards the future use of South African STBs: its standard.

At the CTO conference, MultiChoice’s Gerhard Petrick urged South Africans to revisit the already agreed digital TV standard.

Upgrading to DVB-T2, he said, it would give a 30% increase in bandwidth over the existing DVB-T standard. This move would, according to Petrick, increase the cost of the STB, but he believes it is an investment worth making for the future.

In practical terms, his suggestion is a real karakul among the pigeons. But if African communications are going to get fully airborne, this is the kind of thinking needed.

The take-away point: if digital migration is not going to be postponed, it needs enough investment to ensure a coherent convergence between the current silos of broadcast and internet.

  • Pingback: uberVU - social comments()

  • Robin

    Good article, but let’s face it, logic and practically in the larger sense will never win out in this situation. The reason for this is both simple and obvious. A free market economy requires a constant supply of new products, new markets and new customers. As we have just recently become so painfully aware, this is in reality a frighteningly unstable system.

    What it boils down to is this. Currently, the analogue TV/radio/bandwidth market has been around for quite a while. A vast portion of the hardware market for this technology already own functional devices that will probably not need to be replaced for a number of years.
    Additionally, on the other hand, the technology companies have been spending/investing billions of whatever currency is convenient to them in research of new technologies on the grounds that “if we provide it – they will buy”.

    The only way that this scenario differs from any other in our economic system is that this time, these companies, and their paid representatives (mostly government or quasi-government), must first create an entirely new playing field on which the new toys can be used.

    Is this wildly expensive? Is this completely illogical, not to mention scandalous, particularly in developing countries? Of course!

    Will it happen none-the-less? Will we as consumers & tax payers, who have to foot the bill from both sides go along with it? Of course!

  • fred

    I regret that the immense amount of jargon lacing this blog is incomprehensible.
    It seems though that you oppose modernising Afrika’s telecommunications into the new digital thing because Afrika is backward and can’t pipe water to its people.

    Further you seem to suggest for jargon loaded reasons that the entire [digital] idea is basically a marketer’s confidence trick to bankrupt people, by forcing them to change from one technology to another…??

    I’m sure that if we dig back into the Star newspaper records of the period before the railways came to Johannesburg we’ll find the same arguments presented by, for instance, the [late] Schneeburger [sic] Stagecoach Company, that used to take a leisurely day to get from Jozi to the old Boer Capital.

    Every time i go online i am potentially regaled by podcasts and youTube ‘stuff’ that someone is “broadcasting” or perhaps more appropriately “narrowcasting”. I do not understand why you believe Afrika should stay on the farm and not take part in this.

    Then you seem somehow to imply that the SABC represents the whole of Afrika when most rational citizens know that the SABC is quite probably the worst example possible of a broadcast model… It is bankrupt.

    On my odd forays into the continent over the past four decades it seems that radio in particular is wonderfully well served, by a vibrant and enthusiastic broadcast fraternity… The future lies in small not big… surely that is the message of the micro[processor

  • Kanthan Pillay

    Guy: This is precisely the argument that will have Multichoice laughing all the way to the bank, allowing them to retain exclusivity in the high tech space while the masses make do with 40 year old technology.

    Here are some very basic things made possible by digital terrestrial for the masses:
    -Multiple audio channels
    -Selectable mother-tongue subtitling
    -Broadcast to hand-held devices
    -regional community TV programming
    -National educational programming

    What we desperately need to do is get rid of TV licences. Right now, we can buy a black and white TV set for R125 but need a licence costing R225 per year. What does that do for getting access to TV to the masses?


  • Iain

    @Kanthan, you have hit the nail on the head.

    TV licences are purely a purchase tax on whatever the lawmakers consider to be a TV. It is entirely possible to have a fully functional setup to watch TV without any device that requires purchase of a TV set. For example, Multichoice have been broadcasting TV content to cellphones for two or three years. Should we need a TV licence to buy a cellphone capable of receiving 3G traffic ?

    Asking the 6.5 Million or so PAYE taxpayers to stump up 1% of their income to allow the rest of the population to watch ANC sorry SABC TV for free is a candid admission that TV licences are an anachronism and should be done away with.

    @Guy – one more use for the STB. Take the multi-choice model. Encode the digital signal and decode it at the STB box. No pay, no watch. A much fairer way of doing it, using well-proven technologies and business models.

  • reslfj

    Please note that DVB-T2 in the UK is transmittet with a net bit-rate of 40.2 Mbit/sec, while DVB-T has a bit-rate of 24.13 Mbit/sec using the same TX powerlevel (ERP).

    This is an increase in capacity of 66.6% at same or better robustness.

    If the DTT is using SFN – singel frequency networks – the extra SFN overhead will be significantly reduced with DVB-T2.
    The figures could be DVB-T2 37 Mbit/sec vs DVB-T 19.91 Mbit/sec = 81% more capacity (at same TX-ERP)

    The 30% figure was the lowest design requirement – but DVB-T2 turned out much, much better.

    Lars :)