Friday, December 03, 2010

Why the Internet Won't Solve Everything

There seems to be a common misunderstanding among a lot of journalists, experts, trend analysts and consultants that the Internet will solve all distribution problems in the world. In their opinion there is no need for broadcasting as a way of getting content to the thousands and millions and billions of people out there.

This is a very tabloidized, and a very wrong picture.

Unfortunately this seems to be creating a perceived truth among politicians, regulators, bureaucrats and other decision makers in a variety of countries. The perceived truth that the Internet can handle distribution of all content in the world. But why is this misconception a problem?

Well, because the Internet cannot cope when a lot of people want to watch or listen to something at exactly the same time. That is why broadcasting was invented. A broadcaster (owning TV channels or radio stations) send the signals out in the air, and EVERYONE in the coverage area can receive the signal, even if EVERYONE means 300 million or a billion people. And because telecom operators are trying to lobby governments around the world into taking frequencies from broadcasters, handing them to telecom operators so that there will be more bandwidth for wireless Internet. If they succeed, there will be a shortage of available frequencies (space to broadcast) for broadcasters. The victims will be viewers and listeners not able to receive their favourite TV or radio programmes.

- But broadcasting is old! The internet is new, the internet is the future, new media experts comment. The wheel is also old, but it still works perfectly for what it was designed to do and it will continue to do so in the future. So will broadcasting.
There are various broadcasting standards for radio and mobile TV out there. In Norway we are using DAB and DAB+ for radio and DMB for mobile TV. For TV most countries have already chosen DVB-T, ATSC or ISDB-T. For radio and mobile TV there is only the DMB family of standards (including DAB and DAB+) that is open, accessible through widely available devices and that has allocated frequencies in most countries. I do therefore examplify broadcasting through the usage of DMB, DAB and DAB+.


A list of 20
I have listed often neglected reasons for why broadcasting is still needed (updated from an original list of 8). A shorter version of this blog post can be found here.

1) Telecom networks go down if too may people use them. During emergencies and disasters, a broadcasting network is the only effective (and sometimes the only possible) way of getting information out to those affected. This is a reason why governments are considering legislation that will require a broadcasting chip in all mobile phones.

2) Broadcasters are in editorial control of their content, and should remain so. Having a gatekeeper acting as an additional editor or censorship manager is limiting democracy and free speech. A gatekeeper can be a government that cencors you through threats or force, an ISP or a telecom operator that limits content or users' access to it or someone controlling what applications/contents/services that will be made available on their platform. James Cridland covers this issue (for radio) on his blog.

There are however also other kinds of gatekeepers that can limit democracy and freedom of speech. Let us take Mark Thompson as an example. He ie editor of chief of the BBC, employing 30,000 people, many of whom are journalists. He is, and should be, the editor-in-chief. For some platforms, i.e. Apple or Facebook, he is not. These two examples are essentially walled gardens which are controlled by Tim Cook and Mark Zuckerberg. They suddenly become super editor in chiefs, above Mark Thompson, creating a less than free press as they will have to OK apps or feeds appearing on their devices or websites.

In April 2012, support for this point came from a somewhat unlikely source. Sergey Brin, one of the cofounders of Google warned against a dedemocratization of the internet in The Guardian. The reason? Gatekeepers.

3) The Internet is vulnerable to hacking and sites may relatively easily be taken down or halted. A broadcasting network is more secure. Bigger operations will usually be needed in order to hijack a broadcasted signal or a broadcaster. Viruses are also plentiful on the internet, now also on Apple products. Viruses are easy to create, and stopping them through anti-virus software is a multi billion industry.

4) DMB consumes a lot less power than GPRS/3G/WLAN. It is in other words a greener and much more environmentally friendly technology. And it makes batteries last longer. A DMB/DAB/DAB+ chip uses 25-60mW. A WLAN/GPRS/3G chip uses 1-7W. That means that DMB chipsets use between (updated due to calculation error) 0.36% and 6% of an IP chip. 0,36% is one twohundredandeightieth (1/280). Do also note that this only covere the receiving end of things. The transmitters needed for connected devices (inernet capable gadgets) are much higher than the number of transmitters needed for broadcasting in the same area.

5) Broadcasters must pay around 3 Euro cents per GB that is being transferred from their server park to the Internet. This may not sound like much, but if Norwegian Broadcasting Corporation (NRK) in little Norway with less than 5 million inhabitants were to distribute its radio and TV offering throughout a year via the Internet, we would be loooking at an amount of 4.9 Exabyte (which is 4,901 Petabyte or 4,900,662 Terrabyte). This is based on average viewing and listening times from NRK. The average Norwegian person over the age of 12 listens to NRKs radio stations over 35 hours a month and watches NRKs TV channels more than 40 hours a month. This would set the public service broadcaster back over 150 million Euro a year. This is highly hypotetical since there is no way that the Internet would be able to deliver this

6) Internet is not free and you have to pay for access. A subscription is needed (unless you find an open WLAN zone). If every average Norwegian was to watch TV via his/her mobile broadband subscription (the 11% of TV viewing that happens outside your home in Norway), the telecom operators would be generating 275 million Euro every month from Norwegian subscriptions (again given that the network could actually cope and that every Norwegian subscribed to the cheapest possible plan). Broadcasting is the most cost effective way to reach a vast number of people in a geographically large area.

7) The Internet is not wide enough when it comes to bandwidth. I will examplify this through some numbers which applies to mobile broadband. But of course LTE will solve everything. We are being told.

There are some problems though. To build an LTE network (often referred to as 3.9G or 4G) for 90% of the population in the UK would require 16,000 transmitters (and then a lot of roads and railroads would still not be covered). To cover the same with DMB, DAB and DAB+ would take less than 300 transmitters. Every LTE transmitter costs around 250,000 USD with running costs of around 60,000 USD every year. That requires an investment of 4 billion USD and operation costs of over 1,8 billion USD a year. If a telecom operator was to build such a network, 4 million customers, each paying 550 USD a year would be needed to break even.

LTE furthermore has a capacity of 31,68Mbps (given a 10Mhz channel) per transmitter. If all the users are located very near and with a direct view of the transmitter. If even one user is further away or behind a hill or a house, a different modulation scheme is needed, something that severly lowers available bandwidth. LTE may need up to 11 modulation schemes to reach users at different distances and locations from the transmitter.

31,68Mbps would give 78 users mobile TV at the same time (at a bitrate of 384Kbps) if all of them were very close to the transmitter and no one was using the same transmitter for any other kind of Internet surfing (mail, news, Facebook, downloads, etc.).

The LTE estimates were presented by Arqiva, a British company building and running various kinds of networks in October, 2010.

8) The Internet works poorly when the receiver travels at high speeds, or some times if moving at all. And a lot of people are listening to radio and watching TV while on the move. In cars, trains, trams, boats or even by walking. Streaming services (when large amounts of data are transferred continously as is the case with TV and radio) do not work well, even if there is coverage. And big areas are difficult and expensive to cover with Internet technologies. In Norway, 40% of radio listening is experienced outside peoples own homes, in most cases while travelling. DMB, DAB and DAB+ works well in speeds up to 900 km/h (tested on planes), and such broadcasts can cover vast areas.

9) The Internet through broadband is neither broad nor stable if you are far away from a switch. That includes a lot of households in rural areas. They have always had much poorer quality on their Internet connection than people in cities, sometimes the ISPs even refuse to give them a connection due to costs that they will never recoup.

10) Broadcasting is not only for live radio or TV, but is a very universal distribution technology. Any file can be broadcast, either for live consumption or to the memory of a device. This can include an eMagazine to a tablet (delivered simultaneously to everyone each morning), traffic information to your navigation/GPS device or films and music to your phone. I have covered this in more detail in the blog post Broadband Isn't Only for Radio and TV.

11) The Internet may not be open in the future as the net neutrality is being challenged in various countries, including in the United States, Norway and Canada. The more people want to use the Internet, the more bandwidth is becoming a luxury (see point 7). As with almost all forms of goods and services, a luxury market will then emerge. But the Internet has limited capacity, so if certain websites are going to be able to pay for guaranteed bandwidth, the bandwidth will have to be taken from the available bandwidth of others. Even though all content providers (websites) currently should in principle have the same possibilities to access end users with their content and services. As long as certain websites will be able to pay a premium to reach  the end users, that will hamper free competition and the principle of net neutrality. Some ISPs now also want to charge in both ends. Meaning from both the consumer and the companies providing information there (i.e. Youtube, CNN, NRK). Update: Telenor Takes on Google.

12) In order to cover a big country as Norway (area wise) with DMB between 500 and 600 transmitters are needed as the reach is good and as the signal is error protected and robust. A DMB signal that bounces off a mountain or a big building is also strengthened, while an analogue signal that bounces is destroyed. To cover the same 99.8% of the country (as the 500-600 DMB transmitters will do) with Internet, many more transmitters are needed.

13) The Internet is vulnerable. Imagine that all infrastructure has to go in or on top of the same ditch. Water, electricity, sewage, cable TV, telephone lines, Internet, roads, railroads, etc. If you have to repair only one of these networks, you will have to disrupt all of them. If all distribution of content and services is to be forced into the same 'cable' the scenario is similar. It is easy to shut down, as seen in Egypt and Libya (updated in February, 2011). Only one distribution channel for everything means no backup. Always have a backup plan.

14) Several countries have introduced The Data Retention Directive, EUs Directive 2006/24/EC. This states that all electronic communication will be stored for up to one year (half a year in Norway) in order for national governments (specifically police, security agencies, etc.) to have access to everything in order to prevent crime. Isn't that nice and caring of them? This has however not prevented crime anywhere, and it totally changes the widespread and much stated doctrine from "innocent until proven guilty" to "guilty until proven innocent." In a digital world, everything you do will leave a trace and you cannot hide. The consumation of broadcasted content is still an exception. You can watch or listen to anything without anyone knowing about it. This is now no longer the case on the Internet. I'd recommend you to read more about the directive here, in the words of Peter J Milford, Southampton Business School. Norwegian readers can also find a lot of well formulated thoughts on the issue by NRKs lawyer, Jon Wessel-Aas.

15) It is neither economical, effective nor attractive to continue doing double distribution. The Internet cannot cope with enormous amounts of concurrent viewers or listeners. Broadcasting will  therefore be a needed technology "forever." To build a net that is capable of delivering live television or radio to hundred of millions of users at the same time (as during Super Bowl or the Eurovision Song Contest) will never prove a viable business model.

16) We are still seeing an explosion in data traffic passing through the internet. More data was transferred over the internet in 2010 than all previous years, according to Intel. And the growth is not going to slow down. First of all, we are seeing an explosion in cloud services. People and companies store their data in the cloud rather than on their hard drives. That demands a lot of capacity. And the number of internet capable devices is skyrocketing. In 2011 there were 5 billion connected devices in the world. That number is projected to reach 15 billion by 2015 and 50 billion by 2020. A much higher number than the number of people, in other words (there are 7 billion people in 2012, there will be 7.3 billion in 2015 and 7.6 billion in 2020). How? There are a lot of M2M (machine to machine) services being introduced, i.e. automatic registration of electricity usage or cars that are equipped with an internet connection. To want to transfer all radio and TV traffic to the internet in addition to everything else that is going through that infrastructure does not make any sense, nor will anyone want to fund that. And of course, one of the buzz words of 2012 is cloud computing. Many people and companies put all their data in the cloud (accessible via the internet). This increases the bandwidth needed manifold.

17) Many ISPs and mobile network operators argue that the only way forward is to take frequencies from broadcasting and give them to themselves in order to provide people with more bandwidth. This would hardly help and would be very shortsighted. The usage of bandwidth is estimated to increase 70 times the next 15 years. So even if MNOs can access more spectrum (they are currently trying to get their hands on spectrum currently used by broadcasters), it would hardly help at all. Giving them that would only ensure that valuable means of distributing mobile data will end up in the hands of very few companies, creating monopoly like conditions. Who would want that?

The report The Economic value of Broadcast Innovation - Impact on the U.S. Treasure, was publsihed by Business Analytix Inc. in November 2011 shows this clearly, and concludes.
A linear increase in the supply of spectrum cannot solve a geometric increase in demand for mobile data. The projected growth in mobile data traffic is so great that re-allocation of television spectrum would provide only temporary and barely discernible relief.
Broadcasting spectrum must remain in the hands of broadcasters that operate independently of MNOs and ISPs.

18) Broadcasting does not discriminate. The internet does. How can I claim such a thing? Personalization is about to make a divide between internet users. What you read, watch or listen to is more and more often influenced by your previous behavious online. What are your interests, what do you like reading, where in the world are you and who are your friends? Past behaviour on the internet while you are logged in to sites such as Facebook or Google or cookies monitoring your behaviour on behalf on sites you haven't even logged into help websites personalize what you get to see and what you won't get to see. Even a web search for the exact same phrase may return very different answers depending on who is searching and from where. This fragments web surfing and takes away the equality of users. I'd say it is like discrimination. When watching broadcasted TV, listening to broadcasted radio or even reading news via teletext (Ceefax) you will get exactly the same as everyone else within coverage, and at the same time too. Broadcasting threats everyone equal. Websites should offer users a "button" where they can switch off such monitoring, usually referred to as personalization which sounds like a helpful service. Websites do this as they make more money off advertising, but it may not be the best solution in the long run. More on this here.

19) Radio and television will "always" be best live. Not all content in all situations, of course, but the majority. Why? Neither sports nor news (especially breaking news) is well suited for watching or listening to "on demand". If you don't consume it live, you'll be left out. Every program will also necessarily need to have a premiere and a lot of people want to be the first to enjoy it. Just as is the case with cinema premieres around the world. The rise in discussions during programmes via chats and soacial media is also due to the fact that people watch or listen simultaneously. Otherwise you cannot discuss something equally. Being served, as opposed to having to go get it youself is also an element. Live broadcasts are served to you while you relax in the sofa. Luxury! On demand programs you will have to find and start on your own. Linear stations will also surprise you. New programs, new genres or new music will suddenly appear. And they will do so regardless of your what you have watched or listened to before. Because broadcasting doesn't discriminate. Everyone gets the same at the same time. And it has not even been personalized based on what you have done before, where you are or what your friends like or do not like. Not to mention emergencies. There are, in short, many reasons for the never fading success of live radio and TV.


20) If broadcasters cannot distribute their content via efficient and cost effective broadcasting networks they will have to fight over access via the internet and compete with everyone else that is depending on the internet for their services. These "others" may have built their business models purely on internet distribution. Unless there is an option to the internet, broadcasters will have to fight over bandwidth with these companies. That weakes the bargaining power of broadcasters in order to distribute radio and television programmes and channels, and also puts pressure on non-broadcasters.

Combination is the new king
Does this blog post mean that I am not in favour of the Internet? Of course not. The Internet is amazing and opens up for unlimited possibilities. But it is not constructed to be used as a distribution platform for the same content to many people at the same time. What I do believe in is a combination of technologies. Anyone with a little DIY knowledge knows that you need the right tool for the job. Broadcasting is great for some things, the Internet great for others. Combination is the new king. Combination of broadcasting and on demand content (the Internet), combination of content and services (when a return channel usually is needed, i.e. the Internet), combination of TV and radio and combination when it comes to partnerships between industries and even competitors. Not to forget a combination of technologies, also for live content that is being broadcasted. Some people will in certain situations or contexts prefer i.e. webradio to a stand alone kitchen radio.

All the great TV and radio channels and programmes out there need be distributed unfiltered via at least one distribution platform, without any gatekeepers (i.e. ISPs, telecom operators, Apple) that in effect act as editors by possibly filtering out the content or as bouncers by turning the users away.

The goal of this blog post? To contribute to a more fair and balanced discussion when it comes to distribution platforms. The Internet will not and cannot solve everything.

A shorter version of this blog post can be found here.

Follow @garfors on Twitter for notifications on new posts.


NMTV launched 6 TV channels via DMB as MiniTV in 2009. 4 transmitters cover 25% of the Norwegian population.

12 comments:

  1. Thumb up. Why use valuable web bandwidth for TV and radio when it is anyhow better transported via independent networks?

    ReplyDelete
  2. Why the Internet Won't Solve Everything - Computers are expensive and not all members of society have access to the internet therefore excluding some current listeners this goes against offering free to air radio to all and fulfiling the remit of PSBs.

    ReplyDelete
  3. The internet, DAB, FM and satellite all need to work together to give the consumer a seamless and relevant service. In the UK digital switchover is driven by greedy government rather than what is best for the consumer. www.creativeradio.com

    ReplyDelete
  4. Combination technologies, like RadioDNS, can also help bring new features to broadcast radio by using the internet as it's meant to be used: by adding relevant personalised information and a back channel. There's more info at www.radiodns.org

    What you say above, of course, is equally valid for FM as well as DAB. The only future for radio is a multi-platform future.

    ReplyDelete
  5. I just don't understand some of your calculations.

    "A DMB/DAB/DAB+ chip uses 25-60mW. A WLAN/GPRS/3G chip uses 1-7W. That means that DMB chipsets use between 0.025% and 6% of an IP chip. 0,025% is one twohundredandeightyeth (1/280)."

    "4.9 Exabyte (which is 4,901 Petabyte or 4,900,662 Terrabyte)."

    ReplyDelete
  6. 4.9 Exabyte of content is based on the following:

    The average Norwegian over 12 years (3.9 million people) consumes over 117 hours of radio and TV every month.

    Radio is being broadcast at between 128 and 192Kbps. Television is being broadcast at between 4000 and 6000Kbps, the latter bandwidth for HD.

    If all 3.9 million Norwegians were to listen and watch all their radio and TV programmes via the Internet, we would be looking at 4.9 Exabyte in a year. I have not taken into account that some people will watch/listen together.

    The power consumption figures are based on numbers from chipset manufacturers.

    ReplyDelete
  7. "0,025% is one twohundredandeightyeth (1/280)."

    No, this is 1/4000.

    ReplyDelete
  8. My bad, thanks for letting me know. The right figures (as updated in the blog post) should be:

    DMB chipsets use between 0.36% and 6% of an IP chip. 0,36% is one twohundredandeightieth (1/280).

    ReplyDelete
  9. Hi Gunnar,

    Totally agree that the future is hybrid. Just waiting for the intelligent apps that we can install at radio stations to push broadcast content over broadcast networks to mobiles, and interactive content over 3G. My guess is that several Norwegian companies already have the technology - just waiting for the right programme format - and someone to take the initial risk. Isn't that the role for public broadcasting? Happy New Year and success with your ideas and business in 2011. Hope opportunities arise to work together.

    ReplyDelete
  10. Jonathan,

    Thanks for your comments and wishes! Norwegian Mobile TV Corporation (NMTV) and partners are working on those intelligent apps. The challenge now has been the lack of receivers with a return channel and open APIs towards the DMB chip. Such devices will hit the market in the second quarter of 2011, finally opening up for creative combinations of broadcasting and the Internet. Norwegian Broadcasting Corporation (NRK) which is the public broadcaster here is taking part in this through their 1/3 ownership of NMTV. ANd so are the two biggest private broadcasters who own the rest. That shows a strong belief in a hybrid future, also from a commercial viewpoint.

    ReplyDelete
  11. Then give the broadcaster what is the broadcaster's , and to telecom-carriers what is carriers'...

    Nice to meet you the other day. I full agree with you that the Internet is not constructed to be used as a distribution platform for the same content to many people at the same time. A decentralizing mechanism as it is, the Internet's forte lies in narrow-casting and networking. It will not and shall not supersede public broadcasting, just as a marketplace/gift economy, no matter how robust it is, can never be the replacement of a central bank/ financial system.



    That said, I see things slightly different when talking about the Internet:

    Reliability : Internet can be fragile but is a distributed system thus a totalizing control over it can be too costly, while broadcasting maybe durable, but as a centralized system it can be taken over in minutes in extreme situations (a military coup for instance) .

    Freedom of speech: it's true that having a faceless guy as the gatekeeper is more painful than having someone from the elite in the position. Nevertheless, the faceless guy is supposed to operate under certain stipulated protocols, while editors usually have more latitude to make their own decisions.

    Safety: Hacking can be an issue for everything going online. but what is on stake with Internet broadcasting (and all personal media)is rather insignificant when compared to public broadcasting. Internet broadcasting is protected by its own obscurity.


    my defense for the Internet is rather peripheral. Internet and broadcasting take up different roles in cultural production and agenda setting. Given there is already too much hype about the Internet and the mobile Internet, a solid defender of public broadcasting should be in place to ensure the future of public communication. Good Luck and best wishes!


    Shubo

    ReplyDelete
  12. best part of this post is further away or behind a hill or a house, a different modulation scheme is needed, something that severly lowers available bandwidth. LTE may need up to 11 modulation
    thanks
    Regards
    Emily.Bronte

    ReplyDelete