Video in operators has a long history. In the beginning, AT&T built the first Picturephone system in 1956; by 1964 the “Mod 1″ was tested between exhibits at Disneyland and the New York World’s Fair. The trial results indicated that people found the picture distracting for most conversations, and were not prepared to pay significantly extra – so it didn’t catch on. Once digital technology came on the scene the CCITT (Consultative Committee for International Telegraph and Telephone, later to become the ITU) created the H.120 standard for video conferencing in the early 1980s; video conferencing has been around for over 20 years! As digital signal processing technology improved, ISDN (Integrated Services Digital Network) videophones became available. Because it required an ISDN line, specialized equipment, and customer feedback showed it to be a ‘nice-to-have,’ so the barriers outweighed the benefits.
I joined the ‘video party’ in 1991 when I did the technical due diligence on the PSTN (Public Switched Telephony Network) videophone, soon to become the BT Relate 2000. Customer feedback described the picture as ‘like a moving frying-pan’ to ‘just recognizable.’ The market quite rapidly decided that it really wasn’t good enough. At the same time as having fun with PSTN videophones, I also worked on building the first Video on Demand system, demonstrating the BT adverts running over one of the first DSL (Digital Subscriber Line) systems from Stanford University (John Cioffi had not yet formed Amati). We showed some BT adverts running over a couple of kilometers of telephone line. The BT board loved it and ran a video on demand trial. There was customer interest, the challenge was price points. In essence we were providing customers with an E1 (2 Mbit/s) line and getting about $10 per month; when the rest of BT was charging business customers thousands of dollars per month for an E1 line. Hence VoD had to wait over 10 years before BT started commercial deployment.
Today we have an explosion of video devices and services, from YouTube, through mobile video telephony to HD video-on-demand. YouTube is now approximately 10% of global Internet traffic, and in the UK the BBC’s iPlayer service is now approximately 15% of all UK Internet traffic. Add in video related traffic from other P2P (peer to peer) services, and well over half the internet traffic today is video related. However, the sad fact is that after all the investment operators have made in video over the decades, all this traffic is just using the operator as a dumb pipe. And the two video services you’d expect to be similarly following internet video in terms of traffic, i.e. mobile videotelephony and MobileTV, are clearly not.
Mobile video telephony is a failure, back in 1999 when I was working with operators in creating the 3G business cases, some of the revenue models had video-telephony accounting for 20% of calls by 2008, generating roughly 50% of call revenues. Today, video calls in many operators can be counted in the low thousands per day, while voice calls are counted in tens of millions. It’s the same problem as SMS, unless most people can use it, no one uses it. Less than half the phones shipped this year have a video-telephony capability, so the situation isn’t going to change anytime soon. For MobileTV, the situation is a little more complex. There are now at least 15 separate Mobile TV technologies, this complexity stifles the market. Japan has now shipped more than 20 million ISDB-T (Integrated Services Digital Broadcasting-Terrestrial) mobile handsets, and Korea has 8 million T-DMB (Terrestrial Digital Multimedia Broadcasting) devices, many of which are not handsets. However, the devices are used for the free-to-air services, so it does not improve ARPU (Average Revenue Per User) for mobile operators.
Put simply, customers are prepared to pay for their experiential video (stuff you sit down to watch on the TV), but expect the interactive stuff (newsclips, YouTube, etc.) to be free. The line between experiential and interactive is blurring. I hear often quoted that US students do not buy cable, they watch TV on their PC. In the rest of the world student can not afford cable! And most of those ex-student once they’re earning get their HDTV, PS3/Xbox, and HD cable/FiOS. Video Subscription revenues are not going away anytime soon, in the limit the customer will decide the mix of subscription (Sports/Premium), ad-supported (VoD), and download-to-own – just like people today pay for HBO to get quality content without the annoyance of adverts.
But back to the interactive video services, which account for the bulk of internet traffic. An operator could look at this purely as providing a driver for customers to buy internet access. Unfortunately, for mobile broadband operators the economics are a little tough to take such an approach, as discussed in this previous weblog article. This is another example of why operators need to consider open access, that is the Telco API. By exposing capabilities that make it easy to stream content to mobile phone / STB (Set Top Box), or extract content from mobile phones, or ensure quality of service to the mobile phone / IPTV STB, in addition to the many other capabilities an operator can expose to make an application developer’s life easier. In doing so an operator can then share in the revenue stimulated, whether it be through subscription, usage or advertising.
I’ll be covering these topics and more at the Dialogic One Event in the keynote session “Carrier Video Services Trends and Opportunities” on 21st October in San Diego.