Mobile video goes mainstream

Mobile video is coming -- despite the sceptics, who see the idea of people wanting to watch TV on their handsets as an unlikely scenario. I'd number myself among those sceptics, as the situation is likely to be more nuanced than a simple either/or.

What's sparked this thought is the long-expected announcement this week of a deal between T-Mobile and 3 with BT, under which the two mobile operators will have their data traffic backhauled over the incumbent's shiny new 21st century network.

The agreement was actually cut between BT Wholesale and the mobile operators' joint-venture company, Mobile Broadband Network Ltd (MBNL), which was formed to allow the two operators to combine the infrastructure for the two networks. Instating MBNL allowed them to close around 5,000 base stations, according to one report.

That left MBNL with around 7,500 base stations, which are being increasingly inundated with demands for data, as ever more-capable handsets reach the hands of growing numbers of people. While many business users have been downloading the occasional web page along with their emails for years, the iPhone has been at the forefront of generating a hockey-stick curve in the data stats. That, and the jump in sales of 3G and HSDPA-enabled data dongles, have brought about a transformation of the data market, and which could become part of a financial rescue package for the mobile operators.

So that's all very well, but what kinds of data are people downloading -- and what does it mean for the networks? While there's a pile of applications too, along with web pages, email, IM and so on, the bulkiest items that people will download are likely to be videos.

The operative word here is download, rather than stream. Real-world data is thin on the ground but data streaming in a mobile environment is a tough technical challenge. Reports from real users on web forums and the like suggest that what appears to be streaming is in fact downloading: the video downloads and plays while it downloads, YouTube and NetEventsTV-style.

In practice, downloading a video this way is a much safer and more seamless way to view content than expecting it to appear instantly. Even with a broadband link, streaming a video from an online service doesn't always mean you can view it in real time -- it's often best to wait for a minute or two until enough of it is buffered. Over a less reliable cellular link, many if not most users are likely to select this option.

And a further issue that the mobile operators and content providers aren't going to like hearing is that downloading gives users the option to fast forward through the ads, reducing the providers' opportunity to monetise the offering. It's an choice most users are likely to make.

The other issue is how much video users are likely to watch on a phone. Short YouTube clips under ten minutes long -- often called 'snack video ' -- are the most likely candidates. Few are likely to want to watch a full-length feature film, or even an episode of their favourite sitcom on a phone-sized screen. Given both the form factor, the impact on battery life, and the fact that it could take days to download a feature film, this is hardly surprising.

So mobile video is not just coming -- it's here. But downloading is the most likely delivery mechanism, not streaming. And the operators' networks still need to be updated to take account of the anticipated increase in demand. So the deal between BT Wholesale and MBNL is not likely to be the last of its kind.

Japan's NTT shows the way forward

Just got back from a week in Langkawi, where NetEvents held two conferences back to back.

We were billeted in The Andaman, a large, low-rise hotel snuggled down between the rain-forest and the beach adjoining the bath-warm Andaman Sea. It’s almost invisible until you’re on top of it, and the location, the local wildlife and, above all, the incomparable desire of the people to make your stay a memorable one combine to make it a great venue. Oh yes, and the warm tropical weather...

The first of the two conferences was the company’s inaugural summit meeting for service providers in the Asia Pacific region, which allowed SPs to meet each other to share best practices. It’s an event without parallel, according to the SPs, who were glad of an opportunity to talk to each other without the paraphernalia of a big exhibition.

NetEvents is all about meeting and talking, and that’s what happened at both events, the second being all about the more familiar meeting of vendors and press.

Service providers are often difficult to get in touch with, and I learnt a lot from this valuable opportunity to listen and talk to them. Among the most interesting presentations both to me and to most of the 60 or so attendees I spoke to was the one by Hiromichi Shinohara, from Japan’s biggest telco, NTT.

Here we learnt more about the company’s launch of its award-winning next-generation network or NGN -- alleged to be the first of its kind in the world. It’s analogous to BT’s 21st century network (universally known as 21CN), which has just been officially launched, and offers fibre to the home, which of course enables high-speed broadband access. However, BT's 21CN doesn’t enable fibre to the home but is instead an IP-based backbone network that in the main connects exchanges.

To give you an idea of how advanced NTT's network is, compare most western countries’ ADSL profile, and you’ll find that it’s growing. In the UK, for example, over half the country’s homes now have ADSL-provided broadband access to the Internet. Incumbent telco BT routinely dishes out press releases trumpeting another half-million homes connected -- see here for an example.

By comparison, NTT is way ahead. For example, numbers of Japanese users of ADSL technology, which was once described as an interim technology until we get to universal fibre-connectivity, have been declining. Instead, users are switching to fibre as an access mechanism.

This means NTT reckons it can deliver richer services, for which it can charge more. For example, it plans to offer videophone services, HDTV, video-conferencing and a range of both consumer and enterprise-focused multimedia services. As a result, NTT has been able to increase its average revenue per user or ARPU, a common metric for measuring a telco’s financial performance.

It’s routine for industry observers and analysts to argue that telcos such as BT and NTT ought to add more value to their fat pipes by providing services if they’re to remain profitable in a world where broadband prices fall as demand sky-rockets. If you look at the agenda for the service provider summit, that’s the issue which recurs throughout the event. However, the telcos have not, broadly speaking, been very successful at doing so, often because their networks cannot yet support services such as video on demand. NTT appears to be an exception to this long-standing industry mantra.

While NTT may be first to launch an NGN, depending on how you define “first”, what most at the NetEvents SP summit found interesting is the way that NTT is opening up its network for others to build on by providing a service delivery platform. as part of this process, it set up a forum for third party service providers. In this way, it helps to ensure that service providers help generate revenue for NTT, as well as for themselves. This kind of double-sided business model is one that many telcos are looking at -- and an issue, incidentally, that NetEvents TV will be examining in a future feature.

I hasten to add that I haven’t tasted NTT’s NGN, so have only Shinohara-san’s words to go on. But if the words are translated into reality, then NTT’s model is one that telcos elsewhere might do well to emulate.