Access, IP Transit: Where's the Rub?

So we've been kicking around lots of issues around telecom industry transformation at the Voice Peering Forum June 23 and 24, 2008. An attendee from Telecom New Zealand pointed out something interesting.

"In the U.S. market, contestants seem to spend a lot of time fighting over rights to use or lease the access network," he said. "That's not where the rub is, which is in IP transit."

That might strike you as an incongruous statement. After all, isn't long-haul a fairly easy thing to build? Isn't there lots of fiber?

Well, yes, there's a substantial amount of fiber, even though lots of it might not be in the right places, or lots of it concentrated inside the same cable sheaths, on the same routes.

But there's another issue, not related to fiber but to IP transit costs. If a service provider owns its own facilities, there is not much of a problem on that score. No matter how much Internet bandwidth is required, the incremental cost of supplying that demand is controllable.

That is definitively not the case for a service provider that does not own its own wide area network, and has to lease capacity in the form of IP transit. In that case, it is quite expensive if a service provider's users start to download or stream significant amounts of video.

That isn't to say access is not a crucial problem. For many contestants it is a key problem. But let's not forget that IP transit costs are growing as video consumption is growing. Sooner or later, larger service providers who do not own their own WANs will start looking at buying them or building them. That's one good way to save money on spiraling IP transit costs.

Consumer Video Drives 1/2 of Bandwidth by 2012

If Cisco is correct, and bandwidth keeps growing about 50 to 60 percent a year, and consumers keep adding video behaviors, then the easy extrapolation is that half of global bandwidth will be used for consumer video apps of one sort or another by 2012.

The challenge for wide area network and access providers is that video provides very-low revenue per bit, compared to any other service.

Lots more bandwidth, provided very economically, is going to be the business challenge.

Video on Every Display Surface?

There isn't much doubt that video is going to drive the next wave of global bandwidth consumption. There are bigger questions about the business models underneath all that video, however.

Forrester Research analyst James L. McQuivey, for example, envisages consumers being confronted with “a dozen video platforms per day,” according to Seeking Alpha. But there's more to it than that.

Forrester thinks video will become so compelling that enterprises will “broadcast” video continuously from inside the enterprise. Companies have to have a strategy for communicating every message--internal or external--using video.

“Once video becomes this easy to produce, deliver, store, and share, every agent in society will not only want to participate but will have to participate in order to have a shot at reaching people with its products and services,” McQuivey contends.

Every video surface will become a marketing platform, he predicts. When nearly every surface in your environment can display video, marketers will pay a pretty penny to show up at the bottom of a food bowl or in a bathroom mirror, where their product marketing message will be far more relevant than it is on a TV today.

“The only broker of this ad space in your home is you: We envision ad networks one day paying you for the right to aggregate your ad experiences,” he argues.

That might be stretching matters a bit. What seems more certain is that global bandwidth now is driven by video.

Business FMC Still a No-No?

Major mobile service providers--in general--seem to remain cautious about fixed-mobile convergence for business customers, mostly because of continuing uncertainty about the business model impact.

In fact, many mobile executives continue to believe they stand to lose more revenue than gain if they get aggressive about allowing mobile handsets to communicate using fixed broadband facilities, say researchers at In-State.

"This is clearly the case in the United States, with the exception of T-Mobile, which has primarily been focused on the consumer market," says David Lemelin, In-Stat analyst.

In-Stat estimates that 20 percent of businesses with Wi-Fi use it to make voice calls, which is one reason mobile providers cannot be sure business users would not simply shift former mobile traffic back to the wired network if it is possible to do so.

Most executives would agree that adding FMC can be a marketplace differentiator. Clearly T-Mobile hopes it will be. But there seem to be more questions about whether FMC is anything more than a feature to attract and retain customers.

T-Mobile and Sprint Nextel have more incentive than Verizon and AT&T to try, however. T-Mobile and Sprint do not have fixed access accounts or usage to cannibalize. If either carrier can slice its churn by promoting FMC, especially to business users, that's probably enough reason to do so.

At a higher level, there obviously is some interest in determining the extent to which business customers can be convinced to abandon some or all traditional landlines in favor of wireless-only service.

FMC--as opposed to simply installing indoor signal repeaters (femtocells)--raises revenue issues. With the femtocell approach a mobile provider still can offer the benefits of better indoor signal quality, an advantage touted by FMC proponents.

And if the FMC attraction is lower calling costs, mobile providers can create "home" or "office" zones where calls are less expensive than on the macrocell network, providing the end user lower prices, without shifting traffic to the wired broadband network.

But FMC is likely to get more attention as mobile broadband usage grows and shifting some traffic--especially non-voice applications--to the landline network and off the mobile network.

Of course, that's why fourth generation wireless networks are being built: to handle all that new broadband traffic.

Still, Strategy Analytics thinks dual-mode mobiles with built-in Wi-Fi will proliferate. People will want to use the feature. But choices have to be made: femtocell or dual-mode? Cannibalize some wireless revenue or risk losing share? No wonder carriers are cautious.

Do Browsers Matter? Revisiting an Old Thesis

It has been quite some time since anybody seriously thought "control of the browser leads to control of the customer." But mobility and Web-based applications--especially on mobile devices--might be developing in a new way. In some cases, the browser could be the path to delivering applications to "locked" devices.

Consider the iPhone and Android and LiMo devices. The iPhone is now a true software platform with a rapidly growing installed base. But it isn't an "open" platform. There are things a developer cannot do, and users cannot access, for that reason.

As it turns out, though, the browser is the way around such closed or controlled environments. Web browsers from firms such as Mozilla and Opera obviously want to be available on every device, for example. And the browser, in a mobile environment, can create a different user interface and application experience.

Mozilla’s recently introduced Firefox 3.0 differs significantly in form and function from Apple’s Safari for Mac OS X and Windows. A different rendering engine and fresh take on a browser’s touchscreen-optimized interface might be preferable to to some iPhone users.

A third-party browser could incorporate one of the most-requested features not on the iPhone at the moment: an accessible third-party plug-in architecture that would allow the inclusion of Adobe Flash, RealPlayer and other apps.

We'll have to see where this goes. Still, it's an interesting development. Browsers might matter more in a mobile context than they do in a PC context.

Jaduka: Measurable Progress in Voice Mashup Business

Jaduka Voice Services, which provides Web-based voice services in mash-up mode, has surpassed four million minutes a month of end-user volume. That includes voice conferencing; Web-initiated phone calling; automated voice alerts and notifications; broadcast messaging; and VoIP-based PC-to-phone services.

That may not seem like much, but that's fair volume for voice mash-up applications with a commercial model.

Separately, Jaduka's transaction services initiative supports creation of 28 million new user accounts a day, processing of 300 million database queries a day, and processing 1,285 account transactions per second.

Additionally, Jaduka is interconnected with major global Point of Sale Activation (POSA) networks with connections to over 506,000 retail locations. If you are thinking stored value cards, you have it right.

Peering Potrential

In principle, most international voice traffic is amenable to interconnection using some type of neutral peering federation.

The reason is that it is expensive and time-consuming to negotiate separate bilateral interconnection agreements with the ever-growing number of carriers.

If all one wanted to do was pass traffic back and forth between mobile networks, a company might have to negotiate more than 300 separate agreements.

The advantage peering federations provide is a simpler, faster way to create those business and technology agreements by joining a federated interconnection provider's community, much as Internet service providers peer with each other.

In principle, much interconnection now handled by bilateral agreements could shift, not to mention wholesale traffic, which generally isn't exchanged using a bilateral agreement because the cost of doing so is prohibitive.