5G Really Will be Different

To the extent that core network principles such as network functions virtualization, big data, information-centric networking, cloud computing and software defined networks are part of the vision for fifth generation mobile networks, it might be argued that the fifth generation is the first to be virtually inseparable from the core network.

That is a major departure. Always in the past, the access or edge networks could be upgraded or rebuilt without necessarily affected the core network, and vice versa.

But 5G is the first network--assuming present thinking accurately reflects the future standard--that eradicates the difference between separate access and core networks.

In one sense, that is clear in the access functionality. New 5G networks will “use any available access.” That includes Wi-Fi, and might also include any other mobile or fixed network.

So 5G is quite different from all prior mobile network generations, and the first of the mobile networks not defined by its air interface.

But 5G might also move value to the core networks, or more specifically, the core networks and the content and app services that use the core network. It is a new kind of network.

Global Telecom Business is Getting More Unstable

The global telecom business has been getting more competitive for decades, starting with a wave of privatizations of former government monopolies in the 1980s as well the authorization of competition in the business.

Add on the impact of the Internet, and the growing ability to provide any service over the Internet, and the challenges have grown far beyond the mere threat of competition from other telcos.

Over the past couple of decades, mobile networks have begun to take customer share as well, shifting buying towards mobile services, and away from fixed.

Now satellite services threaten to become a major factor for Internet access globally, especially in regions where the cost of traditional service is prohibitive.

Space Exploration Technologies (Space X) has asked the Federal Communications Commission for permission to create and launch a new  low earth orbit satellite constellation of thousands of satellites that would be able to provide Internet access at unprecedented speeds anywhere on the globe.

The LEO constellations now proposed would provide a key challenge to fixed or mobile facilities in the Internet service provider business, at least in terms of coverage. In principle, every inch of the earth’s surface would be covered.

The unknown issue is the business model. It isn’t clear what the retail pricing would be, or how much market share any LEO constellation might be able to obtain.

The proposed Space X constellation would change Space X from a launch company to an Internet service provider.

That illustrates both the growing instability in the broader communications business, possible in large part because there are increasing incentives for Internet ecosystem providers to enter adjacent niches.

In other words, specific application, device or infrastructure providers can find lucrative an expansion into the actual Internet service provider business.

Orbiting the earth at just an altitude of around 750 miles, the new constellation would orbit at lower than conventional communications satellites at 22,000 miles.

That has huge implications for bandwidth and latency, potentially enabling bandwidth between 50 Mbps and gigabits for any specific end user, a huge and qualitative advance over what has been possible in the past.

In a non-related development, Facebook and Google reportedly have abandoned potential plans to launch their own geosynchronous satellites for Internet access services. One might argue the upsurge in LEO plans makes geosynchronous a largely non-competitive alternative, in terms of the amount of bandwidth any single customers can get, at any specific hard-to-reach location.

Space X Takes Step Towards Becoming an Internet Service Provider

More evidence of new interest in low earth orbit satellite constellations to provide Internet access: Space X has asked the Federal Communications Commission for permission to create and launch a new constellation of LEO satellites.

The low earth orbit satellite constellation likely would be aimed at potential customers anywhere on the globe where traditional fixed or mobile facilities are expensive to create or non-existent.

The proposed Space X constellation would change Space X from a launch company to an Internet service provider.

In a non-related development, Facebook and Google reportedly have abandoned potential plans to launch their own geosynchronous satellites for Internet access services. One might argue the upsurge in LEO plans makes geosynchronous a largely non-competitive alternative, in terms of the amount of bandwidth any single customers can get, at any specific hard-to-reach location.

The move illustrates the porous boundary between Internet ecosystm participants, where an app, device or infrastructure supplier can cross the boundary and conemplate becoming an actual Internet service provider.

The long term implications for Internet service providers (former telcos and cable TV companies or specialized satellite service providers) is unclear, but certainly increasingly unstable and uncertain.

Will tomorrow's Internet service providers be the same as today's providers? It is getting to be an open question.

Facebook Tailors Ads for Bandwidth, Device Capabilities

Facebook knows, like all developers once understood, that bandwidth and devices impose constraints and enable features for content and advertising.

That remains a reality in many markets where 2G might be the expected network for many users. So Facebook advertising cannot make assumptions that all users are on high end devices and 4G networks.

To put it another way, Facebook looks different on a feature phone than it does on a smartphone.

The creative specs for feature phones and older smartphones differ from newer devices.

Bandwidth-based targeting on Facebook gives advertisers the ability to send ads based on the quality of a person’s network connection.

Brands are now able to develop and send rich media ads, such as videos, to people on faster connections and more relevant pieces of content, such as still images, for those accessing Facebook on a weaker connection.

Nikila Srinivasan, Facebook’s product manager of emerging markets monetization, says targeting an ad’s creative elements based on device and bandwidth improves ad effectiveness.

Users on smartphones and faster networks  might see a video, while users with low-end and feature phones just see an image.

In another case, users in urban areas might see different ad than users in rural areas.

The point, all stretched analogies to network neutrality notwithstanding, is that user experience benefits from code, features and services optimized for bandwidth and device constraints.

Pakistan Levies 20% Tax on Internet Access

Taxes are a major cost item for users of communications services, in Pakistan or the United States, and officials increasingly are looking at Internet access as a lucrative source of tax revenue.

In Pakistan, officials have proposed a new 19.5 percent general sales tax on Internet and data usage. The tax applies to fixed or mobile Internet access.

Import taxes also are applied to infrastructure required to build 3G or 4G mobile networks, as well as all fixed networks.

Import duties for telecom equipment range from five percent to 20 percent, said  Sohaib Sheikh, Information Communication and Technology Think Tank (ICT3) president.

Economics matters, and prices matter, so it is easy to predict that adoption rates for mobile Internet access will be negatively affected.

In the U.S. market, Internet service taxes might be coming, as well.

Eventually, it would seem inevitable that high speed access service revenues will be taxed to support universal service programs in the U.S. market, for example.

Will 5G Move Value to Core Networks?

It might not be intuitive why fifth generation networks might be more important for the core network than the “mobile” network, or why 5G could be a step in the direction of further separating applications from access.

Alert readers will immediately grasp the business implications. To the extent 5G moves in the direction of “use any available access,” it essentially further makes all access networks commodities, including the mobile network.

So where is the value? The managed services and apps. The implications for mobile operators could not be clearer: value shifts to apps and managed services, and away from access.

In one sense, that is further deepening of an existing trend, namely the separation of access from apps that is the foundation of all Internet Protocol networks. When and if the day comes that IP itself is transcended as a protocol supporting the Internet, 5G might eventually be seen as a step along the way.

Oddly enough, IP has enabled a shift to “multi-purpose” networks, and away from “single purpose” networks, as once was the case. Where we once built networks specifically to support voice, TV, radio or satellite TV, we now mostly build IP networks that can deliver any sort of application.

If 5G develops as expected, with core networks virtualized, and able to support apps, devices and users across almost any network, one might argue that value will shift further into the core network and away from access.

Oddly enough, as important as mobility will remain, access will be a less important part of the “mobile” business. It will be the features of the core network that really drive value.

That is a huge change. Once upon a time, it was mobility itself that drove the value of the mobile network and its services. That will still be true. But we might be moving towards a network and a business, where mobility is simply a feature of core network apps, whether provided on a managed or third party basis.

One More Coming Internet Tax

In the service provider business, revenue related directly to high speed access drives revenue growth, and is the single most strategic service. That arguably is true even for funding of universal service.

Eventually, it would seem inevitable that high speed access service revenues will be taxed to support universal service programs in the U.S. market, for example.

The current problem for USF administrators is that the traditional funding mechanism--taxes on purchases of interstate and international revenues--is diminishing as use of fixed network voice diminishes.

You might well ask why that is the funding mechanism, and the answer is simply that, historically, that was where the money was, as industry profit margins were highest in those categories. Regulators and lawmakers being anything but dumb where government revenue is concerned, logically decided to tax the highest-margin services driving what once was the bulk of revenue.

It is worth noting that voice revenue trends have been through two fundamental cycles, with a third on the way.

At one time, international long distance was the highest-margin product, followed by domestic long distance. That changed fundamentally between 1997 and 2007. Over that 10-year period, long distance, which represented nearly half of all revenue, was displaced by mobile voice services.

If you want to know why USF officials seemingly constantly talk about “new revenue sources,” the decline of long distance revenues explains why that is the case.

Among proposals to reform the USF are ways to expand the base of assessable revenues (add service providers, add services taxed) or change the contribution methodology or some combination of both.

The single biggest structural issue faced by the USF is the decline in the assessable revenue base as end users move to internet based services.

Currently, service provider revenues from Internet broadband service are not part of the base of assessable revenues that contribute to the USF.

That is partly why the USF tax is about 17.4 percent.

Of the current proposals for change, one approach is to tax use of telephone numbers, which would “broaden the base.” Eventually, regulators are likely to look at taxing Internet access.

The reason is a structural problem. As more end users migrate to IP based services, the base of interstate and international revenues subject to USF contribution has decreased while simultaneously the demand for USF funds has increased dramatically, CCMI argues.

In 1998, the assessable revenue base was $80B, but in 2012 the base had declined to $66.1B, a 17 percent reduction. That is a measure of the decline in use of voice lines, and partly a decrease in use of long distance, as well as huge pricing declines for use of long distance.

In the same period, the demand for USF funds had grown from $3.9 billion to $8.5 billion, an increase of 118 percent

The FCC has capped the overall size of the USF at $12.1 billion. But the move to over the top Internet-based services means that the assessable revenue base will continue to decline, and the quarterly contribution percentage will continue to grow, even with the cap in place.

Eventually, policymakers will conclude they are better off taxing high speed access, which is where the subsidies arguably are needed, as well.