Thursday, January 5, 2017

Build It and They Will Come: Sometimes It Works, Sometimes it Doesn't

It is hard to deny the widespread opinion that 5G is the next big thing for a wide number of industries from consumer electronics to telecom; healthcare to transportation. Those hopes are likely to take longer to emerge than many believe. Some are even skeptical about the need for 5G. Some argue customers do not need the new speeds, or that business models will be unworkable.

There is some element of a "build it and they will come" strategy here. Proponents say the network will create new demand and apps. Skeptics say the risks outweigh the benefits. Every next generation network carries such risks. Not every proposal succeeds, even if deployed.

We sometimes forget that ISDN, then ATM, as well as OSI were touted, and in some cases deployed, as "next generation" networks. More recently, IMS and RCS offer other examples of proposed next generation network networks that have failed to gain adoption as originally proposed.  

Such debates are not unusual. Fixed network architects and executives argued about the wisdom and business model for fiber to the home, fiber to the neighborhood and other network platforms for decades. Google Fiber’s recent difficult experience with deploying fiber to home as an overbuilder has perhaps reignited some debate about the “best” way to supply gigabit connections. It appears Google Fiber has underperformed expectations.

The key issue is not the cost of deploying the network, but the financial returns from doing so, in competitive markets. As apparently happened, revenue was too little, and adoption rates too low, to support the business model. That has been a concern for decades, as well.

On the other hand, there always is skepticism about the next generation network, whatever that network happens to be. In fact, there was skepticism about 4G as well.

There was, in the case of 3G and 4G, a real element of build it and they will come thinking. In fact, supporters “hoped” that the faster speeds would lead to creation of new applications and revenue streams, even if nobody was sure which apps and revenues sources actually would develop.

In fact, some might argue that cost savings, in the the form of lower costs per bit, was the actual driver of 4G. On the other hand, one might well argue that 3G did succeed in making the mobile web a useful consumer experience, while 4G made consumption of video a useful and widespread consumer reality.

Many would argue the possibility that 5G really will be different, in the sense that it is being purpose built to support internet of things applications, that might not--in fact often do not need--the vast increases in bandwidth 5G will enable. Instead, it is the lower latency that will be the key enabler of some new IoT applications.

Not to be ignored, either, are the other changes in networking that will come with 5G, though logically distinct. Those key trends include network virtualization, which will lead, over time, to lower cost networks and a vast increase in ability to create on-demand bandwidth and specialized network features on a virtual basis.

Also, 5G might be the first mobile network that embraces use of unlicensed spectrum and shared spectrum in an extensive way. In fact, some leading mobile service providers (cable companies and other upstart mobile carriers) will rely heavily on use of unlicensed spectrum to support their business models.  

So 5G might well be the next big thing, because it also is happening at the same time, and building on, network virtualization, while underpinning a huge new potential wave of growth based on smart devices and systems of all types.

For the mobile industry, 5G, it is hoped, will fuel the next great wave of industry growth, beyond services and apps used by people, and based on services and apps used by machines.



No comments:

Post a Comment

Is Sora an "iPhone Moment?"

Sora is OpenAI’s new cutting-edge and possibly disruptive AI model that can generate realistic videos based on textual descriptions.  Perhap...