Wednesday, January 9, 2013

1 Gbps Internet Access: Could Verizon Defeat Rivals by Launching an Arms Race?

In January 2011, then Verizon COO Lowell McAdam said FiOS already has been tested for 1-Gbps service, though at that point the marketing effort was on Verizon's 150-Mbps offer.

In January 2013, as Verizon CEO Lowell McAdam said the same thing at the Consumer Electronics Show

That, of course, is the point of fiber to the home: it offers practically unlimited bandwidth. The issue always is how much bandwidth to offer, and that is a market receptiveness question. Verizon, or any other would-be supplier of very-high-speed Internet access, should offer the highest speeds at which that provider can make a positive business case.


Of course, as already is clear, unless the whole of the Internet is architecturally able to support speeds up to 1 Gbps, the bigger access pipe doesn't actually provide a consumer too much. 

Consumer experience with streamed Netflix content on Google Fiber already shows that removing the local access bottleneck on one end only moves the bottleneck elsewhere. 

But even as a marketing tool, spending one's adversary into defeat can work. In other words, Verizon arguably has the easiest task of all major access providers in widely deploying 1-Gbps service, across its footprint, rather quickly. There are real costs to add the faster opto-electronics, so the decision is not one Verizon would undertake lightly.

However, should the need arise, Verizon is best placed to push the matter on a full-production scale, putting nearly all its rivals at a disadvantage, since other competitors would be unable to match 1-Gbps speeds at all, while others (cable operators) would be able to do so, but not without significant, and time-consuming access plant upgrades or severe reconfiguration of video entertainment bandwidth. 

To be sure, executives who argue there is no serious end user benefit for 1-Gbps, today, are right. The rest of the Internet is not set up to support such speeds, so the constraints merely shift to far end servers, application coding, transport links and far-end access links. 

But someday, that might not be true. And just before that point is reached, Verizon could, in principle, make the leap to 1-Gbps service at a time when nearly all of its competitors would be unable to react at all, or would be able to react only over time.

Retail tariffs would be important, as well. Consumers might find that "slower" connections of hundreds of megabits per second actually work well enough that they would be unwilling to pay a premium for full 1-Gbps service. 

But that's partly why Google Fiber is priced the way it is: to create market pressure for lower-priced but very fast access services. Though the analogy is not perfect, U.S. President Ronald Reagan essentially started an arms race he knew the Soviet Union could not afford, leading essentially to the end of the former USSR. 

At least conceptually, Verizon could do something similar. How soon such a move might be useful is not clear. 

Expectations about what is needed change over time, nearly always in a direction of "faster access."  But Verizon wouldn't necessarily gain much by moving too soon. It would want to move just before the demand reaches significant proportions. 



No comments:

"Tokens" are the New "FLOPS," "MIPS" or "Gbps"

Modern computing has some virtually-universal reference metrics. For Gemini 1.5 and other large language models, tokens are a basic measure...