Tuesday, December 27, 2005

Om Malik posted a provocative essay "Need For Speed … How Real?".

I agree that there is "Too Much Bandwidth" for the carriers to maintain the fiction they are special but the glut really represents an opportunity to do new kinds of applications.

For now, though, I agree that there is indeed a bandwidth glut compared with our current applications and thus our current needs. We are in a situation in which the carriers are racing to provide capacity to meet competitive pressures created by the fungible connectivity. If you can't distinguish between services than all you have to compete on is quantity -- hence the need for speed as a marketing checkbox item). Because the costs are actually negligible there is no reason to hold back on speed -- unless the carriers look ahead at a stark future in which their customers are better equipped to create new services than they are.

The carriers can only survive by creating billable events and they can't do that if they don't have privileged access to high capacity connectivity. They talk of keeping some of the capacity aside and using what they are calling a higher speed Internet for their own use but even the lower tier is enough to challenge their business model and thus their existence.

We see this with FIOS. I like my 15mbps service even though it is only one percent of the capacity of the fiber. But even with that one percent I can compete with the other gigabit! Thanks to the power of the video chips in our PCs I can do good video at 1mbps -- the speed of the original 1980's version of ADSL. Verizon is finding that they have to keep increasing the speed of DSL to compete which further demonstrates that their rational for building out fiber -- the need for speed -- is overstated.

It's not that I don't want to speed available with a fiber connection -- I just question whether it's compatible with their business model. The capacity glut is even worse when we realize that each carrier is building out its own high capacity infrastructure.

The tragedy is that this focus on speed misses the more important point -- we need connectivity. If the government policies weren't so fixated on "broadband" we could quickly provide everyone in the country with 24x7 connectivity using existing infrastructure and extending it with short distance wireless connectivity using Wi-Fi and other technologies. Instead of saddling ourselves with an obsolescent 1960's vintage E911 system we could we could be providing vital services rather than just limited emergency services.

Thirty five years ago I was excited by speed of 300bps (or .0003 mbps). Twenty five years ago I tried to explain to a newspaper service provider that 1200 bps wasn't enough even though it was faster than reading speed. Today we do have a glut in the sense that we are not yet taking advantage of the potential.

The carriers' business model is defined by scarcity. The glut means there is no scarcity and the users create their own services without have to give up a portion of the value to the carriers.

Our public policy must come to terms with abundant capacity. Rather than trying to incent the carriers to provide more of the "broadband channel", we should recognize that the capacity is real. What is missing is availability. Our goal should be to provide everyone in the country with connectivity as a fundamental right just like roads and electricity.

Rather than putting billions of dollars into building redundant infrastructure we do far better to make 24x7 connectivity a priority even at modest speeds. If we could use opportunistic connectivity rather than having to maintain multiple paths just for billing purposes we could provide real universal service.

The term "universal service" is not new--it is the name of a program originally created to make sure that everyone had telephone service. It was an enlightened idea even if the implementation was not. Universal phone service allowed everyone to participate in society and it made all phones more valuable.

Today universal service means universal connectivity. Phone service itself is just one application. It's also far less expensive today. Before spending money on new infrastructure we can repurpose the existing infrastructure -- a single copper wire can carry 100mbps but even 1mbps would be sufficient to make a real difference in people's lives.

Speed is important and the glut we see is temporary though it does threaten the carriers' ability to be a privileged provider of services. More important is availability -- the real promise of the Internet comes from connectivity and the opportunities it creates. It's time to stop talking about broadband and start providing connectivity.