Wednesday, December 18, 2013

Can "Internet Access" Be More Than a Commodity?

What makes today’s “Internet access” different from voice, text messaging or video entertainment? The answer explains why service providers spend so much time thinking up ways to make Internet access more like voice, texting or video entertainment.

And might ISPs (at least some tier one ISPs) eventually have a business model more akin to Amazon?

Without overplaying the analogy, Amazon’s network resources are a platform for selling things. People pay for the products they buy, not “access” to Amazon.com.

That is the same model service providers have used for voice, text messaging and video entertainment. Only Internet access has used a distinctly different model, namely, selling the right to access the trading platform.

And though we now commonly count a unit of “Internet access” as merely one of several products sold by an ISP, the business model is quite different from the other products. In other words, the other products are “applications” people want to use that require the use of the network. But access to the network is not what people buy.

In the case of Internet access, the model is inverted. People want access to Google Maps, search, Facebook, Twitter, Amazon, eBay and other Web-based apps.

That makes Internet access, by definition, a “dumb pipe” commodity service.

Precisely how that might change in the future is the issue. Though it is unlikely people will stop wanting general purpose Internet access, which might always be the most-purchased product, there potentially are other models, all based on the traditional way service providers have created their products.

It will not be easy to create such services or attain critical mass. Even the traditional apps have become more “commodity like” in recent years, as a wider range of solutions now exist for the communication needs once supplied exclusively by voice (mobile voice, texting, email, instant messaging and social app point-to-multipoint capabilities).

Still, the principle remains: people will buy apps that use networks. That is the principle behind machine-to-machine (Internet of Things) sensing apps. People want the advantages of instant and continual access to sensor data. They will buy the app functionality, thereby creating revenue models for network service providers.

It will not be easy. Still, in principle, nearly all the ways Internet access can be reimagined will involve the ability to create, or participate in, new apps that solve new problems, using the network, but not based on access to the general purpose Internet.

In principle, those new apps will exist, side by side with general purpose Internet access, much as Internet access now rides on the same networks used to deliver voice and video entertainment.

The interesting problem for mobile service providers is that texting is possible because it is enabled by the signaling system that supports legacy voice. As the older voice networks are shut down, in favor of voice over IP mechanisms, “texting” will become an app enabled by the network, not a byproduct of the operations of the network.

Texting functionality can be provided, in principle, though it will not be a simple byproduct of the signaling system. Precisely how much money mobile service providers will want to spend to replicate text messaging functionality in an IP environment, and how to retain distinctiveness, compared to instant messaging, is the issue.

As we now can conceive of video entertainment packages that emphasize sports, might we also conceive of a sports-optimized service including video, audio, shopping and news functions all related to sports (by extension, U.S. football, soccer, basketball or baseball offer further refinements), and sold as a bundle?

Assume for the moment we don’t have to worry about traditional line of business regulation, and that the only thing that matters is creating the product and then marketing the product to a large enough group of buyers to create a sustainable revenue model.

Health applications obviously offer a similar opportunity, where it is the “health monitoring, information and transaction” application that is the retail offering, not general purpose Internet access.

Sirius XM sells satellite audio, primarily for vehicles. In principle, Sirius XM might also provide the foundation for some connected car services as well, where the same fleet of satellites, possibly augmented by terrestrial access (mobile service providers being the logical suppliers), underpins the service.

One might argue there is no need for mobile operators to use the Sirius point to multipoint capabilities, but such “broadcast or multicast networks” are highly efficient for those parts of a service that deliver content not required to be customized.

The point is that ISPs might someday learn to create apps using the network that change “access” into managed apps. That is a big deal.

No comments:

Many Winners and Losers from Generative AI

Perhaps there is no contradiction between low historical total factor annual productivity gains and high expected generative artificial inte...