Monday, June 10, 2013

Will "Homezones" Compete with or Complement Other ISP Services?

Cable companies are some of the ISPs that might in the future extend Internet access using “homespots” in addition to “hotspots.” The distinction is that ISPs might require, as one of the terms of service, that some portion of at-home bandwidth be reserved for “public” access by other users.

As Fon and Devicescape have done, this would potentially create a “new” network stitched together by amalgamating formerly-private at-home ISP connections.

Homespot Connect, for example, is an Android app that allows smart phone users to connect to Telenet Telenet (Belgium) “homespots.”

Google has for some time been collecting information about U.S. Wi-Fi locations, as part of its Google Maps app. In principle, that could help Google (or users) later if mechanisms to create homespots become more common.

It remains difficult to say with precision whether such “homespot” efforts represent “competition” to mobile or fixed ISP offers.

For some users, who use public Wi-Fi instead of at-home ISP service, there is some amount of competition. But most users tend to use Wi-Fi as a complement to their paid-for ISP services.

In that sense, homespots or public Wi-Fi hotspots offer more coverage, or the ability to offload traffic, and hence mostly are complementary to fixed or mobile ISP service. In principle, that is little different than a mobile operator deploying small cells to provide more capacity in dense urban areas.

But it would be fair to say that in the future, when many subscribers have connections operating at hundreds of megabits per second to a gigabit per second, there should be plenty of bandwidth to enable those at-home connections in “homezone” fashion as well, without affecting the subscriber’s experience.

Still, it will be a matter of business logic, more than anything else, that determines whether a specific homezone network is complementary or competitive to any other ISP operations.

No comments:

Costs of Creating Machine Learning Models is Up Sharply

With the caveat that we must be careful about making linear extrapolations into the future, training costs of state-of-the-art AI models hav...