Text
A plan for free wireless data
Perhaps before the end of this year, ubiquitous city-wide free, unlimited wireless data for mobile devices will be possible. This is important because right now it costs roughly $80-$100 per month per subscriber, well beyond the reach of most people. Why is it so expensive? The infrastructure is expensive (cell towers, base stations, and centralized routing systems) and mobile carriers need to charge high rates to recoup their investments.
But there will soon be an alternative to all this heavy infrastructure currently deployed and run by carriers. The new infrastructure will follow the evolutionary direction of the Internet age: low-cost and decentralized. It will work with your existing mobile phone. Will the carriers fight it? Well, right now it looks like the carriers themselves will be introducing us to these devices. They will look a lot like a wifi router, and cost about the same.
This is the AT&T microcell (sometimes referred to as a “femtocell” by those who have memorized the metric system’s prefixes). From a carrier’s perspective, it acts as a small cell tower, that people install in their homes to get improved reception. It uses the homeowner’s broadband connection to transfer the voice and data information back to the carrier’s central systems.
Bizarrely, the carriers expect you to pay the same per-minute and per-megabyte rates for using these microcells as you do for using the regular cell towers and base stations. But why is this necessary when the subscriber pays for the infrastructure equipment? In the case of mobile data, there is no need for anything beyond the connectivity provided by these microcells. And for voice calls, there already exists data-based alternatives such as Skype and Google Voice. What then are the carriers able to provide?
The answer is that they provide coverage in areas where these microcells don’t exist, eg rural settings. If you squint a bit, rural-only carriers appear to resemble the satellite phone companies from 10 years ago, GlobalStar and Iridium. That story didn’t end with great business success.
So we have the hardware for low cost base stations that can provide data to mobile devices. And we have data-based voice calling systems that are free and (quite) reliable. What else do we need to make this dream of free wireless data come true?
- Radio waves. The carriers have paid billions of dollars for rights to the government for exclusive permission to use frequencies for mobile device communication. Our mobile phones use these frequencies, and are therefore useful only on our carrier’s radio-spectrum property. Earlier this year, the folks at MagicJack previewed a microcell device that would work with GSM frequencies, but it has not appeared in the market yet- perhaps due to legal issues in frequency band exclusivity.
- Hackery. The current crop of microcells restricts access to only the subscriber’s mobile phone. Removing this restriction allows open access to the microcell and data connection. With many of these devices based on Linux, it is reasonable to expect them to be modified to be open. And it is also reasonable to expect consumer-targeted devices without these restrictions to be available, similar to wifi routers.
- More hackery. When an active mobile phone travels beyond the range of a tower, its connection must be transferred to the next tower. This handoff mechanism may not be fully implemented in the existing microcells, and would need to be able to work between microcells. Organizing decentralized nodes in a network can be challenging, but it is possible using existing mesh networking techniques.
There have been attempts to create free metro-area networks based on wifi in the past (including Meraki, fon, SFLan), and these have met with limited success. And in general, most wifi networks are protected by a secret WEP/WPA password to prevent access by unknown people. It’s difficult to ask people to give their existing access without a real expectation that they will receive access from others. But in this case, the microcells provide a new mode of data access to the provider, and should be set up as shared by default. I believe that if people were given a reasonable proposition that would have the potential to save them hundreds of dollars every year, they would welcome the idea of participating for free mobile data. My proposal would be:
- Up-front cost of $100 for the device. No monthly fees.
- Connects to your existing internet router. You continue to pay for that service.
- Works with GSM and CDMA mobile phones to provide data service. Range is limited to a small area, which encourages people to get their own base stations, and should mitigate the frequency exclusivity concerns. (maybe)
- Automatically open to use by the general public. Total bandwidth available will be limited to a fraction of your broadband connection
Would you go for that?
