On Monday, Public Knowledge, along with the New America Foundation’s Open Technology Institute and the Benton Foundation, told the Federal Communications Commission (FCC) that AT&T’s plans to test its new services on consumers must include more thorough data collection plans and more robust consumer protections.
The FCC is now in the process of setting up trials to test new network technologies that could potentially be used to replace our existing phone network in the years to come. These trials have the potential to help us understand how to measure new voice services and how to figure out whether those new services are actually as good or better than what we have now.
But if the FCC approves AT&T’s trial proposal in its current form, it will effectively give its blessing to trials that will at best only give us a couple new data points while risking that real people will lose necessary network features or service quality with no recourse. That’s why the FCC should not approve the trials proposed by AT&T until AT&T submits more information to clarify how the trials will operate and improves the trials’ data collection methods and consumer protections, respectively.
Just because a technology is newer does not mean it is better in all respects. Even if a technology offers benefits like lower build-out costs, it’s not a true step forward for everyone if it also abandons certain calling features supported by the existing network, subjects users to longer or more frequent outages, or results in lower service quality. But, if well designed and carefully conducted, these trials can give the FCC the opportunity to more fully understand where new technologies may improve service for consumers and where those technologies must still be improved before carriers can convert entire communities onto the new networks.
In our comments, Public Knowledge pointed out ways AT&T’s trials should be improved so they produce useful, reliable information:
- The proposal should collect data on many metrics beyond just the number of dropped or blocked calls, like network capacity, call quality, device interoperability, accessibility for users with disabilities, system availability, and 9-1-1 access.
- The proposal should specify a control group to compare results with, ideally one within the same wire center as the trial.
- The proposal should make the trials transparent by involving community representatives and by releasing the collected data.
Of course, even if AT&T improves the trial designs, trials that only cover two small areas in Alabama and Florida won’t tell us everything we need to know unless they are supplemented with more trials. There are a lot of questions that will remain unanswered even after these trials, and the FCC shouldn’t pretend these two trials in Alabama and Florida give us any information at all about the impact of network transitions in, for example, colder climates, mountainous terrains, or urban areas.
Finally, the trial plans need to ensure network users will continue to be protected throughout the tests. No carrier should be allowed to systematically turn customers away from services on the traditional network infrastructure until the FCC has ensured the replacement services are comparable—a conclusion we are still far from reaching. To the extent that customers move voluntarily onto new services, AT&T must explicitly and prominently notify them of the known limitations or risks in the new technology, like losing access to medical alerts, alarm systems, 800 numbers, and collect calling services.
These trials, like the overall network transition, must serve the real people using the network first and foremost. The FCC can’t assume network changes will automatically result in better service, so the FCC must use these experiments to inform its decisions to ensure the transition leaves customers better off. AT&T’s trial proposal is a first step toward a complete proposal, but AT&T need to provide substantially more information about the trials’ data collection and customer protection mechanisms before they can even be fully considered, much less approved.
Photo by Flickr user beana_cheese.