The Telecommunications Service Priority program is the communications heart of U.S. plans for emergency preparedness, mandated by law and implemented on the public switched telephone network … and we don’t know for sure that VoIP can support it. Which begs the question, What else don’t we know about VoIP?
Last month, the FCC hosted a “summit” on a topic that touches everyone yet is almost unrecognized: the Telecommunications Service Priority program. The TSP is the communications heart of U.S. plans for emergency preparedness, mandated by law and implemented on the public switched telephone network . . . and we don’t know for sure that VoIP can support it. Which begs the question, “What else don’t we know about VoIP?”
In late 2003, the government issued a request for information (RFI) to explore the impact of IP convergence on TSP. It had two primary issues to consider. First, are there new forms of communications, enabled via IP and adopted widely by industry and the government, which must now be considered for priority treatment? Second, will TSP’s treatment of voice calls and leased services work in converged networks?
The answer to the first question seems to be yes. E-mail, instant messaging and online collaboration are tools at least as valuable as voice services and thus should be included in any priority program. The problem is that there are thousands of e-mail systems and dozens of IM and collaboration systems. There is one PSTN. It was possible to create a unified TSP plan for the single PSTN, but how practical would it be to even attempt that for the plethora of new-age IP communications options?
The RFI’s requirements seem to answer the second question. They talk about security, non-traceability, restorability and other things that the Internet has not been successful in providing.
In the PSTN, services are in/of the network. In IP, they’re over the network. The PSTN acts as a kind of harmonizer of technology and policy, a place where common membership mandates common standards, and public policy can be applied because there’s a defined and small group of companies to apply it to. The IP world lets anyone create services, which increases flexibility, but reduces the network’s ability to create harmony at the service level and multiplies the number of different and possibly incompatible solutions.
I’ve read the TSP documents and talked with some of the people involved. In my view, voice over the Internet could not meet TSP requirements, nor could current e-mail and IM systems. Internet VPNs couldn’t conform to the priority reconnection goals of the program. Not all “private IP” implementations of VoIP (where the VoIP network is partitioned from the Internet) would be much better than the vanilla Internet at complying with TSP policies. So here we are, with a critical program for priority communications at risk. Why did this happen?
First, regulations today are very PSTN-centric because we haven’t had any new legislation to cover data networks and services. Even the 1996 Telecom Act didn’t do data. The FCC today is trying to meet public policy goals with laws that don’t reflect the new technology of IP.
Second, the standards process isn’t cooperating. The IETF has regularly declined to address public policy issues in standards, including Session Initiation Protocol voice. Part of this is IETF politics, but part is also the fact that the IETF wants an absolute separation of services and networks, which makes it hard to build something as unified as the PSTN using IP.
Third, carriers aren’t disclosing their “non-Internet-IP” plans or the technology on which they’re based. Everything we know about VoIP comes from over-the-Internet players that don’t control infrastructure and therefore can’t meet TSP goals.
Finally, equipment vendors aren’t bringing TSP issues to the fore, so it’s not clear how or if they can support this sort of program.
TSP is a U.S. program, but there is a TSP-like requirement in nearly every major industrial country. There are many public policy goals that converged networks may not be meeting. It’s disquieting that we’ve gotten this far in the convergence debate without exposing all the implications and issues. There are probably others yet to be exposed, and we’re running out of time to deal with them.
Nolle is president of CIMI, a technology assessment firm in Voorhees, N.J. He can be reached at (856) 753-0004 or tnolle@cimicorp.com.
