The Pirate Plan for Voice

Ed. - To warm us up for the forthcoming Telco 2.0 Exec Brainstorms on new business models (next week in London and 9-10 Dec in Orlando, Florida), Telco 2.0 is blogging from eComm this week. Below are some more highlights...

If there was a Telco 2.0 coup at the world's regulators, and they were taken over by crazy extremists determined to revolutionise the industry as fast as possible, what might happen? As you might expect, eComm is a good place to find out. It would probably be a good place to plan the coup, too. Most of the usual suspects are right here.

But the first step, I think, would be to act on the LTE voice fiasco. As Dean Bubley points out, we're now up to six possible options, which is too many. Fortunately, some of them are so awful they can be instantly discarded; an overlay circuit-switched voice network, anyone? The obvious option for the revolutionaries is option six - forget about it, and run VoIP services over the data pipe. The legacy devices can keep using GSM - it's famously difficult to actually turn a network off.

This would, of course, be a huge opportunity for the independent voice & messaging people. It would also be a huge opportunity for the carriers, to replace their Voice 1.0 interface and their efforts in the field of Worse Voice & Messaging with a line of technically superior and more customer-intimate products. Things like the power-saving radio control channel could be provided as carrier APIs. Even during a revolution, the default option is still very important, and therefore there would be money in partnering with one of the independents to get them access to the base of default-option customers.

Making that happen, though, would require action on the numbering problem. Cullen Jennings of Cisco pointed out that telephone numbers as an institution may well survive the PSTN, just because of the layers of social significance they've acquired. People are delighted to get someone's phone number - not their e-mail address. E-mail is work, twitbookspace is public, telephony is emotional. Further, telephone numbers are necessary to make the interworking with older systems happen - Decommissioning Day is a long way off, and the first jurisdictions to do that would have to keep using numbers for international calls for a very long time.

Numbers are also the only identifiers you can use when the only user interface is a dial pad, and they also retain more identifying robustness than the e-mail style user names in most VoIP systems (but notably not Skype, with its cryptographic certificates), which is important for identity/security/authentication applications.

We've got a technical solution; ENUM uses the DNS to map telephone numbers to user names, TELURL does the reverse. But so far, the only available infrastructures are private ones operated by carriers, which keeps the old numbering model in place even as the business model crumbles. Fortunately, there is an example of a solution; some countries, like Finland, coped with mobile number portability by creating an organisation to manage the national numbering plan and map the numbers to their currently-relevant line identifier.

In a public numbering system, we could own our numbers and have anyone we wanted take care of them, with some public agency to sign the root signing certificate (because you don't want to do national ENUM without also doing DNSSEC, the security extensions for the DNS). With that, and open access to the IP pipe, we could get our telephony from anywhere we liked, cross-network.

Further, as Rudolf van der Berg pointed out, changing the interconnection model for telephony to be more like the peering/transit Internet model would end the situation where the mobile operators get to subsidise themselves from the rest of the industry in general and the disruptors in particular.

The final element in this big bang strategy would be to tackle spectrum. Presenter after presenter pointed to the increasingly uncomfortable truth that vast amounts of spectrum are licenced, but unused, or unlicensed, and locally congested but still broadly underutilised. Permitting the opportunistic reuse of spectrum, beginning to open the higher frequencies, and coming up with what OFCOM's William Wood called "a better flavour of unlicensed rules" might make it possible for disruptors to get access to spectrum, where they could use the new generation of cheap and open-source equipment to deliver mobility.

This would leave a connectivity- or utility-only sector and an empowered and technically advanced applications-and-services sector, and perhaps a few successful players from the old telco market who partnered successfully with the new voice applications.

It's possible, here, to imagine that we're going through the same scenarios as we identified in the Online Video report - and voice is just entering the Pirate World phase.