Fixed-mobile convergence is one of the industry's top buzzwords. But is the real convergence between telecoms and the internet?

Everyone likes to talk about fixed-mobile convergence. But one thing that was quite conspicuously missing from 2005 was much clear progress towards it. After all, convergence implies that both parties become more like each other: and far more than that, we saw the mobile sphere progressively undermining the fixed sphere. Perhaps, rather than converging, mobile is simply going to out-compete fixed? Is the convergence more like sheep-wolf convergence?

The function of any telecoms system is to enable the calling party to communicate with someone or something else. Fixed-line phones are, well, fixed. Mobile phones move around with their people. They are also usually more featureful. That only leaves price as an advantage for the fixed world-so long as the famous "mobility premium" exists, there's a case for "convergence" in the sense of routing mobile calls over a fixed line when possible.

That's not a given, though. As price wars raged in developing markets all year, the differential between mobile and fixed-line pricing was eroded. Boris Nemsic, CEO of Mobilkom Austria, told MCI back in April that with good in-building coverage, fixed-mobile convergence was a question of price. Mobilkom followed this by launching a flat rate tariff to all Austrian fixed-line and mobile numbers. If it's no more expensive, why keep the fixed lines? The same phenomenon applies to some forms of technical convergence: BT this year launched its converged device, BT Fusion, which operates as a Bluetooth cordless phone in range of its fixed line and as a GSM phone on Vodafone's network beyond it.

It hasn't been a great success, partly because the tariff isn't much better than a cheap mobile tariff. Further, all inbound calls are routed via GSM and are therefore charged at Vodafone's usual mobile rates, which meant that Fusion users' friends are forced to subsidise them. A further system, codenamed Bluephone 2, was said to be in development earlier this year at BT's Martlesham Heath research centre using Wi-Fi for the short-range radio link, but nothing has been heard for a while.

What can fixed-line phones bring to the convergence party, then? Or is the entire classification essentially on the way out? One possible answer is the range of functions provided in-network by the traditional enterprise PBX (internal numbering, operator service, directories, voicemail and such). At the moment, mobile phones are usually operated outside corporate telephone systems, even if they are company property.

Providing these services for mobile users on traditional systems is difficult. But the real convergence story of 2005, which we will come to, could change that. For example, Sony Ericsson is offering a new version of the P990 handset with its Corporate Telephony solution, which provides PBX-like functions to mobile and fixed users alike and can use company Wi-Fi networks to handle calls within the organisation, rather than incurring mobile operator charges for internal calls.

To do this sort of thing, it's necessary to do some form of overlay networking, simulating a private company network over multiple networks and accepting inbound calls centrally. The technology which enables this is well-known: Voice- over-Internet Protocol, or VoIP. And it's IP that was the real converger of 2005. In essence, doing secure VoIP permits an enterprise to run a virtual private telecoms network in the same way as it runs a virtual private data network over the public internet.

At the same time, the telecoms networks themselves are becoming more internet-like, deploying IP for backhaul and core-network links and looking to solutions like IMS for the applications layer. Vodafone CTO Thomas Geitner told the firm's half-year results press conference that it would become an IP network with "agnostic" access technologies. This, in a sense, means that the carriers themselves are becoming virtual operators of traditional telecoms services over an IP network.

VoIP also permits any individual user with an internet connection to do the same, of course. 2005 was the year Skype became a mass phenomenon, what with the surge in user numbers and minutes, the multi-billion takeover by EBay, and the first mobile Skypers making their appearance. Skype is a converging factor, too-it competes both with fixed and mobile telecoms. After all, assuming you already have an internet connection, you can't get cheaper than free, and the software also offers a variety of useful features that practically no fixed and few mobile devices can at the moment.

Using it on the move, though, is more challenging. Many Skypers (and users of other VoIP software) use public Wi-Fi access points, but this is always a limited option. GPRS should in theory offer sufficient bandwidth for a Skype call, however, latency makes this difficult. One also needs a suitable device, although most top-end phones are now capable of running the software. This would suggest that, rather than converging, mobile systems will dominate the mobile space but face quite tough competition from fixed internet service, with fixed telephony gradually dying out. The appearance of the first UMA (Unlicensed Mobile Access) handset from BenQ, which combines GSM/GPRS and Wi-Fi, as well as the proliferation of Wi-Fi radios in smartphones, gives reason to believe that this is already here.

Even away from the Wi-Fi hotspot, though, VoIP exists. A significant minority of users, essentially intrepid early-adopters, are using Skype over Sprint-Nextel and Verizon's US CDMA2000 EV-DO networks, either with PC data cards or with smartphones. (Interestingly, many of them appear to work in the industry, at least in MCI's experience.) In Germany, where both Vodafone and T-Mobile offer flat-rate data tariffs, a few Skype users are known to be active. Obviously, unlimited or at least flat rate data charging is a pre-requisite for mobile VoIP, which does not tend to please network operators.

In fact, MNOs have been so displeased at this particular manifestation of convergence that most of them, in so far as they offer unlimited or flatrate data transfer, have attempted to prevent or forbid the use of VoIP. Verizon Wireless and Vodafone Germany have both included extremely restrictive terms of service in their subscriber contracts in an effort to legislate out VoIP. It is technically feasible to render most VoIP applications inoperable by blocking access to certain ports, or by filtering packets at the media gateway.

Despite the restrictive TOS, though, it is surprising that incidents of VoIP blocking have been almost unknown in the mobile space although several fixed-line ISPs, normally large telcos, have taken technical measures to bar Skype calls. The head of Vodafone's German management board, Jurgen von Kuczowski, has gone on record as saying that he doesn't believe that port-blocking is a viable business model, as has France Telecom's head of R&D, who went so far as to say: "Voice is free-get used to it."

This mixture of dread-after all, VoIP is the ultimate bit-pipe application-and half-hearted or no enforcement seems to suggest a degree of denial on the part of the big operators. Alternatively, it could be that openness is gradually seeping through-after all, if you can't sell data services it doesn't make sense to bar applications that do use lots of data.

In the end, the real convergence story of 2005 wasn't the much-ballyhooed "fixed-mobile convergence" at the network edge, but the much more significant convergence of the infrastructure from specifically telecoms systems to a homogenous TCP/IP system in the core, which further allows new kinds of converged applications, like the Ericsson enterprise phone system mentioned above. It should come as no surprise that, by the time of the first HSDPA deployments in late 2005, at least one major carrier was planning to use DSL lines for its extra data backhaul, with a future move to Gigabit Ethernet links.