It sounds such an appealing term, ‘convergence’, conjuring up an image of things neatly slotting together. There is some indication of it in the worlds of IT and telecommunications, as proprietary forms of communication and interconnection are being replaced by one unified approach—IP, the internetworking protocol at the foundation of the Internet.

This started from the open systems revolution in the 90s that dumped IT vendors' silo'd networking solutions in favour of a universal Internet. Now IP is being used to deliver telephony and video over the same network and is moving through the cores of traditionally telecoms and television networks—BT's 21st century network, IP to the mobile network cell towers, and IP to the home.

The reality of this convergence is far more likely to be a ‘collision’ with similar repercussions to the IT convergence on open systems. This led to the demise of many once famous industry giants, and at one time it was even suggested that IBM would have to break up to survive. It didn't fracture, but changed dramatically, and as the other giants stumbled, other rapidly growing companies emerged as powerhouses.

Rather than vertically integrated, providing CPU, computer, operating system, network and applications software like databases, they focussed on the individual horizontal layers. Intel, Dell, Microsoft, Cisco and Oracle all became the 800-pound gorillas of their own horizontal markets.

Convergence in the telecommunications industry landscape is likely to be just as dramatic, perhaps more so, as this far older industry has such a long history of tightly, vertically integrated vendors, from the national monopoly fixed telephone suppliers to the current swath of mobile operators. Arguably some of this tight integration is justified, as the service expectations of telephony users are that the phone system ‘just works’, and so control of the infrastructure, often with more efficient, yet proprietary protocols, has been necessary.

Open systems changes all of this. Yes, it removes much of the certainty and predictability of network performance, as, rather than dedicated circuits, individual packets navigate through many interconnected networks, but it also opens up opportunities for new services and with it new, much needed revenue streams. So, with a little effort, the uncertainties can be minimised, traffic flows can be managed and service levels can be set.

New services and new bundles of existing services—for example the ‘quad play’ of voice, video, Internet and mobile—become possible as, rather than simply VoIP (Voice over IP), we rapidly move to XoIP (where ‘X' equals anything or everything). However, beneath the network flow, many different technologies are necessary to deliver the raw speed or bandwidth expected by the many applications. High speed copper connections, even faster fibre optics, and over wireless networks a diverse array from satellite broadband, third and fourth generation cellular data and local connection from Wi-Fi, and perhaps one day, WiMAX. This brings far more complexity to the telco business.

The convergence is now a collision of business models and multiple levels of focus; innovation and services to deliver new revenues, diverse networks to deliver quality of service and seamless connectivity, and infrastructure to do the whole thing as economically and efficiently as possible. Rather than trying to manage all this in one integrated ‘telco’, how about a different split:

• Companies who can combine understanding of end customer needs with the ability to develop and deliver compelling services—many have emerged as leaders from the Internet industry, such as Google and Yahoo, others are the strong consumer brands we all know and love, such as some of the virtual mobile operators, Virgin and Disney. These we might call ‘servcos’.
  
• Managing and provisioning multi-function infrastructure services with the speed and flexibility of an IT rollout while maintaining the quality and availability levels of traditional telecoms companies requires specialist skills and procedures. These could be built on top of existing operator capabilities, but must span multiple types of connectivity—fixed or mobile networks of differing technologies. These are the ‘netcos’.
  
• While many operators are afraid of becoming ‘bit pipes’ there is a valid business model for the provision of raw data transport. Often this could be far more efficiently delivered if several companies shared some of the physical assets—imagine if mobile operators shared cell towers and back haul networks—so why not have specialist providers focussed on doing just that. We might call them ‘bitcos’.

The vertically integrated model so loved by many current operators has many flaws, but the main ones are exposed by the needs of most of us—businesses and consumers—to have multiple forms of connectivity available on whatever mix of technology endpoints we choose. The growth of a pervasive all-encompassing network—the Internet—has not only re-arranged the IT landscape, it now seems certain it will do the same for telecoms.

When UK incumbent BT announced its plans to move to IP completely through its core network—known as 21st century network—it described this as a move from a tangled spaghetti network to layered lasagne. If the entire telecoms industry is heading towards a lasagne model too, some companies are going to be asking, in the words of Spencer Johnson's book, ‘who moved my cheese?’.