Information Age: News, analysis & insight for IT & business leaders

This year, according to a recent research report, 25% of Europe's largest companies will begin migrating their separate voice and data networks on to a single, unified IP (Internet protocol) network. To most, IP-based network convergence is primarily a way to save money. But for some, it is an opportunity to radically enhance their business competitiveness with a new generation of applications that exploit the synergies of unified voice and data communications.

Certainly, there are great savings to be made from adopting a unified IP network infrastructure. Cisco, one of the key vendors driving the network convergence market, is fond of highlighting the virtues of "eating its own dog food". Since consolidating its own multi-protocol networks, and shifting staff from conventional PBX systems to IP telephony, the company claims to have saved $1.9 billion from reduced management, maintenance and operational costs.

Other companies, without the benefit of Cisco's undoubted networking expertise, are able to tell similar stories. Large PBX installations, for instance, report being able to significantly reduce staff costs by migrating, because IP-enabled systems make it much easier to reallocate telephone extensions, for example. Companies with large storage silos make similar claims for the administrative savings that IP-based SANs allow, and other organisations note that IP is a key enabler of hotdesking - a proven resource optimisation strategy. It is no wonder that cost reduction is the main motivation: when a group of customers of networking equipment giant Avaya were asked why they were adopting unified IP, more than two thirds cited a desire to save money.

Of course, saving money is hardly a poor reason for investing in IP, but neither is it necessarily the best one. But many companies are confused about what the other potential benefits may be. A separate survey of 150 UK companies by Omnetica, the IP network integrator, found that 41% did not know what convergence was or were unable to describe its benefits, and just over half saw the "lack of identifiable business benefits" as a major barrier to adoption.

To some extent, the failure of so many businesses to see a clear benefit from investing in IP is a case of the benefit being obscured by a narrow focus on too many point solutions. A primary example is the approach some organisations take to implementing voice systems over IP networks.

Voice-over-IP (VoIP) is one of the fastest growing sectors in the IT market, with spending on IP telephony increasing at 58% a year. But a lot of this is driven by the relatively short-sighted desire to reduce voice network management costs by replacing traditional TDM (time division multiplex) connections with more flexible IP ports.

This, says Avaya's VP of global sales, channels and marketing, Louis D'Ambrosio, blinds organisations to the true value of convergence, which lies less with the simple business of lumping traffic streams generated by different applications onto the same network, and more with the opportunity for converging that traffic in ways that enrich the applications.

In particular, he believes, treating voice as just another element of the IP traffic mix is to waste a golden opportunity to bring an extra dimension to many different applications. "You cannot trivialise voice. It is not just another application; it is the most fundamental business communications medium. It makes business communications more personal, it adds intimacy," he says.

Any organisation that has invested in call centre technology will understand what Ambrosio is talking about. Call centres offer customers conducting complex or high-value transactions online the humanising comfort of voice communication. For their owners, call centres also provide an opportunity to streamline sometimes highly complex processes, by making an incoming or outgoing phone call the trigger for a series of related data transactions.

This truly multimedia approach to convergence increases the speed and quality of customer contacts, improves call centre operative productivity, and reduces the cost of sales or support calls. And some of the businesses that have adopted call centre operations, such as retail banking and insurance, have been genuinely revolutionised by its impact.

Call centres are a big ticket investment though, and the systems they use to integrate voice and data traffic streams so effectively are usually bespoke developments geared to meeting specific business requirements. The challenge facing mainstream businesses that want to optimise their return on an IP network investment is to discover similar application-level convergence synergies, but without requiring the same scale of bespoke application development.

Two years ago, the idea that businesses could routinely create their own multimedia network applications would have been a hard one to swallow. Then, companies that experimented with video-conferencing or computer integrated telephony had learned the hard way that mixing the worlds of telecommunications and data communications can be an exercise akin to alchemy. Under the surface, most corporate telecoms products - such as PBXs - were built on proprietary protocols, and those industry standards that were supported, such as the H.323 teleconferencing protocol, were designed for the use of telecoms engineers, not C++ developers.

NEW ALCHEMY

Since then, however, a new protocol has emerged from the Internet Engineering Task Force (IETF) that promises to provide enterprise developers with a more accessible tool for building bridges between the worlds of data and telecommunications.

Indeed, according to Neal Tilley, solutions marketing manager for Alcatel in northern Europe, the IETF's Session Initiation Protocol (SIP), which is based on the HTTP [the hyper text transfer protocol] "is very widely understood. Anyone who has built their own website can write SIP."

And the things that can be done with SIP (see box) are extremely powerful. Like H.323, SIP is primarily designed to automate the construction and deconstruction of communications sessions. This can be as simple as redirecting a phone call from a desktop handset to SIP-enabled mobile phone, or as complex as merging a variety different IP traffic streams - blending online data, voice, and video to create a virtual conference room involving multiple participants on different continents.

However, in a number of key respects, SIP is very different from traditional telecoms 'stimulus protocols'. H.323, for instance, required participating devices to possess significant processing and memory capacity, and was largely restricted to use in purpose built telecoms gear, such as PBXs and video link boards. SIP, on the other hand, is a 'thin' protocol that can be implemented in virtually any kind of device. This gives SIP the potential to become a ubiquitous feature of personal communications devices. It is already commonplace on PCs and laptops, where it is part of the instant messaging service embedded in Windows XP, and it is set become a standard feature of 3G telephone handsets.

The other key distinguishing feature of SIP is its ability to associate communications with people rather than with devices, and to do so regardless of where they are.

The significance of this should not be underestimated. Now, for the first time, companies have a tool which will allow them to create communications services which follow individual users, and which are automatically modulated to suit the client device they happen to be using, and the quality of the service that it is attached to. For early adopters of SIP, this is often the most compelling reason for doing so.

At the University of Lund in Sweden, head of IT Magnus Svensson discovered SIP during the course of a three-year evaluation of IP telephony systems that would replace an aging Nortel Meridien PBX. The University is proud of the fact that it was the first academic institution in Sweden to provide all its employees with an email address, but when Svensson saw the potential of SIP, he realised that here was another opportunity to radically improve the personal communications services it provides to staff.

"SIP is at least as significant as an emerging communications standard as email was," says Svensson. With 7,000 mobile-loving University staff to support, Svensson had been looking for something which would allow them to be always in touch with their email and voice services, regardless of whether they were in an office or lab somewhere within the University's four campus sites, or overseas. He believes SIP fills the bill. "SIP allows you to provide one address for everything. For voice services, email, voicemail, anything. We want every user to have their own SIP address with presence and instant messaging," he says.

The payback for Lund University is likely to be measured in terms of happier users, but SIP even saved the University money. According to Tilley, other early adopters have noticed that SIP can help to reduce the number of outgoing calls that staff make via their mobile phones. By providing them with SIP-enabled mobiles, those calls can be routed through the corporate PBX, rather than expensive cellphone networks.

However, like Avaya's D'Ambrosio, Tilley is keen to emphasise that SIP's real virtue lies not in saving money but in its potential to enable whole new classes of business service.

Even though SIP is still in its infancy as a commercial protocol, it is already in use at one US hotel chain, which provides guests with SIP phones that act as terminals to a variety of online services provided by the hotel, including a WiFi network, and that are also capable of being registered as an end-point with their employer's enterprise systems. The guests enjoy the convenience of being able to access their work systems as easily as if they were in their own offices, whilst the hotel is able to offer a variety of sophisticated - and billable - services of its own.

Ultimately though, SIP is likely to spawn applications which go beyond merely providing an easy way of integrating existing communications services, and even inspire the use of terminal devices that have so far not featured in most companies' communications strategies.

In theory, for instance, SIP-enabled televisions may soon become a display option for urgent instant messages to home workers, and one Scandinavian hospital is already experimenting with SIP-enabled name tags for clinical staff, allowing them to be tracked and contacted in places where conventional wireless devices might interfere with sensitive medical equipment.

Certainly, as SIP permeates modern communications systems, it promises to make the world of converged IP networks a friendlier and more accessible place for the people best placed to exploit it: the enterprise application development community. "If SIP does nothing else," says Tilley, "it will allow those people to be more creative."